Interventions for preventing falls in older people living in the community

Interventions for preventing falls in older people living in the

community (Review)

Gillespie LD Robertson MC Gillespie WJ Lamb SE Gates S Cumming RG Rowe BH

This is a reprint of a Cochrane review prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library2009 Issue 4

httpwwwthecochranelibrarycom

Interventions for preventing falls in older people living in the community (Review)

Copyright copy 2009 The Cochrane Collaboration Published by John Wiley amp Sons Ltd

Rettig
Text Box
Retrieved December 7 2009 from13httpmrwintersciencewileycomcochraneclsysrevarticlesCD007146pdf_standard_fshtml

T A B L E O F C O N T E N T S

1HEADER 1ABSTRACT 2PLAIN LANGUAGE SUMMARY 3BACKGROUND 3OBJECTIVES 3METHODS 6RESULTS

Figure 1 13Figure 2 14Figure 3 20

21DISCUSSION 24AUTHORSrsquo CONCLUSIONS 25ACKNOWLEDGEMENTS 26REFERENCES 47CHARACTERISTICS OF STUDIES

185DATA AND ANALYSES 194FEEDBACK 195WHATrsquoS NEW 195HISTORY 195CONTRIBUTIONS OF AUTHORS 195DECLARATIONS OF INTEREST 196SOURCES OF SUPPORT 196DIFFERENCES BETWEEN PROTOCOL AND REVIEW 196NOTES 196INDEX TERMS

iInterventions for preventing falls in older people living in the community (Review)


[Intervention Review]

Interventions for preventing falls in older people living in thecommunity

Lesley D Gillespie1 M Clare Robertson1 William J Gillespie2 Sarah E Lamb3 Simon Gates3 Robert G Cumming4 Brian H Rowe5

1Department of Medical and Surgical Sciences Dunedin School of Medicine University of Otago Dunedin New Zealand 2HullYork Medical School University of Hull Hull UK 3Warwick Clinical Trials Unit Warwick Medical School University of WarwickCoventry UK 4Centre for Education and Research on Ageing University of Sydney Concord Australia 5Department of EmergencyMedicine University of Alberta Edmonton Canada

Contact address Lesley D Gillespie Department of Medical and Surgical Sciences Dunedin School of Medicine University ofOtago PO Box 913 Dunedin Otago 9054 New Zealand lesleygillespieotagoacnz lesleygillespieyahooconz (Editorial groupCochrane Bone Joint and Muscle Trauma Group)

Cochrane Database of Systematic Reviews Issue 4 2009 (Status in this issue Edited commented)Copyright copy 2009 The Cochrane Collaboration Published by John Wiley amp Sons LtdDOI 10100214651858CD007146pub2This version first published online 15 April 2009 in Issue 2 2009 Re-published online with edits 7 October 2009 in Issue 4 2009Last assessed as up-to-date 7 October 2008 (Help document - Dates and Statuses explained)

This record should be cited as Gillespie LD Robertson MC Gillespie WJ Lamb SE Gates S Cumming RG Rowe BH Interventionsfor preventing falls in older people living in the community Cochrane Database of Systematic Reviews 2009 Issue 2 Art No CD007146DOI 10100214651858CD007146pub2

A B S T R A C T

Background

Approximately 30 of people over 65 years of age living in the community fall each year

Objectives

To assess the effects of interventions to reduce the incidence of falls in older people living in the community

Search strategy

We searched the Cochrane Bone Joint and Muscle Trauma Group Specialised Register CENTRAL (The Cochrane Library 2008 Issue2) MEDLINE EMBASE CINAHL and Current Controlled Trials (all to May 2008)

Selection criteria

Randomised trials of interventions to reduce falls in community-dwelling older people Primary outcomes were rate of falls and risk offalling

Data collection and analysis

Two review authors independently assessed trial quality and extracted data Data were pooled where appropriate

Main results

We included 111 trials (55303 participants)

Multiple-component group exercise reduced rate of falls and risk of falling (rate ratio (RaR) 078 95CI 071 to 086 risk ratio (RR)083 95CI 072 to 097) as did Tai Chi (RaR 063 95CI 052 to 078 RR 065 95CI 051 to 082) and individually prescribedmultiple-component home-based exercise (RaR 066 95CI 053 to 082 RR 077 95CI 061 to 097)

1Interventions for preventing falls in older people living in the community (Review)


Assessment and multifactorial intervention reduced rate of falls (RaR 075 95CI 065 to 086) but not risk of falling

Overall vitamin D did not reduce falls (RaR 095 95CI 080 to 114 RR 096 95CI 092 to 101) but may do so in people withlower vitamin D levels

Overall home safety interventions did not reduce falls (RaR 090 95CI 079 to 103 RR 089 95CI 080 to 100) but wereeffective in people with severe visual impairment and in others at higher risk of falling An anti-slip shoe device reduced rate of falls inicy conditions (RaR 042 95CI 022 to 078)

Gradual withdrawal of psychotropic medication reduced rate of falls (RaR 034 95CI 016 to 073) but not risk of falling Aprescribing modification programme for primary care physicians significantly reduced risk of falling (RR 061 95CI 041 to 091)

Pacemakers reduced rate of falls in people with carotid sinus hypersensitivity (RaR 042 95CI 023 to 075) First eye cataract surgeryreduced rate of falls (RaR 066 95CI 045 to 095)

There is some evidence that falls prevention strategies can be cost saving

Authorsrsquo conclusions

Exercise interventions reduce risk and rate of falls Research is needed to confirm the contexts in which multifactorial assessment andintervention home safety interventions vitamin D supplementation and other interventions are effective

P L A I N L A N G U A G E S U M M A R Y

Interventions for preventing falls in older people living in the community

As people get older they may fall more often for a variety of reasons including problems with balance poor vision and dementia Up to30 may fall per year Although one in five falls may require medical attention less than one in 10 results in a fracture Fear of fallingcan result in self-restricted activity levels It may not be possible to prevent falls completely but people who tend to fall frequently maybe enabled to fall less often

This review looked at which methods are effective for older people living in the community and includes 111 randomised controlledtrials with a total of 55303 participants

Exercise programmes may target strength balance flexibility or endurance Programmes that contain two or more of these componentsreduce rate of falls and number of people falling Exercising in supervised groups participating in Tai Chi and carrying out individuallyprescribed exercise programmes at home are all effective

Multifactorial interventions assess an individual personrsquos risk of falling and then carry out or arrange referral for treatment to reducetheir risk They have been shown in some studies to be effective but have been ineffective in others Overall current evidence showsthat they do reduce rate of falls in older people living in the community These are complex interventions and their effectiveness maybe dependent on factors yet to be determined

Taking vitamin D supplements probably does not reduce falls except in people who have a low level of vitamin D in the blood Thesesupplements may be associated with high levels of calcium in the blood gastrointestinal discomfort and kidney disorders

Interventions to improve home safety do not seem to be effective except in people at high risk for example with severe visual impairmentAn anti-slip shoe device worn in icy conditions can reduce falls

Some medications increase the risk of falling Ensuring that medications are reviewed and adjusted may be effective in reducing fallsGradual withdrawal from some types of drugs for improving sleep reducing anxiety and treating depression has been shown to reducefalls

Cataract surgery reduces falls in people having the operation on the first affected eye Insertion of a pacemaker can reduce falls inpeople with frequent falls associated with carotid sinus hypersensitivity a condition which may result in changes in heart rate and bloodpressure



B A C K G R O U N D

Description of the condition

About a third of community-dwelling people over 65 years oldfall each year (Campbell 1990 Tinetti 1988) and the rate of fall-related injuries increases with age (Sattin 1992) Falls can haveserious consequences but if injury does occur it is usually minorbruising abrasions lacerations strains and sprains Less than 10of falls result in fracture (Campbell 1990 Tinetti 1988) howeverfall-associated fractures in older people are a significant source ofmorbidity (Sattin 1992) and mortality (Keene 1993)Despite early attempts to achieve a consensus definition of ldquoa fallrdquo(Buchner 1993 Kellogg 1987) many definitions still exist in theliterature Investigators have adapted these consensus definitionsfor use with specific target populations or interventions (Hauer2006 Zecevic 2006) It is particularly important to have a clearsimple definition for studies in which older people document theirown falls their concept of a fall may differ from that of researchersor health care professionals (Zecevic 2006) A recent consensusstatement defines a fall as ldquoan unexpected event in which the par-ticipant comes to rest on the ground floor or lower levelrdquo (Lamb2005) The wording recommended when asking participants isldquoIn the past month have you had any fall including a slip or tripin which you lost your balance and landed on the floor or groundor lower levelrdquo (Lamb 2005)Risk factors for falling have been identified by epidemiologicalstudies of varying quality These are summarised in the guidelineproduced by the American Geriatrics Society British GeriatricsSociety and American Academy of Orthopaedic Surgeons Panelon Falls Prevention (AGSBGS 2001) About 15 of falls resultfrom an external event that would cause most people to fall asimilar proportion have a single identifiable cause such as syncopeor Parkinsonrsquos disease and the remainder result from multipleinteracting factors (Campbell 2006)Since many risk factors appear to interact in those who suffer fall-related fractures (Cummings 1995) it is not clear to what extentinterventions designed to prevent falls will also prevent hip orother fall-associated fractures Falls can also have psychologicalconsequences fear of falling and loss of confidence that can resultin self-restricted activity levels resulting in reduction in physicalfunction and social interactions (Vellas 1997) Falling puts a strainon the family and is an independent predictor of admission to anursing home (Tinetti 1997)

Description of the intervention

Many preventive intervention programmes based on reported riskfactors have been established and evaluated (AGSBGS 2001)

These have included exercise programmes to improve strengthor balance education programmes medication optimisation andenvironmental modification In some studies single interventionshave been evaluated in others interventions with more than onecomponent have been used Delivery of multiple-component in-terventions may be based on individual assessment (a multifac-torial intervention) or the same components are provided to allparticipants (a multiple intervention)

Why it is important to do this review

The best evidence for the efficacy of interventions to prevent fallingshould emerge from large well-conducted randomised controlledtrials or from meta-analysis of smaller trials A systematic reviewis required to identify the large number of trials in this area andsummarise the evidence for health care professionals researcherspolicy makers and others with an interest in this topic We havesplit the previous Cochrane review ldquoInterventions for preventingfalls in elderly peoplerdquo (Gillespie 2003) into two reviews to sepa-rate interventions for preventing falls in older people living in thecommunity from those in nursing care facilities and hospitals (Cameron 2005) This is partly due to the increase in the numberof trials in both settings but also because participant character-istics and the environment may warrant different types of inter-ventions in the different settings possibly implemented by peoplewith different skill mixes Gillespie 2003 has now been withdrawnfrom The Cochrane Library

O B J E C T I V E S

To summarise the best evidence for effectiveness of interventionsdesigned to reduce the incidence of falls in older people living inthe community

M E T H O D S

Criteria for considering studies for this review

Types of studies

We included randomised controlled trials and quasi-randomisedtrials (eg allocation by alternation or date of birth)

Types of participants



We included trials of interventions to prevent falls if they specifiedan inclusion criterion of 60 years or over or clearly recruited par-ticipants described as elderly seniors or older people Trials thatincluded younger participants for example recruited on the ba-sis of a medical condition such as a stroke or Parkinsonrsquos diseasehave been included if the mean age minus one standard deviationwas more than 60 years We included trials where the majority ofparticipants were living in the community either at home or inplaces of residence that on the whole do not provide residentialhealth-related care or rehabilitative services for example hostelsretirement villages or sheltered housing Trials with mixed pop-ulations (community and higher dependency places of residence)were either included in this review or the Cochrane review on fallprevention in nursing care facilities or hospitals (Cameron 2005)however they were eligible for inclusion in both reviews if datawere provided for subgroups based on setting Inclusion in eitherreview was determined by discussion between the authors of bothreviews and based on the proportion of participants from eachsetting

Types of interventions

This review focusses on any intervention designed to reduce fallsin older people (ie designed to minimise exposure to or the effectof any risk factor for falling) We included trials where the inter-vention was compared with rsquousual carersquo (ie no change in usualactivities) or a rsquoplaceborsquo control intervention (ie an interventionthat is not thought to reduce falls for example general health ed-ucation or social visits) Studies comparing two types of fall-pre-vention interventions were also included

Types of outcome measures

We included only trials that reported outcomes relating to rate ornumber of falls or number of participants sustaining at least onefall during follow up (fallers) Prospective daily calendars returnedmonthly for at least one year is the preferred method for recordingfalls (Lamb 2005) However falls outcome measurement in theincluded studies vary and we have included trials where falls wererecorded retrospectively or not monitored continuously through-out the trial The following are the outcomes for the review

Primary outcomes

bull Rate of fallsbull Number of fallers

Secondary outcomes

bull Number of participants sustaining fall-related fracturesbull Adverse effects of the interventionsbull Economic outcomes

Search methods for identification of studies

Electronic searches

We searched the Cochrane Bone Joint and Muscle Trauma GroupSpecialised Register (May 2008) the Cochrane Central Regis-ter of Controlled Trials ( The Cochrane Library 2008 Issue 2)MEDLINE (1950 to May 2008) EMBASE (1988 to May 2008)CINAHL (Cumulative Index to Nursing and Allied Health Lit-erature) (1982 to May 2008) PsycINFO (1967 to Sept 2007)and AMED (Allied and Complementary Medicine) (1985 toSept 2007) Ongoing trials were identified by searching the UKNational Research Register (NRR) Archive (to September 2007)Current Controlled Trials (accessed 31 March 2008) and theAustralian New Zealand Clinical Trials Registry (accessed 31March 2008) We did not apply any language restrictionsIn MEDLINE (OvidSP) subject-specific search terms were com-bined with the sensitivity-maximising version of the MEDLINEtrial search strategy (Lefebvre 2008) but without the drug therapyfloating subheading which produced too many spurious referencesfor this review The strategy was modified for use in The CochraneLibrary EMBASE and CINAHL (see Appendix 1 for details)

Searching other resources

We checked reference lists of articles Ongoing and unpublishedtrials were also identified by contacting researchers in the field


Selection of studies

One review author (LDG) screened the title abstract and descrip-tors of identified studies for possible inclusion From the full texttwo authors independently assessed potentially eligible trials forinclusion and resolved any disagreement through discussion Wecontacted authors for additional information if necessary

Data extraction and management

Data were independently extracted by pairs of review authors usinga pre-tested data extraction form Disagreement was resolved byconsensus or third party adjudication

Assessment of risk of bias in included studies

Two review authors independently assessed risk of bias using therecommendations in the Cochrane Handbook (Higgins 2008a)(see rsquoDifferences between protocol and reviewrsquo) The following do-mains were assessed sequence generation allocation concealmentand blinding of participants personnel and outcome assessors (forfalls and fractures) (see Higgins 2008a for criteria used for judgingrisk of bias) We also included an item assessing risk of bias in



recall of falls (Was ascertainment of fall outcomes reliable) Thiswas coded rsquoyesrsquo (low risk of bias) if the study used active registra-tion of falls for example a falls diary rsquonorsquo (high risk of bias) ifascertainment relied on participant recall at longer intervals dur-ing the study or at its conclusion and rsquounclearrsquo (uncertain risk ofbias) if there was retrospective recall over a short period only ordetails of ascertainment were not described Review authors werenot blinded to author and source institution They did not assesstheir own trials Disagreement was resolved by consensus or thirdparty adjudication

Measures of treatment effect

We used results reported at one year if these were available for trialsthat monitored falls for longer than one yearWe used the generic inverse variance method for the presentationof results and pooling of data separately for rate of falls and numberof people falling (fallers) This option enables pooling of adjustedand unadjusted treatment effect estimates (rate ratios or risk ratios)reported in the paper or calculated from data presented in thepaper The generic inverse variance option requires entering thenatural logarithm of the rate ratio or risk ratio and its standarderror we calculated these in Excel When rate ratios or risk ratioswere not provided by the authors but raw data were availablewe first used Excel to calculate an incidence rate ratio and 95confidence interval and Stata to calculate a risk ratio and 95confidence interval For cluster randomised trials we performedadjustments for clustering if this was not done in the publishedreport (see rsquoUnit of analysis issuesrsquo)

Data relating to rate of falls

For the rate of falling based on the number of falls over a period oftime if appropriate data were available we present a rate ratio and95 confidence interval for each study using the generic inversevariance option The rate ratio compares the rate of events (falls)in the two groups during the trialWe used a rate ratio (for example incidence rate ratio or hazardratio for all falls) and 95 confidence interval if these were re-ported in the paper If both adjusted and unadjusted rate ratioswere reported we have used the unadjusted estimate unless theadjustment was for clustering If a rate ratio was not reported wehave calculated this and a 95 confidence interval if appropriateraw data were reported We used the reported rate of falls (fallsper person year) in each group and the total number of falls forparticipants contributing data or we calculated the rate of fallsin each group from the total number of falls and the actual totallength of time falls were monitored (person years) for participantscontributing data In cases where data were only available for peo-ple who had completed the study or where the trial authors hadstated there were no losses to follow up we assumed that theseparticipants had been followed up for the maximum possible pe-riod

Data relating to number of fallers or participants with fall-

related fractures

For these dichotomous outcomes if appropriate data were avail-able we present a risk ratio and 95 confidence interval for eachstudy using the generic inverse variance option A risk ratio com-pares the number of participants in each group with one or morefall eventsWe used a reported estimate of effect (risk ratio (relative risk) oddsratio or hazard ratio for first fall) and 95 confidence interval ifavailable If both adjusted and unadjusted estimates were reportedwe used the unadjusted estimate unless the adjustment was forclustering If an effect estimate and 95 confidence interval wasnot reported and appropriate data were available we calculateda risk ratio and 95 confidence interval For the calculations weused the number of participants contributing data in each group ifthis was known if not reported we used the number randomisedto each group

Unit of analysis issues

Data from trials which were cluster randomised for example bymedical practice were adjusted for clustering (Higgins 2008b)using an intra-class correlation coefficient (ICC) of 001 reportedin Smeeth 2002 We ignored the possibility of a clustering effectin trials randomising by household

Assessment of heterogeneity

Heterogeneity between pooled trials was assessed using a combi-nation of visual inspection of the graphs along with considerationof the Chi2 test (with statistical significance set at P lt 010) andthe I2 statistic (Higgins 2003)

Data synthesis

We have pooled results of trials with comparable interventionsand participant characteristics using the generic inverse variancemethod in Review Manager (RevMan 5) We calculated pooledrate ratios for falls and risk ratios for fallers with 95 confidenceintervals using the fixed-effect model Where there was substantialstatistical heterogeneity we pooled the data if appropriate usingthe random-effects modelResults from trials in which participants have a single condition(eg stroke Parkinsonrsquos disease) have been included in the analyseswith the conditions shown in footnotes

Grouping of studies for data synthesis

We grouped interventions for pooling using the fall preventionclassification system that has been developed by the Preventionof Falls Network Europe ( ProFaNE) Interventions have beengrouped by combination (single multiple or multifactorial) andthen by the type of intervention (descriptors) The possible in-tervention descriptors are exercises medication (drug target ie



withdrawal dose reduction or increase substitution provision)surgery management of urinary incontinence fluid or nutritiontherapy psychological interventions environmentassistive tech-nology social environment interventions to increase knowledgeother interventions (Lamb 2007)

Subgroup analysis and investigation of heterogeneity

We minimised heterogeneity as much as possible by grouping tri-als as described previously In some categories of intervention forexample surgery data have been pooled within meaningful sub-groups eg cataract surgeryWe explored significant heterogeneity by carrying out the follow-ing subgroup analyses

bull Higher versus lower falls risk at enrolment (ie compar-ing trials with participants selected for inclusion basedon history of falling or other specific risk factors forfalling versus unselected)

bull For the multifactorial interventions we subdivided tri-als that actively provided treatment to address identi-fied risk factors versus those where the intervention con-sisted mainly of referral to other services or the provi-sion of information to increase knowledge

We used the test for subgroup differences available in RevMan 5 forthe fixed-effect model to determine if the results for subgroups werestatistically significantly different when data were pooled usingthis method We used meta-regression in Stata to test for subgroupdifferences when the random-effects model was used

Economics issues

We have noted the results from any comprehensive economic eval-uations incorporated in the included studies and report the costsand consequences of the interventions as stated by the authorsWe also extracted other healthcare cost items when reported

R E S U L T S

Description of studies

See Characteristics of included studies Characteristics ofexcluded studies Characteristics of studies awaiting classificationCharacteristics of ongoing studies

Results of the search

The search strategies identified a total of 4372 references (see Ap-pendix 1) Removal of duplicates and spurious records resulted in3200 references We obtained copies of 621 papers for considera-tion

Included studies

This review contains 111 trials with 55303 participants Detailsare provided in the Characteristics of included studies and arebriefly summarised below Due to the size of the review not alllinks to references have been inserted in the text but can be viewedin Appendix 2

Design

The majority of included studies were individually randomisedTen studies were cluster randomised by community physicianpractice retirement village or senior centre (Assantachai 2002Coleman 1999 Lord 2003 Pit 2007 Reinsch 1992 Rubenstein2007 Spice 2009 Steinberg 2000 Tinetti 1994 Wolf 2003)Four studies included individually randomised participants butalso cluster randomised by household where more than one personin the household was recruited (Brown 2002 Carpenter 1990Stevens 2001 Van Rossum 1993)

Sample sizes

Included trials ranged in sample size from 10 (Lannin 2007) to9940 (Smith 2007) The median sample size was 239 participants

Setting

Location

The included trials were carried out in 15 countries Australia (N= 20) Canada (N = 7) Chile (N = 1) China (N = 1) Finland (N =3) France (N = 3) Germany (N = 3) Japan (N = 3) Netherlands(N = 5) New Zealand (N = 5) Norway (N = 1) Switzerland (N =2) Taiwan (N = 3) Thailand (N = 2) United Kingdom (N = 22)USA (N = 29) (see Appendix 2) Latham 2003 was conducted inAustralia and New Zealand

Sampling frame

Participants were recruited using a variety of sampling frames ninetrials recruited from specialist clinics or disease registers (Ashburn2007 Campbell 2005 Foss 2006 Grant 2005 Green 2002Harwood 2005 Liu-Ambrose 2004 Sato 1999 Swanenburg2007) five from geriatric medicine or falls clinics (Cumming2007 Dhesi 2004 Hill 2000 Steadman 2003 Suzuki 2004)seven from state or private health care databases (Buchner 1997aLi 2005 Lord 2005 Luukinen 2007 Speechley 2008 Wagner1994 Wyman 2005) six recruited participants who had attendedhospital emergency departments after a fall (Close 1999 Davison2005 Kenny 2001 Kingston 2001 Lightbody 2002 Whitehead2003) and two trials enrolled some of their participants from emer-gency departments but also from a primary care setting (Hendriks2008 Prince 2008) Two trials recruited from ambulatory carecentres (Rubenstein 2000 Rubenstein 2007)



Nine trials recruited participants at discharge from in-patient careOf these three (Latham 2003 Nikolaus 2003 Pardessus 2002)included people who had been admitted for investigation of afall or who were considered frail three recruited older peoplewho had sustained a hip fracture (Harwood 2004 Huang 2005Sherrington 2004) two (Hauer 2001 Lannin 2007) recruitedprior to discharge from a rehabilitation unit and Cumming 1999recruited from hospital wards clinics and day care centresThree trials recruited from electoral rolls (Day 2002 Fabacher1994 Stevens 2001) one (Korpelainen 2006) from a birth cohortand four from retirement communities (Lord 2003 Resnick 2002Wolf 1996 Wolf 2003)Participants for 14 trials were recruited from primary care patientregisters (see Appendix 2) One study (Trivedi 2003) recruitedboth from primary care patient registers and from a database ofparticipants in a large cohort study Dukas 2004 recruited fromamongst participants in a long-standing cohort studyThe remaining 48 trials recruited by advertisement or throughsocial organisations such as senior citizens centres or reported thesampling frame as ldquocommunity dwellingrdquo (see Appendix 2)

Participants

The inclusionexclusion criteria and other participant details arelisted for each study in the Characteristics of included studiesAll participants were women in 23 trials (see Appendix 2) twotrials only recruited men (Rubenstein 2000 Speechley 2008) Theremaining studies recruited men and women in varying propor-tions with men in the majority in only nine trials (Ashburn 2007Carter 1997 Coleman 1999 Fabacher 1994 Green 2002 Huang2004 Rubenstein 2007 Schrijnemaekers 1995 Trivedi 2003)Fifty-two included studies specified a history of falling or evidenceof one or more risk factors for falling in their inclusion criteriaThe remaining 59 studies recruited participants without a spe-cific history of falling or risk factors for falling other than age orfrailty (see Appendix 2) Lower serum vitamin D ie vitamin Dinsufficiency or deficiency was an inclusion criterion in three trialsof vitamin D supplementation (Dhesi 2004 Pfeifer 2000 Prince2008)Sixty-six of the 111 included studies specifically excluded partici-pants with cognitive impairment or severe cognitive impairmenteither defined as an exclusion criterion (or its absence as an inclu-sion criterion) or implied by the stated requirement to be able togive informed consent andor to follow instructions (see Appendix2) In four trials (Close 1999 Cumming 1999 Cumming 2007Jitapunkul 1998) participants with poor cognition were includedprovided data could be obtained from carers Poor cognition wasone of a number of falls risk factors indicating eligibility for inclu-sion in Luukinen 2007In the remaining 40 studies cognitive status was not stated as aninclusion or exclusion criterion It is likely given the importanceof adequate cognition for the provision of informed consent forparticipation that the majority of participants in these studies did

not have serious cognitive impairment (see Appendix 2)Seven trials recruited on the basis of a specific condition but alsohad an age inclusion criterion severe visual impairment (Campbell2005) mobility problems one year after a stroke (Green 2002) op-erable cataract (Foss 2006 Harwood 2005) hip fracture (Huang2005) carotid sinus hypersensitivity (Kenny 2001) and Parkin-sonrsquos disease (Sato 1999) while three did not have an age inclusioncriterion Parkinsonrsquos disease (Ashburn 2007) and hip fracture (Harwood 2004 Sherrington 2004) These and 14 other trialsthat did not describe a minimum age inclusion criterion met ourinclusion criterion of having a mean age minus one standard de-viation of more than 60 years

Interventions

Interventions have been grouped by combination (single multipleor multifactorial) and then by the type of intervention (descriptors)as described in rsquoMethodsrsquo rsquoGrouping of studies for data synthesisrsquoTwenty-one trials contain more than two arms therefore trialsmay appear in more than one category of intervention (and morethan one comparison in the analyses)

Single interventions

A single intervention consists of only one major category of in-tervention which is delivered to all participants these have beengrouped by type of intervention

Exercises

Forty-three trials tested the effect of exercise on falls (see Appendix2)The ProFaNE taxonomy classifies exercises as supervised or unsu-pervised Some degree of supervision was described or could beassumed from the structure of classes in all but two trials wherethe intervention was walking (Pereira 1998 Resnick 2002) In thelatter study participants who accepted the option of walking anindoor route at an outpatients department were probably super-vised The term ldquosupervisedrdquo covers a number of different modelsof supervision ranging from direct supervision of either the indi-vidual or group of individuals while exercising to occasional (al-beit regular) telephone follow up to encourage adherence Sometrials reported initial supervision while participants were master-ing exercises but subsequent exercising was unsupervisedIn most trials the intervention was delivered in groups but in12 trials it was carried out on an individual basis (Ashburn 2007(Parkinsonrsquos disease) Campbell 1997 Campbell 1999 Green2002 (stroke) Latham 2003 Lin 2007 Nitz 2004 Protas 2005Robertson 2001a Sherrington 2004 (hip fracture) Steadman2003 Wolf 1996)The trials were grouped by exercise modality into six categoriesusing the ProFaNE taxonomy (see Table 1) In some trials the



interventions fell within one category gait balance and func-tional training (Cornillon 2002 Liu-Ambrose 2004 McMurdo1997 Wolf 1996) strengthresistance training (Fiatarone 1997Latham 2003 Liu-Ambrose 2004 Woo 2007) flexibility training(no trials included flexibility training alone) 3D training Tai Chi(Li 2005 Voukelatos 2007 Wolf 1996 Wolf 2003 Woo 2007)and square stepping (Shigematsu 2008) general physical activity(walking groups Pereira 1998 Resnick 2002 Shigematsu 2008)endurance training (no trials included endurance training alone)The remaining trials with exercise alone as an intervention in-cluded more than one category of exercise

Table 1 Categories of exercise (ProFaNE) in interventions containing exercise alone

Study ID Gait bal-

ancefunctional

training

Strength re-

sistance

training

Flexibility 3D (Tai Chi

dance etc)

General phys-

ical activity

Endurance Other

Ashburn 2007

Ballard 2004

Barnett 2003

Brown 2002

Buchner1997a

Bunout 2005

Campbell1997

Campbell1999

Carter 2002

Cerny 1998

Cornillon2002

Day 2002

Fiatarone1997



Table 1 Categories of exercise (ProFaNE) in interventions containing exercise alone (Continued)

Green 2002 physiother-apy

Hauer 2001

Helbostad2004

Korpelainen2006

dance stamping

Latham 2003

Li 2005

Lin 2007

Liu-Ambrose2004

agility traininggroup

resis-tance traininggroup

Lord 1995

Lord 2003 dance

Luukinen2007

self care

McMurdo1997

Means 2005

Morgan 2004

Nitz 2004

Pereira 1998

Reinsch 1992 standupstep up

standupstep up

Resnick 2002

Robertson2001a




Rubenstein2000

Sherrington2004

Shigematsu2008

squarestepping group

walkinggroup

Skelton 2005

Steadman2003

Suzuki 2004

Voukelatos2007

Weerdesteyn2006

Wolf 1996 bal-ance platformtraining group

Tai Chigroup

Wolf 2003

Woo 2007 resis-tance traininggroup

Tai Chigroup

indicates exercise categories in interventionldquogroupsrdquo are separate arms in the trial ie people were randomised to the separate groups

Four trials compared different exercise programmes (Nitz 2004Shigematsu 2008 Steadman 2003) or method of delivery (groupor home based) (Helbostad 2004)

Medication (drug target)

Thirteen studies (23112 enrolled participants) evaluated the effi-cacy of vitamin D supplementation either alone or with calciumco-supplementation for fall prevention (Bischoff-Ferrari 2006Dhesi 2004 Dukas 2004 Gallagher 2001 Grant 2005 Harwood2004 Latham 2003 Pfeifer 2000 Porthouse 2005 Prince 2008Sato 1999 Smith 2007 Trivedi 2003) Two studies (Grant 2005Harwood 2004) contain multiple intervention arms

Campbell 1999 in a 2 x 2 factorial design reported the resultsof an exercise programme and a placebo-controlled psychotropicmedication withdrawal programmeFalls were a secondary outcome in Gallagher 2001 in which non-osteoporotic women in one arm of the trial received hormonereplacement therapy (HRT)Greenspan 2005 also explored the effect of HRT on falls in womenwho were calcium and vitamin D repleteVellas 1991 studied the effect of administering a vaso-active medi-cation (raubasine-dihydroergocristine) to older people presentingto their medical practitioner with a history of a recent fallOne study (Meredith 2002) investigated the effect of a medicationimprovement programme based on reported problems (including



falls) relating to medication use This targeted therapeutic dupli-cation and use of NSAIDs cardiovascular and psychotropic drugsIn Pit 2007 the intervention involved general practitioners (an ed-ucational intervention to improve prescribing practices) and theirpatients (self-completed risk assessment tool relating to medica-tion) and subsequent medication review

Surgery

One trial (Kenny 2001) reported the effectiveness of cardiac pac-ing in fallers who were found to have cardioinhibitory carotid sinushypersensitivity following a visit to a hospital emergency depart-ment Two other trials investigated the effect of expedited cataractsurgery for the first eye (Harwood 2005) and second affected eye(Foss 2006)

Fluid or nutrition therapy

Gray-Donald 1995 studied the efficacy of a 12-week period ofhigh-energy nutrient-dense dietary supplementation in older peo-ple with low body mass index or recent weight loss

Psychological

Participants in one randomised arm in Reinsch 1992 received acognitive behavioural therapy intervention

EnvironmentAssistive technology

This category includes the following environmental interventions(or assessment and recommendations for intervention) adapta-tions to homes and the provision of aids for personal care and pro-tection and personal mobility aids for communication informa-tion and signalling eg eyeglasses and body worn aids for personalcare and protectionTen studies evaluated the efficacy of environmental interven-tions alone ie home safety (Campbell 2005 (severely visuallyimpaired) Cumming 1999 Day 2002 Lannin 2007 Lin 2007Pardessus 2002 Stevens 2001 Wilder 2001) interventions to im-prove vision (Cumming 2007 Day 2002) and one trial tested theYaktraxreg walker a device worn over usual footwear to increasegrip in winter outdoor conditions (McKiernan 2005)

Knowledgeeducation interventions

Two trials evaluated educational interventions designed to increaseknowledge relating to fall prevention (Robson 2003 Ryan 1996)In Robson 2003 group sessions were led by lay senior facilitatorsRyan 1996 compared nurse-led fall prevention classes with indi-vidual sessions versus a control group in a three arm trial

Multiple interventions

Multiple interventions consist of a fixed combination of two ormore major categories of intervention delivered to all participantsThis category contains 10 studies with numerous combinationsof intervention Eight trials included an exercise component com-bined with various other interventions (vitamin D (Campbell2005) education and home safety (Clemson 2004) home safetywith or without vision assessment (Day 2002) ldquoindividualisedfall prevention advicerdquo (Hill 2000) education and risk assessment(Shumway-Cook 2007) various combinations of home safetyeducation and clinical assessment (Steinberg 2000) protein en-riched nutritional supplementation and vitamin D and calcium (Swanenburg 2007) home safety (Wilder 2001)) In the two trialsthat did not contain an exercise component education was com-bined with free access to a geriatric clinic (Assantachai 2002) andhome safety was combined with medication review (Carter 1997)

Multifactorial interventions

Multifactorial interventions consist of more than one main cate-gory of intervention but participants receive different combina-tions of interventions based on an individual assessmentThis category includes 31 studies (see Appendix 2) some withmore than one intervention arm These were complex interven-tions which differed in the details of the assessment treatmentprotocols and referralThe initial assessment was usually carried out by one or morehealth professionals an intervention was then provided or recom-mendations given or referrals made for further action In Carpenter1990 and Jitapunkul 1998 the assessment and health surveillancewas carried out by a non-health professional who referred partici-pants to a health professional if a change in health status warranteditIn ten trials participants received an assessment and an active inter-vention (Close 1999 Coleman 1999 Davison 2005 Hornbrook1994 Huang 2005 Lord 2005 (extensive intervention group)Salminen 2008 Spice 2009 (secondary care intervention group)Tinetti 1994 Wyman 2005) Two of these trials (Spice 2009 Lord2005) also compared a weaker intervention involving primarilyassessment and referral with a control group Nikolaus 2003 com-pared an assessment and active intervention with assessment andreferral Twenty-one trials contained an intervention that consistedpredominantly of assessment and referral or the provision of in-formation (see Appendix 2)

Outcomes

Rate of falls were reported in 30 trials and could be calculatedfrom a further 35 trials Data on risk of falling (number of fall-ers) were available in 89 trials Some trials met our inclusioncriteria but did not include any data that could be included inthese analyses Reported results from these trials are presentedin the text Twenty-four trials reported the number of partic-



ipants sustaining a fracture five exercise trials (Ashburn 2007Campbell 1999 Korpelainen 2006 McMurdo 1997 Robertson2001a) nine vitamin D trials (Bischoff-Ferrari 2006 Gallagher2001 Grant 2005 Harwood 2004 Pfeifer 2000 Porthouse 2005Sato 1999 Smith 2007 Trivedi 2003) five trials of other sin-gle interventions (Campbell 1999 Cumming 2007 Foss 2006Harwood 2005 Kenny 2001) and six multifactorial interventions(Davison 2005 Hogan 2001 Lightbody 2002 Nikolaus 2003Tinetti 1994 Vetter 1992) The actual fractures included in theseanalyses vary Where possible we only included fall-related frac-tures (hip wrist humerus etc) and not vertebral fracture Thesource of data used for calculating outcomes for each trial forgeneric inverse variance analysis is shown in Appendix 3

Excluded studies

The Characteristics of excluded studies lists 61 studies Fourteenstudies reporting falls outcomes were excluded because they werenot RCTs Of the identified RCTs seven reported falls outcomesbut did not meet the reviews inclusion criterion for age (ie par-ticipants were too young and results were not presented by agegroup) Five trials with falls outcomes were excluded because themajority of participants were not community dwelling Nine stud-ies were excluded because they did not report falls outcomes fivewere excluded because the reported falls were artificially inducedin a laboratory eg during balance testing and 13 were excludedbecause although they reported falls the intervention was not de-

signed to reduce falls Eight other RCTs were excluded for a vari-ety of reasons (Graafmans 1996 Iwamoto 2005 Larsen 2005 Lee2007 Lehtola 2000 Means 1996 Peterson 2004 Protas 2005)

Ongoing studies

We identified 34 trials that are either ongoing or completedbut unpublished in which falls appear to be an outcome (seeCharacteristics of ongoing studies for details) Sixteen are inves-tigating single interventions nine trials of exercises including TaiChi and exercises for vestibular rehabilitation and seven investi-gating other single interventions (enhanced podiatric care a cog-nitive behavioural intervention home safety surgery for pace-maker insertion vitamin D supplementation and two with visualimprovement interventions) Four trials contain various multiplecombinations of intervention one of which is in people who havehad a hip fracture and thirteen include a multifactorial interven-tion two of which are in people who have had a stroke

Studies awaiting classification

Six studies are awaiting classification (see Characteristics of studiesawaiting classification)

Risk of bias in included studies

Details of risk of bias assessment for each trial are shown in theCharacteristics of included studies Summary results are shown inFigure 1



Figure 1 Methodological quality summary review authorsrsquo judgments about each methodological quality

item for each included study



Allocation

We assessed risk of bias in sequence generation as low in 55 ofincluded studies high in only 2 but unclear in the remainingstudies Concealment of allocation prior to group assignment wasjudged to carry low risk of bias in 32 of studies high in 5 andto be unclear in the reports of the remaining 63 of studies (seeFigure 2)

Figure 2 Methodological quality graph review authorsrsquo judgments about each methodological quality item

presented as percentages across all included studies

Blinding

As less than 15 of included studies were placebo controlled par-ticipants would have known their allocation status in most in-cluded studies and falls are self reported Regular contact is a fea-ture of well-conducted research on fall prevention and outcomeassessors may learn of the participantrsquos group allocation in con-versation It is difficult to assess the impact of that fact on ascer-tainment bias one would anticipate that it would be small Weassessed the risk and potential impact of bias as a result of un-blinding of participants or outcome assessors to be unclear for falloutcomes in 80 of studies (see Figure 2)

Other potential sources of bias

Bias in recall of falls

Fifty per cent of included studies were assessed as being at low riskof bias in the recall of falls ie they included active registrationof falls outcomes or use of a diary In 30 of studies there waspotential for a high risk of bias in that ascertainment of fallingepisodes was by participant recall at intervals during the study orat its conclusion In 20 of studies the risk of bias was unclearas retrospective recall was for a short period only or details ofascertainment were not described (see Figure 2)

Effects of interventions


Single interventions consist of only one major category of interven-tion and are delivered to all participants these have been groupedby type of intervention and data have been pooled within types

Exercises

The trials were grouped by exercise modality into six categoriesusing the ProFaNE taxonomy (see Table 1)

Exercise versus control

Exercise classes containing multiple components (ie a combina-tion of two or more categories of exercise) achieved a statisti-cally significant reduction in rate of falls (pooled rate ratio (RaR)078 95 confidence interval (CI) 071 to 086 2364 partici-pants 14 trials Analysis 111) and risk of falling (pooled risk ratio(RR)(random effects) 083 95 CI 072 to 097 2492 partic-ipants 17 trials Analysis 121) The random-effects model wasused to pool data in Analysis 12 due to the combination of sub-stantial amount of heterogeneity present in Analysis 121 (P =0006 I2= 52) and clinical heterogeneity in the interventionsbeing combined



We carried out an a priori subgroup analysis of these group exercisetrials with multiple components based on falls risk at enrolmentand found there was no difference in pooled estimates betweentrials with participants at higher risk of falling (history of fallingor one or more risk factors for falls at enrolment) versus lowerrisk (unselected on falls risk at enrolment) The intervention waseffective in both subgroups for rate of falls (Analysis 21) Forrisk of falling (Analysis 22) the intervention was significant inthe higher risk subgroup but not in the subgroup not so selectedhowever the difference between subgroups was not significant (P= 0684)Home-based exercises including more than one exercise categoryalso achieved a statistically significant reduction in rate of falls(RaR 066 95 CI 053 to 082 666 participants 4 trials Anal-ysis 112) and in risk of falling (RR (random effects) 077 95CI 061 to 097 566 participants 3 trials Analysis 122) Thelatter analysis does not contain two trials with home-based inter-ventions Ashburn 2007 in which all the participants had Parkin-sonrsquos disease and Green 2002 in which all participants had mobil-ity problems one year after a stroke The intervention in Ashburn2007 consisted of hourly sessions with a physiotherapist for sixweeks which resulted in no significant reduction in the number ofpeople falling (RR 094 95 CI 077 to 115 126 participantsAnalysis 123) The intervention in Green 2002 consisted of com-munity physiotherapy compared with usual care which resultedin a non-significant increase in the number of people falling (RR130 95 CI 083 to 204 170 participants Analysis 124)Although considered to be a single category of exercise interven-tion Tai Chi also contains a combination of both strength andbalance training There is evidence that Tai Chi can significantlyreduce both rate of falls (RaR 063 95 CI 052 to 078 1294participants 4 trials Analysis 113) and risk of falling (RR (ran-dom effects) 065 95 CI 051 to 082 1278 participants 4 tri-als Analysis 125)In the remaining trials the intervention was within only one ofthe categories of exercise using the ProFaNE classification Classesthat included just gait balance or functional training significantlyreduced rate of falls (RaR 073 95 CI 054 to 098 461 par-ticipants 3 trials Analysis 114) but not risk of falling (RR (ran-dom effects) 077 95 CI 058 to 103 461 participants 3 trialsAnalysis 126) None of the remaining comparisons achieved astatistically significant reduction in rate of falls or risk of fallingStrengthresistance training delivered in a group setting failed to

achieve a significant reduction in rate of falls (64 participants 1trial Analysis 115) or number of people falling (184 participants2 trials Analysis 127) The intervention in Fiatarone 1997 alsoconsisted of high intensity progressive resistance training in groupsessions but there were insufficient data to include in the meta-analysis The authors reported that ldquono difference between groupswas observed in the frequency of fallsrdquo Home-based resistancetraining in Latham 2003 also failed to achieve a statistically signif-icant reduction in rate of falls (222 participants Analysis 116)and risk of falling (Analysis 128) This trial also reported thatmusculoskeletal injuries were significantly more common in thegroup participating in resistance exercise training (interventiongroup 18112 (16) versus control group 5110 (5) RR 35495 CI 136 to 919) Two trials investigated the effect of gen-eral physical activity in the form of walking groups (Pereira 1998Resnick 2002) There was no reduction in risk of falling in Pereira1998 (Analysis 129) and Resnick 2002 contained insufficientdata to include in an analysis but reported no significant differencein number of fallsPooled data for risk of fracture shows a statistically significantreduction from exercise interventions (RR 036 95 CI 019 to070 719 participants 5 trials Analysis 13) The result remainssignificant when Ashburn 2007 (in which all the participants hadParkinsonrsquos disease) is removed from the analysis The results aredominated by the data from Korpelainen 2006 in which six women(7) in the intervention group and 15 (20) in the control groupsustained a fracture

Exercise versus exercise

Four trials compared different types of exercise or methods ofdelivery There was no significant reduction in rate of falls (Analysis31) or risk of falling (Analysis 32) in any of these trials


Supplementation with vitamin D

Thirteen studies (23112 enrolled participants) evaluated the ef-ficacy for fall prevention of supplementation with vitamin Dor an analogue either alone or with calcium co-supplementa-tion (Bischoff-Ferrari 2006 Dhesi 2004 Dukas 2004 Gallagher2001 Grant 2005 Harwood 2004 Latham 2003 Pfeifer 2000Porthouse 2005 Prince 2008 Sato 1999 Smith 2007 Trivedi2003) (see Table 2 for reported baseline vitamin D levels)



Table 2 Mean baseline vitamin D levels (25(OH)D) in included trials (nmolL)

Study Overall Intervention Control Men Women Selection crite-

rion

Bischoff-Ferrari2006

747 (SD 383) NA NA 829 (SD 449) 664 (SD 317) No

Dhesi 2004 (range 237 to280)

267 (range 255to 280)

250 (range 237to 261)

NA NA Yes25(OH)Dle30

Dukas 2004 726 (SD 279) 746 (SD 290)

706 (SD 267) NA NA No

Gallagher 2001 793 (SD 247) 780 (SD216)

805 (SD 274) NA NA No

Grant 2005 388 (SD 156) 380 (SD 163) 395 (SD 148) NA NA No

Harwood 2004 295 (range 6 to85)

29 (range 6 to85)

30 (range 12 to64)

NA 29 (range 6 to 85) No

Latham 2003 374 (95 CI349 to 449)

474 (95 CI399 to 524)

NA NA No

Pfeifer 2000 252 (SD 129) 257 (SD 136) 246 (SD 121) NA NA Yes25(OH)D lt50

Porthouse 2005 NA NA NA NA NA No

Prince 2008 448 (SD 127) 452 (SD 125) 443 (SD 128) NA NA Yes25(OH)Dlt599

Sato 1999 285 (SD 161) 275 (SD 148) 295 (SD 173) NA NA No(Parkinsonrsquos dis-ease)

Smith 2007 NA NA NA NA NA No

Trivedi 2003 NA NA NA NA NA No

Data from two trial centres only (random as stratified by trial centre) Converted from ngmL (ngmL x 2496 = nmolL) Calcitriol alone intervention groupNA not available25(OH)D 25-hydroxyvitamin D

The overall analysis of vitamin D versus control did not show a



statistically significant difference in rate of falls (RaR (random ef-fects) 095 95 CI 080 to 114 3929 participants 5 studiesAnalysis 41) risk of falling (RR (fixed effect) 096 95 CI 092to 101 21110 participants 10 studies Analysis 42) or risk offracture (RR 098 95 CI 089 to 107 21377 participants 7studies Analysis 43) Adverse effects (hypercalcaemia renal dis-ease gastrointestinal effects) were reported in three trials but nonewere statistically significant (Analysis 44)

A pre-planned subgroup analysis showed no significant differencein either rate of falling (Analysis 51) or risk of falls (Analysis 52)in trials recruiting participants with higher falls risk or trials notso doing and no significant difference in effect size between thesubgroups in either analysis (Analysis 51 and Analysis 52)

We carried out a post hoc subgroup analysis to explore the effectof only enrolling participants with lower vitamin D levels Datafor rate of falls were pooled using the random-effects model asthere was substantial heterogeneity in the subgroup of trials notselecting on the basis of vitamin D levels (I2 = 63 P = 007)The rate of falls (Analysis 61) was significantly reduced in trialsrecruiting participants with lower vitamin D levels (RaR 057037 to 089 260 participants 2 trials) but not in participants notso selected (RaR 102 95 CI 088 to 119 3669 participants3 trials) There was a significant difference between these twosubgroups with a greater reduction in rate of falls in the subgroupof trials only recruiting participants with lower vitamin D levels (P= 001) There was insignificant heterogeneity in the analysis forrisk of falling (Analysis 62) which was significantly reduced inthe lower vitamin D group (RR 065 95 CI 046 to 091 562participants 3 trials) but not in those not so selected (RR 097092 to 102 20548 participants 7 trials) The test for subgroupdifferences was significant (P = 002)

Supplementation with a vitamin D analogue

For vitamin D analogues (calcitriol (125 dihydroxy-vitamin D)and alfacalcidol (1-alpha hydroxyl vitamin D)) there was no ev-idence of effect for alfacalcidol on rate of falls (80 participants1 trial Analysis 711) or risk of falling (378 participants 1 trialAnalysis 721) but a statistically significant reduction in the num-ber of people sustaining a fracture (RR 013 95 CI 002 to 08980 participants Analysis 73) In participants taking calcitriol therewas a statistically significant reduction in rate of falls (RaR 06495 CI 049 to 082 213 participants 1 trial Analysis 712) andrisk of falling (RR 054 95 CI 031 to 093 213 participants 1trial Analysis 722) There was however a statistically significantincrease in the risk of hypercalcaemia with these analogues (RR233 95 CI 102 to 531 624 participants 2 trials Analysis74)

Other medication (drug target) interventions

Gradual withdrawal of psychotropic medication in a placebo-con-trolled trial significantly reduced rate of falls (RaR 034 95 CI016 to 073 93 participants 1 trial Analysis 811) but not riskof falling (RR 061 95 CI 032 to 117 Analysis 821) or riskof fracture (RR 283 95 CI 012 to 6770 Analysis 831)There is no evidence to support the use of HRT for reducing rate offalls (212 participants 1 trial Analysis 812) or risk of falling (585participants 2 trials Analysis 822) An intervention involvingmedication review and modification was not effective in reducingrisk of falls (259 participants 1 trial Analysis 823)Pit 2007 included an major educational component for familyphysicians that included academic detailing feedback on prescrib-ing practices and financial rewards This combined with self-as-sessment of medication use by their patients and subsequent med-ication review and modification resulted in a significantly reducedrisk of falling (RR 061 95 CI 041 to 091 659 participantsAnalysis 824)Vellas 1991 (95 participants) reported that participants with ahistory of a recent fall who received six months of therapy withthe vaso-active medication raubasine-dihydroergocristine ldquoshowedfewer new falls than the group receiving placebordquo however insuf-ficient data were reported to determine whether this was a signif-icant reduction

Surgery

Cardiac pacemaker insertion

Cardiac pacing in fallers with cardioinhibitory carotid sinus hy-persensitivity (Kenny 2001) was associated with a statistically sig-nificant reduction in rate of falls (RaR 042 95 CI 023 to 075171 participants Analysis 911) but not in number of peoplesustaining a fracture (Analysis 931)

Cataract surgery

In Harwood 2005 there was a significant reduction in rate of fallsin people receiving expedited cataract surgery for the first eye (RaR066 045 to 095 306 participants Analysis 912) but not inrisk of falling (RR 095 95 CI 068 to 133 Analysis 921) orrisk of fracture (Analysis 932) In participants receiving cataractsurgery for a second eye (Foss 2006) there was no evidence ofeffect on rate of falls (239 participants Analysis 913) risk offalling (Analysis 922) or risk of fracture (Analysis 933)


In Gray-Donald 1995 risk of falling was not significantly reducedin frail older women receiving oral nutritional supplementation(46 participants Analysis 101)

Psychological



The cognitive behavioural intervention in Reinsch 1992 did notresult in a statistically significant reduction in risk of falling (230participants Analysis 111)


Environment (home safety and aids for personal mobility)

Six studies contributed data on the effectiveness of home hazardmodification in participants not selected on the basis of a specificcondition (Cumming 1999 Day 2002 Lannin 2007 Lin 2007Pardessus 2002 Stevens 2001) Home safety interventions did notresult in a statistically significant difference in rate of falls (RaR090 95 CI 079 to 103 2367 participants 3 trials Analysis1211) or number of people falling (RR 089 95 CI 080 to100 2610 participants 5 trials Analysis 1221) Wilder 2001did not report any results for the group receiving ldquosimple homemodificationsrdquo versus control Data for fractures were not availableIn participants with severe visual impairment (visual acuity 624or worse) (Campbell 2005) a home safety programme significantlyreduced the rate of falls (RaR 059 95 CI 042 to 082 391participants Analysis 1212) and number of fallers (RR 07695 CI 062 to 095 391 participants Analysis 1222)We carried out a subgroup analysis by falls risk at enrolment totest whether the intervention effect was greater in participantsat higher risk of falling ie with a history of falling or one ormore risk factors Rate of falling (Analysis 131) was significantlyreduced in the higher risk subgroup (Campbell 2005 Lin 2007)(RaR 056 95 CI 042 to 076 491 participants) but not thelower risk subgroup (Cumming 1999 Stevens 2001) (RaR 09295 CI 080 to 106 2267 participants) There was a statisticallysignificant difference between subgroups with a greater reductionin rate of falling in the higher risk group (Chi2 = 842 P = 0004 I2

= 881) The risk of falling (Analysis 132) was also significantlyreduced in the higher risk subgroup (Campbell 2005 Pardessus2002) (RR 078 95 CI 064 to 095 451 participants) but notthe lower risk subgroup (RR 090 95 CI 080 to 100 4 trials2550 participants) although in this case the test for subgroupdifferences was not significant (Chi2 = 145 P = 023 I2 = 310)

Environment (aids for communication information and

signalling)

Two trials (Cumming 2007 Day 2002) investigated the effect ofinterventions to improve vision In Cumming 2007 this involvedvision assessment and eye examination and if required the provi-sion of new spectacles referral for expedited ophthalmology treat-ment mobility training and canes This intervention resulted in astatistically significant increase in both rate of falls (RaR 157 95CI 119 to 206 616 participants Analysis 1213) and numberof participants falling (RR 154 95 CI 124 to 191 Analysis1223) There was also an increase in risk of fracture although thiswas not statistically significant (RR 173 95 CI 096 to 312

Analysis 123) Day 2002 compared people who received a visualacuity assessment and referral with those who did not There wasno significant reduction in risk of falling (276 participants Anal-ysis 1224)

Environment (body worn aids for personal care and

protection)

McKiernan 2005 tested the effect of wearing a non-slip device( Yaktraxreg walker) on outdoor shoes in winter conditions andachieved a statistically significant reduction in rate of outdoorfalls (RaR 042 95 CI 022 to 078 109 participants Analysis1214)


Two trials tested interventions designed to reduce falls by increas-ing knowledge about fall prevention (Robson 2003 Ryan 1996)There was no evidence of reduction in rate of falls (45 participants1 trial Analysis 141) or risk of falling (516 participants 2 trialsAnalysis 142)


Multiple interventions consist of a fixed combination of majorcategories of intervention delivered to all participants these havebeen grouped by combinations of interventions for analysis andeach combination analysed separatelyAll trials with rate of falls outcomes (Analysis 151) included anexercise component of varying intensity combined with one ormore other interventions Clemson 2004 using a combinationof exercise education and a home safety intervention achieved asignificant reduction in rate of falls (RaR 069 95 CI 050 to096 285 participants Analysis 1514) Swanenburg 2007 inves-tigated the effect of exercise plus nutritional supplementation invitamin D and calcium replete women Although a highly signif-icant reduction in rate of falls was achieved (RaR 019 95 CI005 to 068 20 participants Analysis 1515) these results shouldbe treated with caution due to the small sample size None of theremaining comparisons in Analysis 151 achieved a significant re-duction in rate of falls including Campbell 2005 in which theintervention consisted of the Otago Exercise Programme and vi-tamin D in participants with severe visual impairmentThirteen different combinations of interventions provided data onrisk of falling (Analysis 152) of which 11 contained an exercisecomponent In Day 2002 the risk of falling was significantly re-duced in the three arms receiving an exercise component exerciseplus home safety (RR 076 95 CI 060 to 097 272 participantsAnalysis 1521) exercise plus vision assessment (RR 073 95CI 059 to 091 273 participant Analysis 1522) and exerciseplus vision assessment plus home safety (RR 067 95 CI 051to 088 272 participants Analysis 1523) In Assantachai 2002there was a statistically significant reduction in risk of falling in aneducational intervention combined with free access to a geriatric



clinic in Thailand (RR 077 95 CI 063 to 094 815 partici-pants Analysis 1529) but in the remaining combinations of in-terventions in Analysis 152 there was no significant reduction inthe number of people falling Wilder 2001 did not contain databut reported ldquopost hoc testsrdquo which showed that the home safetyand exercise group was ldquosignificantly different from the other twogroupsrdquo (control group and ldquosimple home modificationrdquo group)in number of falls


Multifactorial interventions consist of more than one main cate-gory of intervention but participants receive different combina-tions of interventions based on an individual assessment Thesetrials have been grouped together as each contains numerous dif-ferent combinations of intervention based on individual assess-mentMultifactorial interventions significantly reduced the rate of falls(RaR (random effects) 075 95 CI 065 to 086 8141 partici-pants 15 trials Analysis 161) but there is substantial heterogene-ity between individual studies in the pooled data (I2 = 85 P lt000001) Review of the funnel plot (see Figure 3) shows two out-liers (Carpenter 1990 Close 1999) When both are removed fromthe analysis heterogeneity is reduced (I2 = 52 P = 002) butthe results remain significant (RaR (random effects) 082 95CI 076 to 090) Current evidence does not confirm a significantreduction in risk of falling (RR (random effects) 095 95 CI088 to 102 11173 participants 26 trials Analysis 162) or riskof fracture (RR 070 95 CI 047 to 104 2195 participants 7trials Analysis 163) There were insufficient data in Van Rossum1993 to include this study in these analyses The authors reportedldquono differences between the two groups with respect to these healthaspectsrdquo which included falls Vetter 1992 also contained insuffi-cient data for inclusion in these analyses and reported ldquono differ-ence between groupsrdquo



Figure 3 Funnel plot of Analysis 161 Multifactorial intervention after assessment vs control Rate of falls



The pre-planned subgroup analysis by falls risk at enrolmentshowed no evidence of difference in treatment effect between sub-groups for both rate of falls (Analysis 171) and risk of falling(Analysis 172)The pre-planned subgroup analysis by scope and intensity of in-tervention showed no evidence of difference in treatment effectbetween subgroups for both rate of falls (Analysis 181) and riskof falling (Analysis 182)

Economic evaluations

A total of 15 studies included in this review reported the costeffectiveness of the intervention the cost of delivering the inter-vention or other healthcare cost items as an outcome measure (seeAppendix 4 for details) A comprehensive cost effectiveness eval-uation with the control group as the comparator was reported ineight studies A further four studies provided the cost of deliveringthe intervention and a total of 12 of the 15 studies reported otherhealthcare resource cost itemsA cost effectiveness analysis compares the costs and consequencesof alternative treatments or approaches with the same clinicallyrelevant outcome (eg falls) Cost effectiveness was established fora home safety assessment and modification programme deliveredto those with severe vision loss in Campbell 2005 and those re-cently in hospital in Cumming 1999 (Salkeld 2000) 16 weeksof Tai Chi classes in Voukelatos 2007 (Haas 2006) a multifacto-rial programme in Tinetti 1994 (Rizzo 1996) the Otago ExerciseProgramme in Campbell 1997 (Robertson 2001c) and Robertson2001a the double blind gradual withdrawal of psychotropic medi-cation in Campbell 1999 (Robertson 2001b) and first eye cataractsurgery within one month after randomisation compared with theroutine 12-month wait in Harwood 2005 (Sach 2007) The timeperiod for these analyses was the trial duration but the perspec-tives taken and the cost items measured and methods for valuingthe items varied so that comparison of incremental cost effective-ness ratios for the interventions (cost per fall prevented) is difficulteven for evaluations carried out within similar health systemsThe results from three studies demonstrated the potential forcost savings from delivering the intervention (Cumming 1999Robertson 2001a Tinetti 1994) One trial of the Otago ExerciseProgramme showed savings in the costs of hospital admissions asa result of falls (Robertson 2001a) and the incremental cost effec-tive ratios for particular high risk subgroups of older people wasless than zero (indicating cost savings) in two studies (Cumming1999 Tinetti 1994) The incremental cost effectiveness ratio forfalls prevented indicated cost savings for a home safety programme(Cumming 1999) when delivered to the subgroup of participantswith a previous fall (Salkeld 2000) A multifactorial intervention(Tinetti 1994) was cost saving for those with four or more of theeight targeted risk factors but not for those with fewer risk factorsboth in terms of number of falls prevented and falls resulting inmedical treatment prevented (Rizzo 1996)In addition a cost utility analysis was reported for the study thattested first eye cataract surgery (Harwood 2005) Cost utility anal-

ysis compares outcomes in terms of quality adjusted life years(QALYs) gained The incremental cost utility ratio was pound35704(at 2004 prices) which is above a currently accepted UK thresholdof willingness to pay per QALY gained of pound30000 (Sach 2007)However if the time period of the analysis was extended fromthe 12-month trial period and modelled for the personrsquos expectedlifetime the incremental cost per QALY gained was much lowerat pound13172

D I S C U S S I O N

In this review through the use of the generic inverse variancemethod for the analyses we have been able to include data onboth rate of falls and risk of falling and appropriately adjusteddata from cluster randomised studies We believe that this offersmore confidence in the overall results and thus in the conclusionsdrawn from them

In the analyses we used a mix of reported rate ratios (N = 30trials) and rate ratios we calculated from raw data when thesewere available (N = 35 trials) (see Appendix 3 for details) Wedid a sensitivity analysis testing the effect of removing calculatedrate ratios Removing these from the analyses did not change thesignificance of the results (analysis not shown)

Statistical and clinical heterogeneity in our analyses presentedsome difficulties particularly for multifactorial interventions dueto variation in populations sampled and particularly to the de-tails of the nature and context of the intervention studied Inthe previous review covering this topic (Gillespie 2003) we notedthat ldquoas the number of studies has increased the picture beginsto emerge that interventions which target an unselected group ofolder people with a health or environmental intervention on thebasis of risk factors or age are less likely to be effective than thosewhich target known fallersrdquo We approached the problem of clini-cal heterogeneity through planned subgroup analyses which wereconducted in four intervention categories exercise the adminis-tration of vitamin D environmental interventions (home safety)and multifactorial interventions

Summary of main results

Exercises

Overall multiple-component exercise interventions are effectivein reducing rate and risk of falling Subgroup analysis failed toidentify evidence of difference between studies targeting peoplewith known falls risk or people who were not enrolled on thebasis of risk interventions containing multiple components ofexercise were effective in reducing both rate and risk of falls inboth subgroups Within the exercise category there is evidence forthe effectiveness of three different approaches in reducing bothrate of falls and risk of falling multiple component group exercise



Tai Chi as a group exercise and individually prescribed multiplecomponent exercise carried out at homeMedication (drug target)

Vitamin D supplementation

Despite evaluation in a number of large studies the effectivenessof vitamin D for reducing falls with or without calcium remainsunclear In the overall analysis and in the subgroup analysis com-paring participant populations with higher and lower falls risk atenrolment we found that vitamin D did not significantly reduceeither rate of falls or risk of falling However subgroup analysisshowed that when administered to older people selected on thebasis of low vitamin D level supplementation was effective in re-ducing rate of falls and risk of falling This significant findingshould be considered provisional until data from additional trialsbecomes available as the subgroup differences are based on sub-groups containing only two (Analysis 611) and three (Analysis621) trialsVitamin D analogues (calcitriol (125 dihydroxy-vitamin D) andalfacalcidol (1-alpha hydroxyl vitamin D) may be effective but theevidence base is limited and their use is associated with a signifi-cantly raised incidence of reported hypercalcaemia compared withplacebo (Dukas 2004 Gallagher 2001)Other medication interventions

An educational programme for primary care physicians on med-ication use significantly reduced risk of falling in older peopleunder their care (Pit 2007) Gradual withdrawal of psychotropicmedication reduces rate of falls but not risk of falling (Campbell1999)


Home safety interventions failed to significantly reduce rate offalls or risk of falling although subgroup analysis by falls risk atenrolment suggests that these interventions may be effective inparticipants who are at higher risk (Campbell 2005 Lin 2007Pardessus 2002) compared with those not selected on the basis ofriskAn anti-slip shoe device for icy conditions significantly reducedwinter outside falls (McKiernan 2005)


We found that assessment and multifactorial intervention is effec-tive in reducing rate of falls but does not overall have a signifi-cant effect on risk of falling Using subgroup analyses we exploredwhether recruitment by falls risk was important and whether theintensity of the intervention might be important Heterogeneitybetween studies in the multifactorial category was high and wedecided that pooling of data using the random-effects model waspreferable This did not confirm significant differences betweensubgroups for recruitment by risk or for intensity of interventionThe effectiveness of multifactorial interventions may be sensitiveto differences between health care systems structures and net-works at local and national level Hendriks 2008 reported the re-sults of a study which aimed to reproduce in The Netherlands

the successful integrated multifactorial intervention reported byClose 1999 from the UK The major differences in the health op-erational networks in The Netherlands health system comparedwith those in the UK appear to have made timely direct contactwith the appropriate health professionals impossible to achieve (Lord 2008) That risk of falling was not reduced in Hendriks 2008may be due to these systematic differences rather than to samplevariation as negative results were also reported by Van Haastregt2000 and Van Rossum 1993 in the same health-care settingPrevention of falling in people with particular health

problems

Poor vision

For people with poor vision home safety intervention appearseffective in reducing both rate of falls and risk of falling (Campbell2005) The effectiveness of other interventions for this group ofolder people is uncertain Accelerating first eye cataract surgeryfor older people on a waiting list significantly reduced rate of fallscompared with waiting list controls (Harwood 2005) but thereduction in number of fallers was not significant Acceleratingsecond eye surgery did not significantly reduce either measure (Foss2006) Assessment and correction of visual impairment did notreduce falls in two trials (Cumming 2007 Day 2002) Indeed theintervention in Cumming 2007 resulted in a significant increasein both rate and risk of falling A number of possible reasons forthis are discussed in Cumming 2007 including the fact that neweyeglasses were the most common intervention in this study andmost required major changes in prescription The trialists suggestthat rdquoold frail people may need a considerable period of time toadjust to new eyeglasses and could be at greater risk of fallingduring this timeldquoCardiovascular disorders

Cardiac pacing in people with carotid sinus hypersensitivity and ahistory of syncope andor falls reduces rate of falls (Kenny 2001)Neurological disorders

Risk of falling was not significantly reduced by home-based phys-iotherapy for people with Parkinsonrsquos disease (Ashburn 2007) orcommunity physiotherapy for people with stroke-related mobilityproblems (Green 2002) Vitamin D analogues were not effectivein reducing rate of falls in people with Parkinsonrsquos disease (Sato1999)

Post hip fracture

The vitamin D intervention in Harwood 2004 was effective inreducing the number of people who fell after a hip fracture butneither discharge planning by a specialist gerontological nurse (Huang 2005) nor physiotherapist prescribed home-based exer-cises (Sherrington 2004) were effective in reducing the numberof people fallingEconomic evaluations

In eight studies the authors had reported a comprehensive eco-nomic evaluation which provided an indication of value for money



for the interventions being tested but variations in the methodsused makes comparison of the incremental cost-effectiveness ra-tios across studies difficult There was some although limited ev-idence that falls prevention strategies can be cost saving during thetrial period and may also be cost effective over the participantsrsquoremaining lifetime The results indicate that to obtain maximumvalue for money effective strategies need to be targeted at partic-ular subgroups of older people

Overall completeness and applicability ofevidence

We sought data for rate of falls number of people falling andnumber of people sustaining a fracture However few studies pro-vided fracture data As the analyses and Appendix 3 demonstratesome studies provided data for both falls and fallers but othersprovided data only for one or other fall outcome In most inter-ventions we were able to pool more data on risk of falling thanon rate of falls Since robust statistical methods are now availableto deal with comparison of the number of falls occurring in eachgroup of a study the use of rate of falls has a number of attractionsFirst it improves power In the sense that every fall carries a riskof injury an intervention which reduces the number of times thefallers fall even if not the number of fallers has clinical publichealth and economic relevance But from a public health perspec-tive fall prevention lies across the threshold between primary andsecondary prevention Older people who are not yet ldquofallersrdquo how-ever defined might wish to know how best to prolong the timeuntil they cross the threshold For this reason and because currentconsensus recommends that both outcomes be collected (Lamb2005) we have provided meta-analyses for both using generic in-verse varianceThis review shows that the effect of exercise programmes in re-ducing the risk and rate of falling should now be regarded as es-tablished Heterogeneity between studies in this category was notlarge given the complex nature of these interventions Howeverfurther research exploring the best combination of componentswithin the exercise category might be justified Trials need to belarge in order to have power to discern any differencesThe place of vitamin D supplementation with or without calciumin fall prevention remains somewhat unclear We found no overallevidence of effectiveness in fall prevention in older people livingin the community The evidence for effectiveness in reducing rateof falls in participants selected for study inclusion on the basis oflow vitamin D levels although statistically significant is limitedbeing derived from a sub-group analysis comparing data fromonly 260 participants (selected for study inclusion on the basisof low vitamin D) with 21100 participants not so selected Thedefinition of low vitamin D and the level of supplementationdiffered between studies The findings of this subgroup analysisindicate that further research appears justified to establish the cost-effectiveness of administration of vitamin D to older people with

low serum vitamin D levelsAssessment with individualised multifactorial intervention pro-grammes overall appear effective in reducing the rate of falls instudies from different health care systems However further re-search appears justified to explore the difference between pro-grammes which provide integration of assessment and interven-tion by a multidisciplinary team and programmes which provideassessment but rely on referral to other providers and agencies forthe interventionAs the majority of trials specifically excluded older people whowere cognitively impaired the results of this review may not begeneralisable to this important group of people at risk Researchon the impact of management programmes for other risk factorssuch as cognitive impairment and urinary incontinence on riskand rate of falling appears justifiedFurther research appears justified to confirm the emerging evi-dence of effectiveness of home safety interventions in high-riskgroups and to clarify the impact of strategies to optimise care forpeople with different visual impairments

Quality of the evidence

Falls trials are difficult to design but conduct and methodologycould be improved considerably The fact that the outcome ofinterest falling was not always defined is a continuing concernThe use of two definitions in Wolf 1996 demonstrated that thedefinition of falling used can alter the significance of the resultsA consensus definition of a fall such as the one developed by thePrevention of Falls Network Europe (Lamb 2005) needs to beadopted in order to facilitate comparisons of research findingsThe included studies also illustrated the wider problems of varia-tion in the methods of ascertaining recording analysing and re-porting falls described in the Hauer 2006 systematic review Rec-ommendations on how these should be approached are also con-tained in Lamb 2005We included many small studies and were able through the use ofgeneric inverse variance to pool data from cluster randomised andfactorial studies A clearer framework for standards is emergingStudies evaluating fall prevention should be adequately poweredand use a contemporary standard for definition of a fall methodsof ascertainment recording analysis and reporting of data (Lamb2005)Design and reporting of trials should meet the contemporary stan-dards of the CONSORT statement (Boutron 2008) includingthose for cluster-randomised trials (Campbell 2004)Where factorial designs are employed data for each treatment cellshould be reported to allow interpretation of possible interactionsbetween different intervention components (McAlister 2003)

Potential biases in the review process

During the preparation of the review we attempted to minimisepublication bias but encountered a number of other potential



biases Although our search was comprehensive and we includedstudies identified in languages other than English we cannot ruleout the possibility that some studies have been missed We ob-tained unpublished falls data from a number of studies and weincluded four abstracts which have yet to be published as full pa-pers (Cerny 1998 Fiatarone 1997 Hill 2000 Wilder 2001) Weconstructed funnel plots from analyses of rate ratio and risk ra-tio for four larger categories of study For exercise interventionsasymmetry in the funnel plots is slight For vitamin D administra-tion home safety interventions and multifactorial interventionsthe plots are somewhat asymmetric suggesting the possibility ofnegative publication biasMany studies were reported in more than one paper but in the ma-jority of cases the relevant outcome data were available in a singlepaper A small number of studies reported data more than oncesometimes with apparent small discrepancies which required care-ful interpretation or communication with authors Ten excludedtrials reported falls as adverse effects although in some instancesthe intervention might plausibly have reduced falls This raises thepossibility of a form of outcome reporting bias Increased publi-cation of protocols in trials registers will make it easier to establishthe a priori hypotheses

Agreements and disagreements with otherstudies or reviews

Seven relevant systematic reviews published since 2006 were iden-tified through our search for randomised trials for inclusion (Beswick 2008 Campbell 2007 Gates 2008 Goodwin 2008Jackson 2007 Richy 2008 Sherrington 2008)

Exercise

Two systematic reviews addressed the effectiveness of exercise in-terventions Goodwin 2008 in a review of exercise in people withParkinsonrsquos disease identified two trials with falls outcomes bothidentified for this review Ashburn 2007 was included and Protas2005 (with 18 participants) was excluded from this review (seeCharacteristics of excluded studies)Sherrington 2008 pooled data from 44 trials with 9603 partici-pants and found a significant reduction in rate of falls (RaR 08395 CI 075 to 091) They found greater relative effects in pro-grammes that included exercises which challenged balance used ahigher dose of exercise or did not include a walking programmeAlthough their inclusion criteria and methods of analysis differedsomewhat from ours the overall findings are similar


We identified three systematic reviews Beswick 2008 focused onmultifactorial interventions and included 12 trials with falls out-comes all of which are included in this review They found thatrisk of falling was reduced (RR 092 95 CI 087 to 097) Thisanalysis differs from ours which was based on 26 studies andfound a risk ratio of 095 95 CI 088 to 102

Our results for rate of falls were very similar to those of Campbell2007 (RaR 078 95 CI 068 to 089) which included six trialsthat reported a rate ratioGates 2008 included 19 trials of multifactorial interventions 17 ofwhich are in this review We excluded Gill 2002 which although acommunity-based intervention reported falls as an adverse eventand Shaw 2003 in which 79 per cent of the participants werenot community dwelling but were living in institutions providingintermediate to high level nursing care Their analysis found thatthe risk of falling was not reduced (RR 091 95 CI 082 to 10218 trials) Their finding is similar to that of this review for thisoutcome Our subgroup analysis by intensity of intervention failedto confirm the finding of Gates 2008 possibly due to differences inthe inclusion criteria and the number of completed trials availablefor inclusion in their review

Vitamin D

Two systematic reviews explored the evidence for the effect of vi-tamin D on falls Jackson 2007 included five studies in a meta-analysis of risk of falling of which three are included in this reviewand two were excluded either because they were not an RCT (Graafmans 1996) or because their participants were older peoplein institutional care (Bischoff 2003) We agree with their conclu-sion of a trend towards a reduction in the risk of falling amongpeople treated with vitamin D3 compared with placebo but thedifference is not significantRichy 2008 included 11 studies in a meta-analysis of which sixwere included in this review The other five did not meet our in-clusion criteria either because they were not RCTs (Graafmans1996) or because their participants were older people in insti-tutional care (Bischoff 2003 Broe 2007 Chapuy 2002 Flicker2005) Richy 2008 used indirect comparisons to shape their con-clusion that D-hormone analogues prevent falls to a greater extentthan their native compound We agree that this may be the caseHowever more data would be needed to confirm this hypothesisin older people living in the community and we found evidenceof an increased risk of adverse effects with these agents

A U T H O R S rsquo C O N C L U S I O N S

Implications for practice

By June 2008 we found the following evidence of effectivenessfor a number of different approaches to fall prevention in thecommunity in older people Please note that this evidence may notbe applicable to older people with dementia as a majority of theincluded studies specifically excluded them from participation

Exercise

Overall exercise is an effective intervention to reduce the risk andrate of falls Three different approaches to exercise appear to have



significant beneficial effects Multiple-component group exercisereduces rate of falls and risk of falling Tai Chi as a group exercisereduces rate of falls and risk of falling Individually prescribedexercise carried out at home reduces rate of falls and risk of fallingbut there is no evidence to support this intervention in peoplewith severe visual impairment or mobility problems after a strokeParkinsonrsquos disease or after a hip fracture


Multifactorial interventions integrating assessment with individ-ualised intervention usually involving a multi-professional teamare effective in reducing rate of falls but not risk of falling Thereis no evidence that assessment and intervention is more effectivethan assessment and referral or that multifactorial interventionsare more effective in participants selected as being at higher riskof falling

Environmental assessment and intervention

Overall home safety interventions do not appear to reduce rateof falls or risk of falling Although evidence so far published isrelatively limited people at higher risk of falling may benefit Ananti-slip shoe device for icy conditions significantly reduced winteroutside falls in one study

Medication interventions

There is limited evidence for the effectiveness of interventions tar-geting medications (eg withdrawal of psychotropics educationalprogrammes for family physicians) Overall vitamin D does notappear to be an effective intervention for preventing falls in olderpeople living in the community but there is provisional evidencethat it may reduce falls risk in people with low vitamin D levels

Prevention of falling in people with particularhealth problems

Poor vision

In people who are severely visually impaired there is evidence fromone trial for the effectiveness of a home safety intervention butnot an exercise intervention The effectiveness of other interven-tions for visual impairment in older people is uncertain althoughaccelerating first eye cataract surgery for people on a waiting listsignificantly reduces rate of falls compared with waiting list con-trols Older people may be at increased risk of falling while adjust-ing to new spectacles or major changes in prescriptionCardiovascular disorders

Evidence from a single study indicates that cardiac pacing in peoplewith carotid sinus hypersensitivity and a history of syncope andor falls reduces rate of falls

Neurological disorders

Risk of falling was not significantly reduced by home-based phys-iotherapy for people with Parkinsonrsquos disease or community phys-iotherapy for people with stroke-related mobility problems Vi-tamin D analogues were not effective in reducing rate of falls inpeople with Parkinsonrsquos disease

Implications for research

This review shows that the effect of exercise programmes in re-ducing the risk and rate of falling should now be regarded as es-tablished Further research exploring the balance of componentswithin the exercise category might be justified but would need tobe large in order to have power to discern any differences

Assessment and individualised multifactorial intervention pro-grammes appear effective in reducing the rate of falls in studiesfrom different health care systems Further research appears justi-fied to explore the difference between programmes which provideintegration of assessment and intervention by a multidisciplinaryteam and programmes which provide assessment but rely on re-ferral to other providers and agencies for the intervention

Further research appears justified to confirm the emerging evi-dence of effectiveness of home safety interventions in higher riskgroups and vitamin D in people with lower vitamin D levels andto clarify the impact of strategies to optimise care for people withdifferent visual impairments

Research on the impact of management programmes for other riskfactors such as cognitive impairment and urinary incontinence onrate and risk of falling appears justified

Studies evaluating fall prevention should be adequately poweredand use a contemporary standard for definition of a fall methodsof ascertainment recording analysis and reporting of data

Design and reporting of trials should meet the contemporary stan-dards of the CONSORT statement including those for cluster-randomised trials

Where factorial designs are employed data for each treatment cellshould be reported to allow interpretation of possible interactionsbetween different intervention components

A C K N O W L E D G E M E N T S

The authors would like to thank Lindsey Elstub and Joanne Elliottfor their support at the editorial base We would also like to thankthe following for their useful and constructive comments on earlierversions of the protocol andor review Dr Jacqueline Close DrHelen Handoll Assoc Prof Peter Herbison Prof Rajan Madhokand Dr Janet Wale In addition we would also like to thank DrGeoff Murray for his assistance with data extraction and qualityassessment We are grateful to N Freeman and Dr Aiko Osawa fortheir assistance with translations



R E F E R E N C E S

References to studies included in this review

Ashburn 2007 published data only

Ashburn A Randomised controlled trial of a home-based exerciseprogramme to reduce fall frequency among people with Parkin-sonrsquos disease (PD) Current Controlled Trials wwwcontrolled-tri-alscomISRCTN63503875 (accessed 27 March 2008)lowast Ashburn A Fazakarley L Ballinger C Pickering R McLellan LDFitton C A randomised controlled trial of a home based exercise pro-gramme to reduce the risk of falling among people with Parkinsonrsquosdisease Journal of Neurology Neurosurgery and Psychiatry 200778

(7)678ndash84 [PUBMED 17119004 ]Ashburn A Pickering RM Fazakarley L Ballinger C McLellan DLFitton C Recruitment to a clinical trial from the databases of special-ists in Parkinsonrsquos disease Parkinsonism and Related Disorders 200713(1)35ndash9 [PUBMED 16928464]

Assantachai 2002 published and unpublished data

Assantachai P personal communication June 11 2007lowast Assantachai P Chatthanawaree W Thamlikitkul V PraditsuwanR Pisalsarakij D Strategy to prevent falls in the Thai elderly acontrolled study integrated health research program for the Thaielderly Journal of the Medical Association of Thailand 200285(2)215ndash22 [PUBMED 12081122]

Ballard 2004 published data only

Ballard JE McFarland C Wallace LS Holiday DB Roberson G Theeffect of 15 weeks of exercise on balance leg strength and reduc-tion in falls in 40 women aged 65 to 89 years Journal of the Amer-ican Medical Womenrsquos Association 200459(4)255ndash61 [PUBMED16845754]

Barnett 2003 published data only

Barnett A Smith B Lord SR Williams M Baumand A Community-based group exercise improves balance and reduces falls in at-riskolder people a randomised controlled trial Age and Ageing 200332

(4)407ndash14 [PUBMED 12851185]

Bischoff-Ferrari 2006 published data only

Bischoff-Ferrari HA Orav EJ Dawson-Hughes B Additive bene-fit of higher testosterone levels and vitamin D plus calcium sup-plementation in regard to fall risk reduction among older men andwomen Osteoporosis International 200819(9)1307ndash14 [MED-LINE 18348447]lowast Bischoff-Ferrari HA Orav EJ Dawson-Hughes B Effect of chole-calciferol plus calcium on falling in ambulatory older men andwomen a 3-year randomized controlled trial Archives of Internal

Medicine 2006166(4)424ndash30 [PUBMED 16505262]Bischoff-Ferrari HA Orav EJ Dawson-Hughes B Effect of vitaminD3 plus calcium on fall risk in older men and women a 3-yearrandomized controlled trial [abstract] Journal of Bone and Mineral

Research 200419(Suppl 1)S57Dawson-Hughes B Harris SS Krall EA Dallal GE Effect of calciumand vitamin D supplementation on bone density in men and women

65 years of age or older New England Journal of Medicine 1997337

(10)670ndash6 [PUBMED 9278463]

Brown 2002 published data onlylowast Brown AI Functional adaptation to exercise in elderly subjects [thesis]httpadtcurtineduauthesesavailableadt-WCU20030423094914Perth (WA) Curtin Univ of Technology 2002 (accessed 31 March2008)Brown AP Reducing falls in elderly people a review of exerciseinterventions Physiotherapy Theory and Practice 199915(2)59ndash68[EMBASE 1999232158]Piotrowski A Cole J Allison G The influence of functional abilityand physical and social intervention on falls in elderly subjects [ab-stract] XVIth Congress of the International Association of Geron-tology 1997Aug 19-23 Adelaide Australia 581

Buchner 1997a published data onlylowast Buchner DM Cress ME de Lateur BJ Esselman PC MargheritaAJ Price R et alThe effect of strength and endurance training ongait balance fall risk and health services use in community-livingolder adults Journals of Gerontology Series A Biological Sciences andMedical Sciences 199752(4)M218ndash24 [PUBMED 9224433]Buchner DM Cress ME Wagner EH de Lateur BJ The role of exer-cise in fall prevention Developing targeting criteria for exercise pro-grams In Vellas B Toupet M Rubenstein L Albarede JL ChristenY editor(s) Falls balance and gait disorders in the elderly AmsterdamElsevier 199255ndash68Buchner DM Cress ME Wagner EH de Lateur BJ Price R AbrassIB The Seattle FICSITMoveIt study the effect of exercise on gaitand balance in older adults Journal of the American Geriatrics Society

199341321ndash5 [PUBMED 8440857]

Bunout 2005 published and unpublished data

Bunout D personal communication Feb 1 2005lowast Bunout D Barrera G Avendano M de la Maza P Gattas V Leiva Let alResults of a community-based weight-bearing resistance trainingprogramme for healthy Chilean elderly subjects Age and Ageing

200534(1)80ndash3 [PUBMED 15591487]

Campbell 1997 published and unpublished data

Campbell AJ Robertson MC Gardner MM Norton RN Buch-ner DM Falls prevention over 2 years a randomized controlledtrial in women 80 years and older Age and Ageing 199928513ndash8[PUBMED 10604501]lowast Campbell AJ Robertson MC Gardner MM Norton RN TilyardMW Buchner DM Randomised controlled trial of a general practiceprogramme of home based exercise to prevent falls in elderly womenBMJ 19973151065ndash9 [PUBMED 9366737]Gardner M Home-based exercises to prevent falls in elderly womenNew Zealand Journal of Physiotherapy 199826(3)6 [ CINAHLAN 1999044632]Gardner MM Buchner DM Robertson MC Campbell AJ Practicalimplementation of an exercise-based falls prevention programmeAge and Ageing 200130(1)77ndash83 [MEDLINE 11322678]Robertson MC Development of a falls prevention programme for el-



derly people evaluation of efficacy effectiveness and efficiency [thesis]Dunedin New Zealand Univ of Otago 2001Robertson MC Campbell AJ Gardner MM Devlin N Preventinginjuries in older people by preventing falls a meta-analysis of indi-vidual-level data Journal of the American Geriatrics Society 200250905ndash11 [PUBMED 12028179]Robertson MC Devlin N Scuffham P Gardner MM Buchner DMCampbell AJ Economic evaluation of a community based exerciseprogramme to prevent falls Journal of Epidemiology and Community

Health 200155(8)600ndash6

Campbell 1999 published and unpublished datalowast Campbell AJ Robertson MC Gardner MM Norton RN BuchnerDM Psychotropic medication withdrawal and a home-based exerciseprogram to prevent falls a randomized controlled trial Journalof the American Geriatrics Society 199947(7)850ndash3 [PUBMED10404930]Gardner MM Buchner DM Robertson MC Campbell AJ Practicalimplementation of an exercise-based falls prevention programmeAge and Ageing 200130(1)77ndash83 [MEDLINE 11322678]Robertson MC Development of a falls prevention programme for elderlypeople evaluation of efficacy effectiveness and efficiency [PhD thesis]Dunedin New Zealand Univ of Otago 2001Robertson MC Campbell AJ Gardner MM Devlin N Preventinginjuries in older people by preventing falls a meta-analysis of indi-vidual-level data Journal of the American Geriatrics Society 200250905ndash11 [PUBMED 12028179]

Campbell 2005 published data onlylowast Campbell AJ Robertson MC La Grow SJ Kerse NM SandersonGF Jacobs RJ et alRandomised controlled trial of prevention of fallsin people aged gt or =75 with severe visual impairment the VIP trialBMJ 2005331(7520)817 [PUBMED 16183652]La Grow SJ Robertson MC Campbell AJ Clarke GA Kerse NMReducing hazard related falls in people 75 years and older with signif-icant visual impairment how did a successful program work InjuryPrevention 200612(5)296ndash301 [MEDLINE 17018669]

Carpenter 1990 published data only

Carpenter GI Demopoulos GR Screening the elderly in the com-munity controlled trial of dependency surveillance using a ques-tionnaire administered by volunteers BMJ 1990300(6734)1253ndash6 [PUBMED 2354297]

Carter 1997 unpublished data only

Carter S Campbell E Sanson-Fisher R Tiller K Gillespie WJ Trialdata (as supplied 1997) Data on file

Carter 2002 published data onlylowast Carter ND Khan KM McKay HA Petit MA Waterman CHeinonen A et alCommunity-based exercise program reduces riskfactors for falls in 65- to 75-year-old women with osteoporosis Ran-domized controlled trial CMAJ Canadian Medical Association Jour-

nal 2002167(9)997ndash1004 [PUBMED 12403738 ]Carter ND Khan KM Petit MA Heinonen A Waterman C Don-aldson MG et alResults of a 10 week community based strengthand balance training programme to reduce fall risk factors a ran-domised controlled trial in 65-75 year old women with osteoporosisBritish Journal of Sports Medicine 200135(5)348ndash51 [PUBMED11579072 ]

Cerny 1998 published and unpublished data

Cerny K personal communication October 22 2002lowast Cerny K Blanks R Mohamed O Schwab D Robinson B RussoA Zizz C The effect of a multidimensional exercise program onstrength range of motion balance and gait in the well elderly [ab-stract] Gait and Posture 19987(2)185ndash6

Clemson 2004 published data only

Clemson L Stepping On reducing falls and building confidencea practical program that works [abstract] Falls prevention in olderpeople from research to practice Proceedings of the 1st Australianfalls prevention conference 2004 Nov 21-23 Sydney (AU) Rand-wick NSW Australia Prince of Wales Medical Research Institute200468lowast Clemson L Cumming RG Kendig H Swann M Heard R TaylorK The effectiveness of a community-based program for reducingthe incidence of falls in the elderly a randomized trial Journal of

the American Geriatrics Society 200452(9)1487ndash94 [PUBMED15341550 ]Clemson L Taylor K Kendig H Cumming RG Swann M Recruit-ing older participants to a randomised trial of a community-basedfall prevention program Australasian Journal on Ageing 200726(1)35ndash9 [ CINAHL AN 2009512824]Swann M Clemson L Evaluating falls efficacy following a commu-nity based falls prevention program for older people [abstract] Fallsprevention in older people from research to practice Proceedingsof the 1st Australian falls prevention conference 2004 Nov 21-23Sydney (AU) Randwick NSW Australia Prince of Wales MedicalResearch Institute 200434

Close 1999 published and unpublished data

Close J personal communication Dec 9 2008Close J Can the incidence of falls in the elderly be reduced by asecondary prevention protocol National Research Register (NRR)Archive httpsportalnihracuk (accessed 26 March 2008) [NRR Publication ID F0300115]lowast Close J Ellis M Hooper R Glucksman E Jackson S Swift CPrevention of falls in the elderly trial (PROFET) a randomised con-trolled trial Lancet 1999353(9147)93ndash7 [PUBMED 10023893]Close J Hooper R Glucksman E Jackson S Swift C Predictors offalls in a high risk population - results from the prevention of fallsin the elderly trial (PROFET) [abstract] Journal of the AmericanGeriatrics Society 200048(8)S79Close JCT Ellis M Hooper R Glucksman E Jackson SHD SwiftCG Predictors of falls - results from prevention of falls in the elderlytrial (PROFET) [abstract] Age and Ageing 199928(Suppl 1)14Close JCT Ellis M Jackson SHD Glucksman E Swift CG Inter-disciplinary assessment of elderly people presenting to AampE with afall [abstract] Age and Ageing 199827(Suppl 1)20Close JCT Patel A Hooper R Glucksman E Jackson SHD SwiftCG PROFET - Improved clinical outcomes at no additional cost[abstract] Age and Ageing 200029(Suppl 1)48

Coleman 1999 published data only

Coleman EA Grothaus LC Sandhu N Wagner EH Chronic careclinics a randomized controlled trial of a new model of primary carefor frail older adults Journal of the American Geriatrics Society 199947(7)775ndash83 [PUBMED 10404919]



Cornillon 2002 published data only

Cornillon E Blanchon MA Ramboatsisetraina P Braize C BeauchetO Dubost V et alEffectiveness of falls prevention strategies for el-derly subjects who live in the community with performance assess-ment of physical activities (before-after) [Impact drsquoun programmede prevention multidisciplinaire de la chute chez le sujet age au-tonome vivant a domicile avec analyse avantndashapres des performancesphysiques] Annales de Readaptation et de Medecine Physique 200245(9)493ndash504 [PUBMED 12495822 ]

Cumming 1999 published data only

Cumming RG Thomas M Szonyi G Frampton G Salkeld G Clem-son L Adherence to occupational therapist recommendations forhome modifications for falls prevention American Journal of Occu-

pational Therapy 200155(6)641ndash8 [PUBMED 12959228]lowast Cumming RG Thomas M Szonyi G Salkeld G OrsquoNeill E West-bury C et alHome visits by an occupational therapist for assessmentand modification of environmental hazards a randomized trial offalls prevention Journal of the American Geriatrics Society 199947

(12)1397ndash1402 [PUBMED 10591231]Salkeld G Cumming RG OrsquoNeill E Thomas M Szonyi G West-bury C The cost effectiveness of a home hazard reduction program toreduce falls among older persons Australian and New Zealand Jour-nal of Public Health 200024(3)265ndash71 [PUBMED 10937402]


Cumming RG Ivers R Clemson L Cullen J Hayes MF TanzerM et alImproving vision to prevent falls in frail older people Arandomized trial Journal of the American Geriatrics Society 200755

(2)175ndash81 [PUBMED 17302652]

Davison 2005 published data only

Aske J Can the incidence of falls in the elderly be reduced by asecondary falls prevention protocol National Research Register(NRR) Archive httpsportalnihracuk (accessed 26 April 2007)[ NRR Publication ID N0116069489]Davis M SAFER2 - Syncope and falls in the emergency room - anexplanatory randomised controlled trial of a multidisciplinary post-fall assessment and intervention strategy in elderly recurrent fallers at-tending casualty National Research Register (NRR) Archive httpsportalnihracuk (accessed 26 May 2008) [ NRR Publication IDN0009027144]lowast Davison J Bond J Dawson P Steen IN Kenny RA Patients withrecurrent falls attending Accident amp Emergency benefit from multi-factorial intervention - a randomised controlled trial Age and Ageing

200534(2)162ndash8 [PUBMED 15716246]Kenny RA A post-fall intervention strategy after presentation tocasualty National Research Register (NRR) Archive httpsportalnihracuk (accessed 26 May 2008) [ NRR Publication IDN0145036249]Kenny RA A post-fall intervention strategy after presentation tocasualty - Safer 2 National Research Register (NRR) Archive httpsportalnihracuk (accessed 26 March 2008) [ NRR PublicationID N0145049230]Kenny RA SAFER 2 - Syncope and falls in the emergency room -The Tyneside casualty falls intervention project National ResearchRegister (NRR) Archive httpsportalnihracuk (accessed 26 April2007) [ NRR Publication ID N0503055776]

Day 2002 published and unpublished data

Day L Fildes B Gordon I Fitzharris M Flamer H Lord S Ran-domised factorial trial of falls prevention among older people livingin their own homes BMJ 2002325(7356)128ndash31 [PUBMED12130606 ]

Dhesi 2004 published data only

Dhesi JK Bearne L Jackson SH Moniz C Hurley M Swift CG etalVitamin D supplementation improves the balance and functionalperformance of older people who fall [abstract] Journal of the Amer-ican Geriatrics Society 200250(4 Suppl)S5lowast Dhesi JK Jackson SH Bearne LM Moniz C Hurley MV SwiftCG et alVitamin D supplementation improves neuromuscular func-tion in older people who fall Age and Ageing 200433(6)589ndash95[PUBMED 15501836]Swift C A controlled intervention study of vitamin D supplemen-tation on neuromuscular and psychomotor function in elderly peo-ple who fall National Research Register (NRR) Archive httpsportalnihracuk (accessed 26 March 2008) [ NRR PublicationID N0116016083]

Dukas 2004 published data onlylowast Dukas L Bischoff HA Lindpaintner LS Schacht E Birkner-BinderD Damm TN et alAlfacalcidol reduces the number of fallers in acommunity-dwelling elderly population with a minimum calciumintake of more than 500 mg daily Journal of the American GeriatricsSociety 200452(2)230ndash6 [PUBMED 14728632]Dukas L Schacht E Mazor Z Stahelin HB Treatment with alfacal-cidol in elderly people significantly decreases the high risk of falls as-sociated with a low creatinine clearance of lt65 mlmin OsteoporosisInternational 200516(2)198ndash203 [MEDLINE 15221207]Dukas LC Schacht E Mazor Z Stahelin HB A new significant andindependent risk factor for falls in elderly men and women a lowcreatinine clearance of less than 65 mlmin Osteoporosis International200516(3)332ndash8 [MEDLINE 15241585]

Elley 2008 published data only

Falls Assessment Clinical Trial randomised controlled trial of amulti-component intervention in primary health care to reduce fallsamongst over 75 year old adults with a history of falling AustralianNew Zealand Clinical Trials Registry httpwwwanzctrorgau(accessed 31 March 2008) [ ACTRN12605000054617]lowast Elley CR Robertson MC Garrett S Kerse NM MacKinlay ELawton B et alEffectiveness of a falls-and-fracture nurse coordina-tor to reduce falls a randomized controlled trial of at-risk olderadults Journal of the American Geriatrics Society 200856(8)1383ndash9[MEDLINE 18808597]Elley CR Robertson MC Kerse NM Garrett S McKinlay E LawtonB et alFalls Assessment Clinical Trial (FACT) design interventionsrecruitment strategies and participant characteristics BMC PublicHealth 20077185 [MEDLINE 17662156]

Fabacher 1994 published data only

Fabacher D Josephson K Pietruszka F Linderborn K Morley JERubenstein LZ An in-home preventive assessment program for in-dependent older adults a randomized controlled trial Journalof the American Geriatrics Society 199442(6)630ndash8 [PUBMED8201149]

Fiatarone 1997 published data only

Fiatarone MA OrsquoNeill EF Doyle RN Clements K Efficacy of home-based resistance training in frail elders (Abstract 985) Abstracts of



the 16th Congress of the International Association of GerontologyBedford Park South Australia World Congress of Gerontology Inc1997323 [CENTRAL CNndash00405155]

Foss 2006 published data onlylowast Foss AJ Harwood RH Osborn F Gregson RM Zaman A MasudT Falls and health status in elderly women following second eyecataract surgery a randomised controlled trial Age and Ageing 200635(1)66ndash71 [PUBMED 16364936 ]Foss AJE Randomised controlled trial of second eye cataract extrac-tion to prevent falls in elderly women National Research Register(NRR) Archive httpsportalnihracuk (accessed 26 March 2008)[ NRR Publication ID N0192080923]

Gallagher 1996 published data only

Gallagher EM Brunt H Head over heels impact of a health pro-motion program to reduce falls in the elderly Canadian Journal on

Aging 199615(1)84ndash96 [ EMBASE 1996164172]


Gallagher JC The effects of calcitriol on falls and fractures and phys-ical performance tests Journal of Steroid Biochemistry and Molecular

Biology 200489-90(1-5)497ndash501 [MEDLINE 15225827]Gallagher JC Fowler S Effect of estrogen calcitriol and a combina-tion of estrogen and calcitriol on bone mineral density and fracturesin elderly women [abstract] Journal of Bone and Mineral Research

199914(Suppl 1)S209lowast Gallagher JC Fowler SE Detter JR Sherman SS Combinationtreatment with estrogen and calcitriol in the prevention of age-relatedbone loss Journal of Clinical Endocrinology and Metabolism 200186

(8)3618ndash28 [PUBMED 11502787]Gallagher JC Haynatski G Fowler S Calcitriol therapy reduces fallsand fractures in elderly women [abstract] Calcified Tissue Interna-tional 200372334Gallagher JC Haynatzki G Fowler S Effect of estrogen calcitriolor the combination of both on falls and non vertebral fractures inelderly women [abstract] Journal of Bone and Mineral Research 200217(Suppl 1)S210Gallagher JC Rapuri P Smith L Falls are associated with decreasedrenal function and insufficient calcitriol production by the kidneyJournal of Steroid Biochemistry and Molecular Biology 2007103(3-5)610ndash3 [MEDLINE 17236758]Gallagher JC Rapuri PB Haynatzki G Detter JR Effect of discon-tinuation of estrogen calcitriol and the combination of both onbone density and bone markers Journal of Clinical Endocrinologyand Metabolism 200287(11)4914ndash23 [MEDLINE 12414850]Gallagher JC Rapuri PB Smith LM An age-related decrease in cre-atinine clearance is associated with an increase in number of falls inuntreated women but not in women receiving calcitriol treatmentJournal of Clinical Endocrinology and Metabolism 200792(1)51ndash8[MEDLINE 17032712]

Grant 2005 published and unpublished data

Andrew JG Randomised placebo-controlled trial of daily oral vita-min D and calcium for the secondary prevention of osteoporosis re-lated fractures in the elderly (RECORD) National Research Register(NRR) Archive httpsportalnihracuk (accessed 26 March 2008)[ Publication ID N0217084004]Armstrong A MREC 9707 The MRC RECORD Study Ran-domised placebo-controlled trial of daily oral vitamin D and calciumfor the secondary prevention of osteoporosis related fractures in the

elderly In National Research Register Oxford Update Software2003 issue 2Chikanza I Vitamin D and Calcium for secondary prevention ofosteoporosis related fractures in the elderly In National ResearchRegister Oxford Update Software 2003 issue 2 [ PublicationID N0147078505]Chuck A The MRC Record study - Randomised trial vitamin D andcalcium for the secondary prevention of osteoporosis related fracturesin the elderly In National Research Register Oxford UpdateSoftware 2003 issue 2 [ Publication ID N0521092364]Francis RM Randomised trial of Vitamin D and calcium for thesecondary prevention of osteoporosis related fractures in the elderlyIn National Research Register Oxford Update Software 2003issue 2Francis RM Grant AM RECORD Trial Group The RECORDtrial a randomised double-blind study of calcium andor vitamin Din the secondary prevention of low trauma fractures [abstract] Age

and Ageing 200534(Suppl 2)ii16Gillespie WJ Randomised trial of Vitamin D and Calcium for thesecondary prevention of osteoporosis related fractures in the elderlyRECORD STUDY In National Research Register Oxford UpdateSoftware 2003 issue 2 [ Publication ID N0519058601]Grant AM Randomised trial of vitamin D and calcium for the sec-ondary prevention of osteoporosis related fractures in the elderly(MRC RECORD study) In National Research Register OxfordUpdate Software 2003 issue 2 [ Publication ID N0411050637]lowast Grant AM Avenell A Campbell MK McDonald AM MacLennanGS McPherson GC et alOral vitamin D3 and calcium for secondaryprevention of low-trauma fractures in elderly people (RandomisedEvaluation of Calcium Or vitamin D RECORD) a randomisedplacebo-controlled trial Lancet 2005 Vol 365 issue 94711621ndash8[MEDLINE 15885294]Howell F Randomised placebo-controlled trial of daily oral vitaminD and calcium for the secondary prevention of osteoporosis relatedfractures in the elderly In National Research Register OxfordUpdate Software 2003 issue 2Poulton S MRC RECORD TRIAL Randomised placebo controlledtrial of daily oral vitamin D and calcium for the secondary preventionof osteoporosis related fractures in the elderly In National ResearchRegister Oxford Update Software 2003 issue 2 [ PublicationID N0187062340]Rowley DI Multicentre randomised trial of vitamin D and calciumfor the secondary prevention of osteoporosis related fractures in theelderly In National Research Register Oxford Update Software2003 issue 2 [ Publication ID N0405042439]Summers GD A randomised trial of vitamin D and calcium for thesecondary prevention of osteoporosis related fractures in the elderlyIn National Research Register Oxford Update Software 2003issue 2 [ Publication ID N0077049118]Wallace WA Randomised trial of vitamin D and calcium for thesecondary prevention of osteoporosis related fractures in the elderly(the RECORD study) ISRCTN 51647438 In National ResearchRegister Oxford Update Software 2003 issue 2 [ PublicationID N0192080910]

Gray-Donald 1995 published data only

Gray-Donald K Payette H Boutier V Randomized clinical trial ofnutritional supplementation shows little effect on functional status



among free-living frail elderly Journal of Nutrition 1995125(12)2965ndash71 [PUBMED 7500174]

Green 2002 published data only

Green J A randomised trial of community physiotherapy one yearpost stroke National Research Register (NRR) Archive httpsportalnihracuk (accessed 1 April 2008) [ NRR Publication IDN0049004427]lowast Green J Forster A Bogle S Young J Physiotherapy for patientswith mobility problems more than 1 year after stroke a randomisedcontrolled trial Lancet 2002359(9302)199ndash203 [PUBMED11812553]

Greenspan 2005 published data only

Greenspan SL Resnick NM Parker RA Combination therapy withhormone replacement and alendronate for prevention of bone lossin elderly women a randomized controlled trial JAMA 2003289

(19)2525ndash33 [MEDLINE 12759324]lowast Greenspan SL Resnick NM Parker RA The effect of hormonereplacement on physical performance in community-dwelling el-derly women American Journal of Medicine 2005118(11)1232ndash9[PUBMED 16271907]

Harwood 2004 published data only

The Nottingham Neck of Femur Study the optimal role ofvitamin D and calcium in elderly patients with established os-teoporosis National Research Register (NRR) Archive httpsportalnihracukPagesNRRArchiveaspx (accessed 02 December2008) [ NRR Publication ID N0192080773]lowast Harwood RH Sahota O Gaynor K Masud T Hosking DJ Arandomised controlled comparison of different calcium and vitaminD supplementation regimens in elderly women after hip fractureThe Nottingham Neck of Femur (NoNOF) study Age and Ageing

200433(1)45ndash51 [MEDLINE 14695863]


Foss AJE Randomised trial to assess the efficacy of expedited cataractextraction in the prevention of falls in elderly people awaitingcataract surgery National Research Register (NRR) Archive httpsportalnihracuk (accessed 27 March 2008) [ NRR PublicationID 192080923]Harwood R Does expedited cataract extraction reduce therisk of falls in elderly people - a randomised controlledtrial National Research Register (NRR) Archive httpsportalnihracukPagesNRRArchiveaspx (accessed 26 March2008)Harwood RH Foss A Osborn F Gregson R Zaman A Masud TFalls and health status in elderly women following first eye cataractsurgery a randomised controlled trial [abstract] Age and Ageing200534(Suppl 1)i21lowast Harwood RH Foss AJ Osborn F Gregson RM Zaman A MasudT Falls and health status in elderly women following first eye cataractsurgery a randomised controlled trial British Journal of Ophthal-mology 200589(1)53ndash9 [PUBMED 15615747]Sach TH Foss AJ Gregson RM Zaman A Osborn F Masud T etalFalls and health status in elderly women following first eye cataractsurgery an economic evaluation conducted alongside a randomisedcontrolled trial British Journal of Ophthalmology 200791(12)1675ndash9 [MEDLINE 17585002]

Hauer 2001 published data only

Hauer K Pfisterer M Schuler M Bartsch P Oster P Two yearslater A prospective long-term follow-up of a training interventionin geriatric patients with a history of severe falls Archives of PhysicalMedicine and Rehabilitation 200384(10)1426ndash32 [MEDLINE14586908]lowast Hauer K Rost B Rutschle K Opitz H Specht N Bartsch P etalExercise training for rehabilitation and secondary prevention offalls in geriatric patients with a history of injurious falls Journal

of the American Geriatrics Society 200149(1)10ndash20 [PUBMED11207837]Hauer K Specht N Schuler M Bartsch P Oster P Intensive physicaltraining in geriatric patients after severe falls and hip surgery Age

and Ageing 200231(1)49ndash57 [MEDLINE 11850308]Oster P Hauer K Specht N Rost B Baertsch P Schlierf G Strengthand coordination training for prevention of falls in the elderly [Kraftndashund Koordinationstraining zur Sturzpraumlvention im Alter] Zeitschrift

fur Gerontologie und Geriatrie 199730(4)289ndash92 [MEDLINE9410508]

Helbostad 2004 published data only

Helbostad JL Moe-Nilssen R Sletvold O Comparison of two typesof exercise regimes on selected functional abilities for community-dwelling elderly at risk of falling [abstract] XVI Conference of theInternational Society for Postural Gait Research 2003 March 23-27 Sydney (Australia) httpwwwpowmriunsweduauispg2003(accessed 240703)lowast Helbostad JL Sletvold O Moe-Nilssen R Effects of home ex-ercises and group training on functional abilities in home-dwellingolder persons with mobility and balance problems A randomizedstudy Aging - Clinical and Experimental Research 200416(2)113ndash21 [PUBMED 15195985]Helbostad JL Sletvold O Moe-Nilssen R Home training with andwithout additional group training in physically frail old people livingat home effect on health-related quality of life and ambulationClinical Rehabilitation 2004 Vol 18 issue 5498ndash508 [PUBMED15293484]

Hendriks 2008 published data only

Hendriks M Preventing further falls and functional decline amongelderly persons presented to the Accident and Emergency (AampE)department with a fall randomised controlled trial Current Con-trolled Trials httpcontrolled-trialscom (accessed 31 March 2008)Hendriks MR Bleijlevens MH Van Haastregt JC Crebolder HFDiederiks JP Evers SM et alLack of effectiveness of a multidisci-plinary fall-prevention program in elderly people at risk a random-ized controlled trial Journal of the American Geriatrics Society 200856(8)1390-7 [MEDLINE 18662214]Hendriks MR Bleijlevens MH Van Haastregt JC De Bruijn FHDiederiks JP Mulder WJ et alA multidisciplinary fall preventionprogram for elderly persons a feasibility study Geriatric Nursing200829(3)186ndash96 [MEDLINE 18555160]lowast Hendriks MR Evers SM Bleijlevens MH Van Haastregt JC Cre-bolder HF Van Eijk JT Cost-effectiveness of a multidisciplinary fallprevention program in community-dwelling elderly people A ran-domized controlled trial (ISRCTN 64716113) International Jour-

nal of Technology Assessment in Health Care 200824(2)193ndash202[MEDLINE 18400123]Hendriks MR Van Haastregt JC Diederiks JP Evers SM Crebolder



HF Van Eijk JT Effectiveness and cost-effectiveness of a multidisci-plinary intervention programme to prevent new falls and functionaldecline among elderly persons at risk design of a replicated ran-domised controlled trial [ISRCTN64716113] BMC Public Health200556 [MEDLINE 15651990]

Hill 2000 published data only

Crome P personal communication August 29 2006Crome P Hill S Mossman J Stockdale P A randomised controlledtrial of a nurse led falls prevention clinic [abstract] Journal of the

American Geriatrics Society 200048(8)S78lowast Hill S Mossman J Stockdale P Crome P A randomised controlledtrial of a nurse-led falls prevention clinic [abstract] Age amp Ageing200029(Suppl 2)20

Hogan 2001 published data only

Hogan DB MacDonald FA Betts J Bricker S Ebly EM DelarueB et alA randomized controlled trial of a community-based consul-tation service to prevent falls CMAJ Canadian Medical AssociationJournal 2001165(5)537ndash43 [PUBMED 11563205]

Hornbrook 1994 published data only

Hornbrook MC Stevens VJ Wingfield DJ Seniorsrsquo program for in-jury control and education Journal of the American Geriatrics Society

199341(3)309ndash14 [MEDLINE 8440855]lowast Hornbrook MC Stevens VJ Wingfield DJ Hollis JF GreenlickMR Ory MG Preventing falls among community-dwelling olderpersons results from a randomized trial Gerontologist 199434(1)16ndash23 [PUBMED 8150304]Stevens VJ Hornbrook MC Wingfield DJ Hollis JF Greenlick MROry MG Design and implementation of a falls prevention interven-tion for community-dwelling older persons Behavior Health and

Aging 1991922(1)57ndash73

Huang 2004 published data only

Huang TT Acton GJ Effectiveness of home visit falls preventionstrategy for Taiwanese community-dwelling elders randomized trialPublic Health Nursing 200421(3)247ndash56 [PUBMED 15144369]


Huang TT Liang SH A randomized clinical trial of the effectivenessof a discharge planning intervention in hospitalized elders with hipfracture due to falling Journal of Clinical Nursing 200514(10)1193ndash201 [PUBMED 16238765]

Jitapunkul 1998 published data only

Jitapunkul S A randomised controlled trial of regular surveillancein Thai elderly using a simple questionnaire administered by non-professional personnel Journal of the Medical Association of Thailand

199881(5)352ndash6 [PUBMED 9623035]

Kenny 2001 published data only

Kenny RA Richardson DA Carotid sinus syndrome and falls inolder adults American Journal of Geriatric Cardiology 200110(2)97ndash9 [PUBMED 11253467]lowast Kenny RA Richardson DA Steen N Bexton RS Shaw FE BondJ Carotid sinus syndrome a modifiable risk factor for nonaccidentalfalls in older adults (SAFE PACE) Journal of the American College ofCardiology 200138(5)1491ndash6 [PUBMED 11691528]Kenny RA Seifer CM SAFE PACE - Syncope and falls in the el-derly - pacing and carotid sinus evaluation a randomised controlledtrial of cardiac pacing in older patients with falls and carotid sinus

hypersensitivity American Journal of Geriatric Cardiology 19998(2)87ndash90Richardson DA Steen N Bond J Bexton R Kenny RA Cardiacpacing reduces falls in carotid sinus hypersensitivity [abstract] Ageand Ageing 200029(Suppl 1)46

Kingston 2001 published data only

Kingston P Elderly people and accidents a prospective analysis ofaccidental causation among elderly populations and their post dis-charge requirements National Research Register (NRR) Archivehttpsportalnihracuk (accessed 1 April 2008) [ NRR Publica-tion ID N0498009612]Kingston P Jones M Crome P A RCT of health visitor (HV) inter-vention in falls [abstract] Age and Ageing 200130(Suppl 1)40lowast Kingston P Jones M Lally F Crome P Older people and fallsA randomized controlled trial of a health visitor (HV) interven-tion Reviews in Clinical Gerontology 200111(3)209ndash14 [EM-BASE 2002061828]Kingston PA Older people and rsquofallsrsquo a randomised control trial of healthvisitor intervention [thesis] Stoke-on-Trent Keele University 1998

Korpelainen 2006 published data only

Korpelainen R Keinanen-Kiukaanniemi S Heikkinen J VaananenK Korpelainen J Effect of impact exercise on bone mineral densityin elderly women with low BMD a population-based randomizedcontrolled 30-month intervention Osteoporosis International 200617(1)109ndash18 [PUBMED 15889312]

Lannin 2007 published data only

Lannin NA Clemson L McCluskey A Lin CW Cameron ID Bar-ras S Feasibility and results of a randomised pilot-study of pre-dis-charge occupational therapy home visits BMC Health Services Re-search 2007742 [PUBMED 17355644]

Latham 2003 published data only

Latham NK Anderson CS Lee A Bennett DA Moseley A CameronID A randomized controlled trial of quadriceps resistance exerciseand vitamin D in frail older people The Frailty Interventions Trialin Elderly Subjects (FITNESS) Journal of the American GeriatricsSociety 200351291ndash9 [PUBMED 12588571]

Li 2005 published data only

Li F Harmer P Fisher KJ McAuley E Tai Chi improving functionalbalance and predicting subsequent falls in older persons Medicineand Science in Sports and Exercise 200436(12)2046ndash52 [MED-LINE 15570138]lowast Li F Harmer P Fisher KJ McAuley E Chaumeton N Eckstrom Eet alTai Chi and fall reductions in older adults a randomized con-trolled trial The Journals of Gerontology Series A Biological Sciences

and Medical Sciences 200560(2)187ndash94 [PUBMED 5814861]

Lightbody 2002 published data only

Leathley M Fallers attending casualty National Research Register(NRR) Archive httpsportalnihracuk (accessed 1 April 2008) [NRR Publication ID N0500000414]lowast Lightbody E Watkins C Leathley M Sharma A Lye M Evalu-ation of a nurse-led falls prevention programme versus usual carea randomized controlled trial Age and Ageing 200231(3)203ndash10[PUBMED 12006310]

Lin 2007 published and unpublished data

Lin MR Wolf SL Hwang HF Gong SY Chen CY A randomizedcontrolled trial of fall prevention programs and quality of life in older



fallers Journal of the American Geriatrics Society 200755(4)499ndash506 [PUBMED 17397426]

Liu-Ambrose 2004 published data only

Liu-Ambrose T Khan KM Eng JJ Lord SR McKay HA Balanceconfidence improves with resistance or agility training Increase isnot correlated with objective changes in fall risk and physical abilitiesGerontology 200450(6)373ndash82 [MEDLINE 15477698]Liu-Ambrose T Khan KM Eng JJ Lord SR McKay HA Strengthor agility training significantly reduces fall risk compared to posturetraining in 75 to 85 year old women with low bone density a sixmonth RCT [abstract] XVI th conference of the International So-ciety for Postural and Gait Research 2003 March 23-27Sydney(Australia) httpwwwpowmriunsweduauispg2003 (accessed 24August 2003)Liu-Ambrose TY Khan KM Eng JJ Gillies GL Lord SR McKayHA The beneficial effects of group-based exercises on fall risk profileand physical activity persist 1 year postintervention in older womenwith low bone mass follow-up after withdrawal of exercise Journal ofthe American Geriatrics Society 200553(10)1767ndash73 [PUBMED16181178]lowast Lui-Ambrose T Khan KM Eng JJ Janssen PA Lord SR McKayHA Resistance and agility training reduce fall risk in women aged75 to 85 with low bone mass a 6-month randomized controlledtrial Journal of the American Geriatrics Society 200452(5)657ndash65[PUBMED 15086643]

Lord 1995 published data onlylowast Lord SR Ward JA Williams P Strudwick M The effect of a 12-month exercise trial on balance strength and falls in older women arandomized controlled trial Journal of the American Geriatrics Society1995431198ndash206 [PUBMED 7594152]Lord SR Ward JA Williams P Zivanovic E The effects of a com-munity exercise program on fracture risk factors in older womenOsteoporosis International 19966(5)361ndash7 [PUBMED 8931030]

Lord 2003 published data only

Lord SR Castell S Corcoran J Dayhew J Matters B Shan A etalThe effect of group exercise on physical functioning and falls in frailolder people living in retirement villages a randomized controlledtrial Journal of the American Geriatrics Society 200351(12)1685ndash92 [MEDLINE 14687345]


Lord SR Tiedemann A Chapman K Munro B Murray SM Geron-tology M et alThe effect of an individualized fall prevention pro-gram on fall risk and falls in older people a randomized controlledtrial Journal of the American Geriatrics Society 200553(8)1296ndash304 [PUBMED 16078954]

Luukinen 2007 published data onlylowast Luukinen H Lehtola S Jokelainen J Vaananen-Sainio R LotvonenS Koistinen P Pragmatic exercise-oriented prevention of falls amongthe elderly A population-based randomized controlled trial Pre-ventive Medicine 200744(3)265ndash71 [PUBMED 17174387]Luukinen H Lehtola S Jokelainen J Vaananen-Sainio R Lotvo-nen S Koistinen P Prevention of disability by exercise among theelderly a population-based randomized controlled trial Scandina-vian Journal of Primary Health Care 200624(4)199ndash205 [MED-LINE 17118858]

Mahoney 2007 published data only

Mahoney JE Shea TA Przybelski R Jaros L Gangnon R Cech S etalKenosha County falls prevention study a randomized controlledtrial of an intermediate-intensity community-based multifactorialfalls intervention Journal of the American Geriatrics Society 200755

(4)489ndash98 [PUBMED 17397425]

McKiernan 2005 published data only

McKiernan FE A simple gait-stabilizing device reduces outdoor fallsand nonserious injurious falls in fall-prone older people during thewinter Journal of the American Geriatrics Society 200553(6)943ndash7[PUBMED 15935015]

McMurdo 1997 published data only

McMurdo ME Mole PA Paterson CR Controlled trial of weightbearing exercise in older women in relation to bone density and fallsBMJ 1997314(7080)596 [PUBMED 9055716]

Means 2005 published data only

Means KM Rodell DE OrsquoSullivan PS Balance mobility and fallsamong community-dwelling elderly persons effects of a rehabilita-tion exercise program American Journal of Physical Medicine andRehabilitation 200584(4)238ndash50 [PUBMED 15785256]

Meredith 2002 published data only

Meredith S Feldman P Frey D Giammarco L Hall K Arnold Ket alImproving medication use in newly admitted home healthcarepatients a randomized controlled trial Journal of the American Geri-atrics Society 200250(9)1484ndash91 [PUBMED 12383144]

Morgan 2004 published data only

DeVito CA Morgan RO Safe-Grip fallinjuries intervention a ran-domized controlled trial httpclinicaltrialsgov (accessed 1 April2008)DeVito CA Morgan RO Duque M Abdel-Moty E Virnig BAPhysical performance effects of low-intensity exercise among clin-ically defined high-risk elders Gerontology 200349(3)146ndash54[PUBMED 12679604]lowast Morgan RO Virnig BA Duque M Abdel-Moty E DeVito CALow-intensity exercise and reduction of the risk for falls among at-risk elders Journals of Gerontology Series A Biological Sciences andMedical Sciences 200459(10)1062ndash7 [PUBMED 15528779]

Newbury 2001 published data only

Newbury J Marley J Preventive home visits to elderly people in thecommunity Visits are most useful for people aged gt75 [letter] BMJ2000321(7529)512lowast Newbury JW Marley JE Beilby J A randomised controlled trialof the outcome of health assessment of people aged 75 years andover Medical Journal of Australia 2001175(2)104ndash7 [PUBMED11556409]

Nikolaus 2003 published data onlylowast Nikolaus T Bach M Preventing falls in community-dwelling frailolder people using a home intervention team (HIT) Results fromthe randomized falls-HIT trial Journal of the American GeriatricsSociety 200351(3)300ndash5 [PUBMED 12588572]Nikolaus T Specht-Leible N Bach M Wittmann-Jennewein C Os-ter P Schlierf G Effectiveness of hospital-based geriatric evaluationand management and home intervention team (GEM-HIT) Ratio-nale and design of a 5-year randomized trial Zeitschrift fur Geron-

tologie und Geriatrie 199528(1)47ndash53 [MEDLINE 7773832]



Nitz 2004 published and unpublished data

Nitz JC personal communication May 6 2005lowast Nitz JC Choy NL The efficacy of a specific balance-strategytraining programme for preventing falls among older people a pi-lot randomised controlled trial Age and Ageing 200433(1)52ndash8[PUBMED 14695864]

Pardessus 2002 published data only

Pardessus V Puisieux F Di P Gaudefroy C Thevenon A DewaillyP Benefits of home visits for falls and autonomy in the elderly Arandomized trial study American Journal of Physical Medicine and

Rehabilitation 200281(4)247ndash52 [PUBMED 11953541]

Pereira 1998 published data only

Kriska AM Bayles C Cauley JA LaPorte RE Sandler RB PambiancoG A randomized exercise trial in older women increased activityover two years and the factors associated with compliance Medicineand Science in Sports and Exercise 198618(5)557ndash62Pereira MA Ten year follow-up of a randomized exercise trial in post-menopausal women [PhD thesis] Pittsburgh (PA) Univ of Pitts-burgh 1996 [ Proquest Digital Dissertations Publication NumberAAT 97 16627]lowast Pereira MA Kriska AM Day RD Cauley JA LaPorte RE KullerLH A randomized walking trial in postmenopausal women effectson physical activity and health 10 years later Archives of InternalMedicine 1998158(15)1695ndash701 [PUBMED 9701104]

Pfeifer 2000 published data onlylowast Pfeifer M Begerow B Minne HW Abrams C Nachtigall DHansen C Effects of a short-term vitamin D and calcium supplemen-tation on body sway and secondary hyperparathyroidism in elderlywomen Journal of Bone and Mineral Research 200015(6)1113ndash8[PUBMED 10841179]Pfeifer M Begerow B Nachtigall D Hansen C Prevention of falls-related fractures vitamin D reduces body sway in the elderly - aprospective randomized double blind study [abstract] Bone 199823(5 Suppl 1)1110

Pit 2007 published data only

Pit SW Byles JE Henry DA Holt L Hansen V Bowman DA AQuality Use of Medicines program for general practitioners and olderpeople a cluster randomised controlled trial Medical Journal ofAustralia 2007187(1)23ndash30 [PUBMED 17605699]

Porthouse 2005 published and unpublished data

Baverstock M A randomised controlled trial of calcium and vitaminD supplementation for fracture and falls prevention In NationalResearch Register Oxford Update Software 2006 Issue 3Baverstock M A randomised-controlled trial of nurse led clinics forcalcium and vitamin D supplementation to prevent fractures InNational Research Register Oxford Update Software 2006 Issue3Cochayne S personal communication August 16 2005lowast Porthouse J Cochayne S King C Saxon L Steele E Aspray Tet alRandomised controlled trial of calcium and supplementationwith cholecalciferol (vitamin D3) for prevention of fractures in pri-mary care BMJ 2005 Vol 330 issue 74981003 [PUBMED15860827]Puffer S Calcium and vitamin D in primary care Compliance re-sults from a randomised controlled trial [abstract] Osteoporosis In-

ternational 200314(Suppl 4)S8

Prince 2008 published data only

Prince R Effects of vitamin D and calcium on bone and fallsin an elderly population of Australian women selected for theirhistory of falling Australian New Zealand Clinical Trials Reg-istry httpwwwanzctrorgau (accessed 31 March 2008) [ AC-TRN12606000331538]lowast Prince RL Austin N Devine A Dick IM Bruce D Zhu K Ef-fects of ergocalciferol added to calcium on the risk of falls in elderlyhigh-risk women Archives of Internal Medicine 2008168(1)103ndash8[PUBMED 18195202]

Reinsch 1992 published data only

El-Faizy M Reinsch S Home safety intervention for the preventionof falls Physical amp Occupational Therapy in Geriatrics 199412(3)33ndash49 [ EMBASE 1994365778]MacRae PG Feltner ME Reinsch S A 1-year exercise program forolder women effects on falls injuries and physical performanceJournal of Aging and Physical Activity 19942127ndash42lowast Reinsch S MacRae P Lachenbruch PA Tobis JS Attempts to pre-vent falls and injury a prospective community study Gerontologist

199232450ndash6 [PUBMED 1427246]Tobis J Reinsch S McRae P Lachenbruch T Experimental interven-tion at senior centres for the prevention of falls [abstract] Journal ofthe American Geriatrics Society 199038(8)A28

Resnick 2002 published data only

Resnick B Testing the effect of the WALC intervention on exerciseadherence in older adults Journal of Gerontological Nursing 200228

(6)40ndash9 [PUBMED 12071273]

Robertson 2001a published and unpublished data

Gardner MM Buchner DM Robertson MC Campbell AJ Practicalimplementation of an exercise-based falls prevention programmeAge and Ageing 200130(1)77ndash83 [MEDLINE 11322678]Robertson MC Development of a falls prevention programme for el-

derly people evaluation of efficacy effectiveness and efficiency [thesis]Dunedin New Zealand Univ of Otago 2001Robertson MC Campbell AJ Gardner MM Devlin N Preventinginjuries in older people by preventing falls a meta-analysis of indi-vidual-level data Journal of the American Geriatrics Society 200250

(5)905ndash11lowast Robertson MC Devlin N Gardner MM Campbell AJ Effective-ness and economic evaluation of a nurse delivered home exercise pro-gramme to prevent falls 1 Randomised controlled trial BMJ 2001322(7288)697ndash701 [PUBMED 11264206]

Robson 2003 published data only

Robson E Edwards J Gallagher E Baker D Steady as you go(SAYGO) A falls-prevention program for seniors living in the com-munity Canadian Journal on Aging 200322(2)207ndash16 [EMBASE2003344777]

Rubenstein 2000 published data only

Rubenstein LZ Josephson KR Trueblood PR Loy S Harker JOPietruszka FM et alEffects of a group exercise program on strengthmobility and falls among fall-prone elderly men Journals of Geron-tology Series A Biological Sciences and Medical Sciences 200055(6)M317ndash21 [PUBMED 10843351]



Rubenstein 2007 published and unpublished data

Alessi C personal communication June 10 2007Josephson K personal communication November 20 2007lowast Rubenstein LZ Alessi CA Josephson KR Trinidad Hoyl M HarkerJO Pietruszka FM A randomized trial of a screening case findingand referral system for older veterans in primary care Journal ofthe American Geriatrics Society 200755(2)166ndash74 [MEDLINE17302651]

Ryan 1996 published data only

Ryan JW Spellbring AM Implementing strategies to decrease risk offalls in older women Journal of Gerontological Nursing 199622(12)25ndash31 [PUBMED 9060344]

Salminen 2008 unpublished data only

Kivela S-L Aarnio P Asikainen E Hyttinen H Isoaho R Karra E etalPrevention of injurious falls and fractures in ageing and aged pop-ulation [abstract] ProFaNE (Prevention of Falls Network Europe)meeting 2004 June 11-13 Manchester (UK)lowast Salminen MJ Vahlberg TJ Salonoja MT Aarnio PT Kivelauml S-LFalls data (as supplied 20 May 2008) Data on fileSalonoja M Kivelauml S-L Prevention of falls and injurious falls amongelderly people wwwclinicaltrialsgov (accessed 26 March 2008)Sjosten NM Salonoja M Piirtola M Vahlberg T Isoaho R HyttinenH et alA multifactorial fall prevention programme in home-dwellingelderly people A randomized-controlled trial Public Health 2007121(4)308ndash18 [MEDLINE 17320125]Sjosten NM Salonoja M Piirtola M Vahlberg TJ Isoaho R Hyt-tinen HK et alA multifactorial fall prevention programme in thecommunity-dwelling aged predictors of adherence European Jour-

nal of Public Health 200717(5)464ndash70 [MEDLINE 17208952]Sjosten NM Vahlberg TJ Kivela S-L The effects of multifactorialfall prevention on depressive symptoms among the aged at increasedrisk of falling International Journal of Geriatric Psychiatry 200823

(5)504ndash10 [EMBASE 2008251008]Vaapio S Salminen M Vahlberg T Sjosten N Isoaho R Aarnio Pet alEffects of risk-based multifactorial fall prevention on health-related quality of life among the community-dwelling aged a ran-domized controlled trial Health amp Quality of Life Outcomes 2007520 [MEDLINE 17462083]

Sato 1999 published data only

Sato Y Manabe S Kuno H Oizumi K Amelioration of osteope-nia and hypovitaminosis D by 1alpha-hydroxyvitamin D3 in elderlypatients with Parkinsonrsquos disease Journal of Neurology Neurosurgery

and Psychiatry 199966(1)64ndash8

Schrijnemaekers 1995 published data only

Schrijnemaekers VJ Haveman MJ Effects of preventive outpatientgeriatric assessment short-term results of a randomized controlledstudy Home Health Care Services Quarterly 199515(2)81ndash97[MEDLINE 10143898]

Sherrington 2004 published and unpublished data

Sherrington C Personal communication October 30 2004Sherrington C The effects of exercise on physical ability following fall-related hip fracture [thesis] Sydney (Australia) Univ of New SouthWales 2001Sherrington C Lord SR Herbert RD A randomised controlled trialof weight-bearing versus non-weight-bearing exercise for improvingphysical ability after hip fracture and completion of usual care [ab-stract] XVI th conference of the International Society for Postu-

ral and Gait Research 2003 March 23-27Sydney (Australia) httpwwwpowmriunsweduauispg2003 (accessed 240703)Sherrington C Lord SR Herbert RD A randomised trial of weight-bearing versus non-weight-bearing exercise for improving physicalability in inpatients after hip fracture Australian Journal of Physio-

therapy 200349(1)15ndash22 [MEDLINE 12600250]lowast Sherrington C Lord SR Herbert RD A randomized controlledtrial of weight-bearing versus non-weight-bearing exercise for im-proving physical ability after usual care for hip fracture Archives

of Physical Medicine and Rehabilitation 200485(5)710ndash6 [MED-LINE 15129393]

Shigematsu 2008 published data onlylowast Shigematsu R Okura T Nakagaichi M Tanaka K Sakai T Ki-tazumi S et alSquare-stepping exercise and fall risk factors in olderadults a single-blind randomized controlled trial Journals of Geron-

tology Series A-Biological Sciences amp Medical Sciences 200863(1)76ndash82 [MEDLINE 18245764]Shigematsu R Okura T Sakai T Rantanen T Square-stepping exer-cise versus strength and balance training for fall risk factors Aging-

Clinical amp Experimental Research 200820(1)19ndash24 [MEDLINE18283224]

Shumway-Cook 2007 published data only

Shumway-Cook A Silver I Mary L York S Cummings P Koepsell TThe effectiveness of a community-based multifactorial interventionon falls and fall risk factors in community living older adults arandomized controlled trial CSM 2007 [abstract] Journal ofGeriatric Physical Therapy 200629(3)117lowast Shumway-Cook A Silver IF LeMier M York S Cummings PKoepsell TD Effectiveness of a community-based multifactorial in-tervention on falls and fall risk factors in community-living olderadults a randomized controlled trial Journals of Gerontology Se-ries A Biological Sciences and Medical Sciences 2007 Vol 62 issue121420ndash7 [PUBMED 18166695]

Skelton 2005 published data only

Skelton D personal communication February 1 2005lowast Skelton D Dinan S Campbell M Rutherford O Tailored groupexercise (Falls Management Exercise -- FaME) reduces falls in com-munity-dwelling older frequent fallers (an RCT) Age and Ageing200534(6)636ndash9 [EMBASE 2005539610]Skelton DA Dinan SM Exercise for falls management Rationalefor an exercise programme aimed at reducing postural instabilityPhysiotherapy Theory and Practice 199915(2)105ndash20 [EMBASE1999232161]Skelton DA Dinan SM Campbell M Rutherford OM FaME(Falls Management Exercise) An RCT on the effects of a 9-monthgroup exercise programme in frequently falling community dwellingwomen aged 65 and over [abstract] Journal of Aging and Physical

Activity 200412(3)457ndash8Skelton DA Stranzinger K Dinan S Rutherford OM BMD im-provements following FaME (Falls Management Exercise) in fre-quently falling women age 65 and over an RCT 7th WorldCongress on Aging and Physical Activity [abstract] Journal of Agingand Physical Activity 200816 SupplS89ndash90

Smith 2007 published data only

Anderson FH Smith HE Raphael HM Cooper C Intramuscularvitamin D increased serum 125-dihydroxycholecalciferol but didnot affect 25-hydroxy-cholecalciferol levels in healthy older adults



[abstract] Journal of Bone and Mineral Research 200015(Suppl 1)S315Anderson FH Smith HE Raphael HM Crozier SR Cooper C Ef-fect of annual intramuscular vitamin D3 supplementation on frac-ture risk in 9440 community-living older people the Wessex frac-ture prevention trial [abstract] Journal of Bone and Mineral Research200419(Suppl 1)S57Arden NK Crozier S Smith H Anderson F Edwards C Raphael Het alKnee pain knee osteoarthritis and the risk of fracture Arthritis

and Rheumatism 200655(4)610ndash5 [MEDLINE 16874784]Ellis B Wessex fracture prevention study In National Re-search Register Oxford Update Software 2006 Issue 3wwwnrrnhsukViewDocumentaspID=N0187062321 (accessed24 August 2006) [ NRR Publication ID N0187062321]Raphael H Smith H Anderson F Cooper C Tackling the problemsof trial management in primary care - experience from the Wessexresearch network fracture prevention study of annual vitamin D in-jection in older people [abstract] Osteoporosis International 200011

(Suppl 1)S63ndash4Smith H Primary prevention of fractures in the elderly eval-uating the effectiveness of annual vitamin D supplementationlinked with primary care in influenza immunisation In Na-tional Research Register Oxford Update Software 2006 Is-sue 3 wwwnrrnhsukViewDocumentaspID=N0108081272(accessed 24 August 2006) [ NRR Publication ID N0108081272]Smith H Anderson F Raphael H Cooper C The Wessex researchnetwork fracture prevention study - a large pragmatic trial of annualvitamin D injection in older people [abstract] Osteoporosis Interna-tional 200011(Suppl 1)S64Smith H Anderson F Raphael H Crozier S Cooper C Effect of an-nual intramuscular vitamin D supplementation on fracture risk pop-ulation-based randomised double-blind placebo-controlled trial[abstract] Osteoporosis International 200415(Suppl 1)S8lowast Smith H Anderson F Raphael H Maslin P Crozier S CooperC Effect of annual intramuscular vitamin D on fracture risk in el-derly men and women - a population-based randomised double-blind placebo-controlled trial Rheumatology 200746(12)1852ndash7[MEDLINE 17998225]

Speechley 2008 published and unpublished data

Gill DP Zou GY Jones GR Speechley M Injurious falls are associ-ated with lower household but higher recreational physical activitiesin community-dwelling older male veterans Gerontology 200854

(2)106ndash15 [MEDLINE 18259094]lowast Speechley M Falls data (as supplied 03 June 2008) Data on file

Spice 2009 published and unpublished data

Gordon C The Winchester Falls Project A randomisedcontrolled trial of multidisciplinary assessment in the sec-ondary prevention of falls National Research Register (NRR)Archive httpsportalnihracukProfilesNRRaspxPublication_ID=N0278078805 (accessed 26 March 2008) [ NRR PublicationID N0278078805]Gordon CJ Spice C The Winchester Falls Project A Cluster Ran-domised Community Intervention Trial of Secondary Prevention ofFalls in Community-Dwelling Older People ClinicalTrialsgov httpclinicaltrialsgovshowNCT00130624 (accessed 26 March 2008)

[ ClinicalTrialsgov Identifier NCT00130624]Spice C personal communication December 24 2006Spice C Morotti W Dent T George S Rose J Gordon C TheWinchester Falls Project A randomised controlled trial of secondaryfalls prevention [abstract] Age amp Ageing 200534(Suppl 2)ii18lowast Spice C Morotti W George S Dent T Rose J Harris S et alTheWinchester falls project a randomised controlled trial of secondaryprevention of falls in older people Age and Ageing 2009 Vol 38issue 133ndash40 [PUBMED 18829689]

Steadman 2003 published and unpublished data

Kalra L personal communication March 27 2006Kalra L Can an enhanced balance training programme improve mo-bility amp reduce falls in elderly patients presenting to Health ServicesIn National Research Register Oxford Update Software 2003 is-sue 2lowast Steadman J Donaldson N Kalra L A randomized controlled trialof an enhanced balance training program to improve mobility andreduce falls in elderly patients Journal of the American GeriatricsSociety 200351(6)847ndash52 [MEDLINE 12757574]

Steinberg 2000 published and unpublished data

Peel N personal communication October 10 2007Peel N Cartwright C Steinberg M Monitoring slips trips and falls inthe older community preliminary results Health Promotion Journalof Australia 19988(2)148ndash50Peel N Steinberg M Williams G Home safety assessment in theprevention of falls among older people Australian and New Zealand

Journal of Public Health 200024(5)536ndash9 [PUBMED 11109693]lowast Steinberg M Cartwright C Peel N Williams G A sustainableprogramme to prevent falls and near falls in community dwellingolder people results of a randomised trial Journal of Epidemiology

and Community Health 200054(3)227ndash32

Stevens 2001 published data only

Stevens M Holman CD Bennett N Preventing falls in older peopleImpact of an intervention to reduce environmental hazards in thehome Journal of the American Geriatrics Society 200149(11)1442ndash7 [PUBMED 11890581]lowast Stevens M Holman CD Bennett N De Klerk N Preventing fallsin older people Outcome evaluation of a randomized controlledtrial Journal of the American Geriatrics Society 200149(11)1448ndash55 [PUBMED 11890582]

Suzuki 2004 published data only

Suzuki T Kim H Yoshida H Ishizaki T Randomized controlledtrial of exercise intervention for the prevention of falls in commu-nity-dwelling elderly Japanese women Journal of Bone and MineralMetabolism 200422(6)602ndash11 [MEDLINE 15490272]

Swanenburg 2007 published data only

Swanenburg J De Bruin ED Stauffacher M Mulder T Uebelhart DEffects of exercise and nutrition on postural balance and risk of fallingin elderly people with decreased bone mineral density randomizedcontrolled trial pilot study Clinical Rehabilitation 200721(6)523ndash34 [MEDLINE 17613583]

Tinetti 1994 published data only

King MB Tinetti ME A multifactorial approach to reducing inju-rious falls Clinics in Geriatric Medicine 199612(4)745ndash59Koch M Gottschalk M Baker DI Palumbo S Tinetti ME An im-pairment and disability assessment and treatment protocol for com-



munity-living elderly persons Physical Therapy 199474286-94discussion 295-8Rizzo JA Baker DI McAvay G Tinetti ME The cost-effectivenessof a multifactorial targeted prevention program for falls among com-munity elderly persons Medical Care 199634(9)954ndash69Tinetti ME Prevention of falls and fall injuries in elderly persons aresearch agenda Preventive Medicine 199423756ndash62Tinetti ME Baker DI Garrett PA Gottschalk M Koch ML HorwitzRI Yale FICSIT risk factor abatement strategy for fall preventionJournal of the American Geriatrics Society 199341315ndash20lowast Tinetti ME Baker DI McAvay G Claus EB Garrett P GottschalkM et alA multifactorial intervention to reduce the risk of fallingamong elderly people living in the community New England Journal

of Medicine 1994331(13)821ndash7Tinetti ME McAvay G Claus E Does multiple risk factor reductionexplain the reduction in fall rate in the Yale FICSIT Trial Frailty andInjuries Cooperative Studies of Intervention Techniques American

Journal of Epidemiology 1996144(4)389ndash99

Trivedi 2003 published data only

Trivedi DP Doll R Tee Khaw K Effect of four monthly oral vita-min D3 (cholecalciferol) supplementation on fractures and mortalityin men and women living in the community randomised doubleblind controlled trial BMJ 2003326(7387)469ndash72 [MEDLINE12609940]

Van Haastregt 2000 published data onlylowast Van Haastregt JC Diederiks JP Van Rossum E De Witte LPVoorhoeve PM Crebolder HF Effects of a programme of multifac-torial home visits on falls and mobility impairments in elderly peopleat risk randomised controlled trial BMJ 2000321(7267)994ndash8[PUBMED 11039967]Van Haastregt JC Van Rossum E Diederiks JP De Witte LP Voorho-eve PM Crebolder HF Process-evaluation of a home visit programmeto prevent falls and mobility impairments among elderly people atrisk Patient Education and Counseling 200247(4)301ndash9 [MED-LINE 12135821]Van Haastregt JC Van Rossum E Diederiks JP Voorhoeve PMDe Witte LP Crebolder HF Preventing falls and mobility prob-lems in community-dwelling elders the process of creating a newintervention Geriatric Nursing 200021(6)309ndash14 [MEDLINE11135129]

Van Rossum 1993 published data only

Van Rossum E Frederiks CM Philipsen H Portengen K WiskerkeJ Knipschild P Effects of preventive home visits to elderly peopleBMJ 1993307(6895)27ndash32 [PUBMED 8343668]

Vellas 1991 published data only

Vellas B Albarede JL A randomized clinical trial on the valueof raubasine-dihydroergocristine (Iskedyl(TM)) in the preven-tion of post fall syndrome [Effet de lrsquoassociation raubasinendashdihydroergocristine (Iskedyl(TM)) sur le syndrome postndashchute et surla prevention de la chute chez le sujet age] Psychologie Medicale 199123(7)831ndash9 [ EMBASE 1991275391]

Vetter 1992 published data only

Vetter NJ Lewis PA Ford D Can health visitors prevent fracturesin elderly people BMJ 1992304(6831)888ndash90 [PUBMED1392755]

Voukelatos 2007 published and unpublished data

Haas M Economic analysis of tai chi as a means of prevent-ing falls and related injuries among older adults CHEREworking paper 20064 Sydney Australia Centre forHealth Economics Research and Evaluation University ofTechnology httpdatasearchutseduauchereresearchworking_paperscfm (accessed 27 March 2008)Rissel C VoukelatosA Cumming B Lord S Central Sydney Tai Chi trial AustralianResource Centre for Health Care Innovations wwwarchinetaue-libraryhealth_administrationbaxter05effectiveness_of_health_carecentral_sydney (accessed 17 August 2006)Voukelatos A Central Sydney Tai Chi trial personal communicationJuly 25 2003lowast Voukelatos A Cumming RG Lord SR Rissel C A randomizedcontrolled trial of tai chi for the prevention of falls the CentralSydney Tai Chi trial Journal of the American Geriatrics Society 200755(8)1185ndash91 [PUBMED 17661956]Voukelatos A Metcalfe A Central Sydney Tai Chi Trial methodol-ogy New South Wales Public Health Bulletin 200213(1-2)19Voukelatos A Rissel C Cumming R Lord S The Central Sydney Tai

Chi Trial a randomised controlled trial of the effectiveness of tai chi inreducing risk of falls in older people Sydney NSW Department ofHealth 2006 (wwwhealthnswgovau)

Wagner 1994 published data only

Wagner EH LaCroix AZ Grothaus L Leveille SG Hecht JA ArtzK et alPreventing disability and falls in older adults a population-based randomized trial American Journal of Public Health 199484

(11)1800ndash6 [PUBMED 7977921]

Weerdesteyn 2006 published and unpublished data

Weerdesteyn V personal communication September 06 2006lowast Weerdesteyn V Rijken H Geurts AC Smits-Engelsman BC Mul-der T Duysens J A five-week exercise program can reduce falls andimprove obstacle avoidance in the elderly Gerontology 200652(3)131ndash41 [MEDLINE 16645293]

Whitehead 2003 published data only

Whitehead C Wundke R Crotty M Finucane P Evidence-basedclinical practice in falls prevention a randomised controlled trial ofa falls prevention service Australian Health Review 200326(3)88ndash96 [MEDLINE 15368824]

Wilder 2001 published data only

Wilder P Seniors to seniors exercise program a cost effective way toprevent falls in the frail elderly living at home [abstract] Journal ofGeriatric Physical Therapy 200124(3)13

Wolf 1996 published data only

Kutner NG Barnhart H Wolf SL McNeely E Xu T Self-reportbenefits of Tai Chi practice by older adults Journals of GerontologySeries B Psychological Sciences and Social Sciences 199752(5)242ndash6[MEDLINE 9310093]McNeely E Clements SD Wolf SL A program to reduce frailty inthe elderly In Funk SG Tornquist EM Champagne MT WeiseRA editor(s) Key aspects of elder care managing falls incontinence

and cognitive impairment New York Springer 199289ndash96OrsquoGrady M Wolf SL Barnhart HX Kutner N McNeely E TaiChi effect on falls in frail older adults [abstract] Archives of Physi-



cal Medicine and Rehabilitation 1997781028 [CENTRAL CNndash00507025]Wolf SL Barnhart HX Ellison GL Coogler CE Horak FB Theeffect of Tai Chi Quan and computerized balance training on posturalstability in older subjects Physical Therapy 199777(4)371ndash84lowast Wolf SL Barnhart HX Kutner NG McNeely E Coogler C XuT Reducing frailty and falls in older persons an investigation ofTai Chi and computerized balance training Journal of the AmericanGeriatrics Society 199644489ndash97Wolf SL Kutner NG Green RC McNeely E The Atlanta FICSITstudy two exercise interventions to reduce frailty in elders Journal

of the American Geriatrics Society 199341(3)329ndash32


Greenspan AI Wolf SL Kelley ME OrsquoGrady M Tai chi and per-ceived health status in older adults who are transitionally frail arandomized controlled trial Physical Therapy 200787(5)525ndash35[MEDLINE 17405808]Sattin RW Easley KA Wolf SL Chen Y Kutner MH Reductionin fear of falling through intense tai chi exercise training in oldertransitionally frail adults Journal of the American Geriatrics Society

200553(7)1168ndash78 [MEDLINE 16108935]Wolf SL OrsquoGrady M Easley KA Guo Y Kressig RW Kutner M Theinfluence of intense Tai Chi training on physical performance andhemodynamic outcomes in transitionally frail older adults Journals

of Gerontology Series A Biological Sciences and Medical Sciences 200661(2)184ndash9 [MEDLINE 16510864]lowast Wolf SL Sattin RW Kutner M OrsquoGrady M Greenspan AI GregorRJ Intense Tai Chi exercise training and fall occurrences in oldertransitionally frail adults a randomized controlled trial Journal ofthe American Geriatrics Society 2003 Vol 51 issue 121693ndash701[MEDLINE 14687346]Wolf SL Sattin RW OrsquoGrady M Freret N Ricci L Greenspan AIet alA study design to investigate the effect of intense Tai Chi inreducing falls among older adults transitioning to frailty Controlled

Clinical Trials 200122(6)689ndash704 [MEDLINE 11738125]

Woo 2007 published and unpublished data

Woo J Hong A Lau E Lynn H A randomised controlled trial ofTai Chi and resistance exercise on bone health muscle strength andbalance in community-living elderly people Age and Ageing 200736(3)262ndash8 [MEDLINE 17356003]

Wyman 2005 published data only

Findorff MJ Stock HH Gross CR Wyman JF Does the Transthe-oretical Model (TTM) explain exercise behavior in a community-based sample of older women Journal of Aging amp Health 200719

(6)985ndash1003 [MEDLINE 18165292]Findorff MJ Wyman JF Nyman JA Croghan CF Measuring thedirect healthcare costs of a fall injury event Nursing Research 200756(4)283ndash7 [MEDLINE 17625468]Lindquist R Wyman JF Talley KM Findorff M Gross CR Design ofcontrol-group conditions in clinical trials of behavioral interventionsJournal of Nursing Scholarship 200739(3)214ndash21 [MEDLINE17760793]Nachreiner NM Findorff MJ Wyman JF McCarthy TC Cir-cumstances and consequences of falls in community-dwelling olderwomen Journal of Womenrsquos Health 200716(10)1437ndash46 [MED-LINE 18062759]Wyman J A home-

based fall prevention intervention for high risk older women httpwwwdhsstatemnusmaingroupsagingdocumentspubdhs16_137823pdf (accessed 141007)Wyman J DiFabio R Gross C Konstan JA LindquistR McCarthy T et alDesign of the Fall Evaluation andPrevention Program (FEPP) a randomized trial of exerciseand risk reduction education in high-risk older women [ab-stract] ICADI International conference on agingdisabilityand independence 2003 Dec 4-6 Washington (DC) httpwwwicadiphhpufledu2003presentationphpPresID=151(accessed 14 October 2007)lowast Wyman J Gross C DiFabio R Nyman J Lindquist R McCarthyT et alA randomized trial of exercise education and risk reduc-tion counseling to prevent falls in population-based sample of olderwomen [abstract] Gerontologist 200545(Special Issue II)297Wyman J Gross C DiFabio R Nyman J Lindquist R McCarthy Tet alEfficacy of exercise education and tailored counseling in reduc-ing falls at 1- and 2-years in older women [abstract] Gerontologist200646(Special Issue 1)141Wyman JF Croghan CF Nachreiner NM Gross CR Stock HHTalley K et alEffectiveness of education and individualized coun-seling in reducing environmental hazards in the homes of commu-nity-dwelling older women Journal of the American Geriatrics Society

200755(10)1548ndash56 [MEDLINE 17908058]

References to studies excluded from this review

Alexander 2003 published data only

Alexander N personal communication August 23 2006lowast Alexander NB Bentur N Strasburg D Nyquist LV Fall risk reduc-tion in Israeli day care center attendees using exercise and behaviorstrategies [abstract] Journal of the American Geriatrics Society 200351(Suppl 4)S117

Alp 2007 published data only

Alp A Kanat E Yurtkuran M Efficacy of a self-management programfor osteoporotic subjects American Journal of Physical Medicine and

Rehabilitation 200786(8)633ndash40 [MEDLINE 17667193]

Armstrong 1996 published data only

Armstrong AL Hormone replacement therapy - effects on strength bal-ance and bone density [thesis] Nottingham Univ of Nottingham1996Armstrong AL Coupland CAC Pye DW Wallace WA A study ofthe effects of hormone replacement therapy (HRT) on bone densitystrength and balance in post-menopausal women [abstract] Journal

of Bone and Joint Surgery British Volume 199476 Suppl 142lowast Armstrong AL Oborne J Coupland CAC Macpherson MB BasseyEJ Wallace WA Effects of hormone replacement therapy on muscleperformance and balance in post-menopausal women Clinical Sci-

ence 199691(6)685ndash90 [MEDLINE 8976803]

Barr 2005 published data only

Barr RJ Stewart A Torgerson DJ Seymour DG Reid DM Screen-ing elderly women for risk of future fractures - participation rates andimpact on incidence of falls and fractures Calcified Tissue Interna-tional 200576(4)243ndash8 [MEDLINE 15812582]

Bogaerts 2007 published data only

Bogaerts A Verschueren S Delecluse C Claessens AL Boonen SEffects of whole body vibration training on postural control in older



individuals A 1 year randomized controlled trial Gait and Posture

200726(2)309ndash16 [MEDLINE 17074485]

Buchner 1997b published data only

Buchner DM Cress ME de Lateur BJ Esselman PC Margherita AJPrice R et alA comparison of the effects of three types of endurancetraining on balance and other fall risk factors in older adults Aging-

Clinical and Experimental Research 19979(1-2)112ndash9 [PUBMED9177594]

Byles 2004 published data onlylowast Byles JE Tavener M OrsquoConnell RL Nair BR Higginbotham NHJackson CL et alRandomised controlled trial of health assessmentsfor older Australian veterans and war widows Medical Journal of

Australia 2004181(4)186ndash90 [MEDLINE 15310251]Mackenzie L Byles J DrsquoEste C Validation of self-reported fall eventsin intervention studies Clinical Rehabilitation 200620(4)331ndash9[MEDLINE 16719031]Mackenzie L Byles J Higginbotham N A prospective community-based study of falls among older people in Australia frequency cir-cumstances and consequences Occupational Therapy Journal of Re-search 200222(4)143ndash52 [EMBASE 2003110930]

Chapuy 2002 published data only

Chapuy MC Pamphile R Paris E Kempf C Schlichting M ArnaudS et alCombined calcium and vitamin D3 supplementation in el-derly women confirmation of reversal of secondary hyperparathy-roidism and hip fracture risk the Decalyos II study Osteoporosis

International 200213(3)257ndash64 [MEDLINE 11991447]

Cheng 2001 published data only

Cheng P-T Wu S-H Liaw M-Y Wong AM Tang F-T Symmetricalbody-weight distribution training in stroke patients and its effect onfall prevention Archives of Physical Medicine and Rehabilitation 2001821650ndash4

Crotty 2002 published data only

Crotty M Kittel A Hayball N Home rehabilitation for older adultswith fractured hips how many will take part Journal of Quality inClinical Practice 200020(2-3)65ndash8Crotty M Whitehead C Gray S Finucane P Hayball N Rehabilita-tion in the home (RITHOM) for patients with fractured neck of fe-mur preliminary results [abstract] Internal Medicine Journal 200232 SupplA38lowast Crotty M Whitehead CH Gray S Finucane PM Early dischargeand home rehabilitation after hip fracture achieves functional im-provements a randomised controlled trial Clinical Rehabilitation200216(4)406ndash13

De Deyn 2005 published data only

De Deyn P Jeste DV Swanink R Kostic D Breder C Carson WHet alAripiprazole for the treatment of psychosis in patients withAlzheimerrsquos disease a randomized placebo-controlled study Jour-nal of Clinical Psychopharmacology 200525(5)463ndash7 [MEDLINE16160622]

Ebrahim 1997 published data only

Ebrahim S Thompson PW Baskaran V Evans K Randomizedplacebo-controlled trial of brisk walking in the prevention of post-menopausal osteoporosis Age and Ageing 199726(4)253ndash60[MEDLINE 9271287]

Elley 2003 published data onlylowast Elley CR Kerse N Arroll B Robinson E Effectiveness of coun-selling patients on physical activity in general practice cluster ran-domised controlled trial BMJ 2003326(7393)793ndash6 [MED-LINE 12689976]Elley CR Kerse NM Arroll B Why target sedentary adults in pri-mary health care Baseline results from the Waikato Heart Healthand Activity Study Preventive Medicine 200337(4)342ndash8 [MED-LINE 14507491]Kerse N Elley CR Robinson E Arroll B Is physical activity coun-seling effective for older people A cluster randomized controlledtrial in primary care Journal of the American Geriatrics Society 200553(11)1951ndash6 [MEDLINE 16274377]

Faber 2006 published and unpublished data

Faber M personal communication Aug 30 2006lowast Faber MJ Bosscher RJ Chin A Paw MJ Van Wieringen PC Effectsof exercise programs on falls and mobility in frail and pre-frail olderadults A multicenter randomized controlled trial Archives of Phys-

ical Medicine and Rehabilitation 200687(7)885ndash96 [MEDLINE16813773]

Freiberger 2007 published and unpublished data

Freiberger E Menz HB Characteristics of falls in physically activecommunity-dwelling older people Findings from the rsquoStandfest imAlterrsquo study Zeitschrift fur Gerontologie und Geriatrie 200639(4)261ndash7 [PUBMED 16900444 ]lowast Freiberger E Menz HB Abu-Omar K Rutten A Preventing fallsin physically active community-dwelling older people a comparisonof two intervention techniques Gerontology 200753(5)298ndash305[PUBMED 17536207]Frieberger E personal communication December 12 2007

Gill 2002 published data onlylowast Gill TM Baker DI Gottschalk M Peduzzi PN Allore H Byers AA program to prevent functional decline in physically frail elderlypersons who live at home New England Journal of Medicine 2002347(14)1068ndash74 [MEDLINE 12362007]Gill TM McGloin JM Gahbauer EA Shepard DM Bianco LMTwo recruitment strategies for a clinical trial of physically frail com-munity-living older persons Journal of the American Geriatrics Soci-

ety 200149(8)1039ndash45 [MEDLINE 11555064]

Graafmans 1996 published data onlylowast Graafmans WC Ooms ME Hofstee HMA Bezemer PD BouterLM Lips P Falls in the elderly a prospective study of risk factorsand risk profiles American Journal of Epidemiology 1996143(11)1129ndash36 [MEDLINE 8633602]Lips P Graafmans WC Ooms ME Bezemer PD Bouter LM Vi-tamin D supplementation and fracture incidence in elderly per-sons Annals of Internal Medicine 1996124(4)400ndash6 [MEDLINE8554248]

Hirsch 2003 published data only

Hirsch MA Toole T Maitland CG Rider RA The effects of bal-ance training and high-intensity resistance training on persons withidiopathic Parkinsonrsquos disease Archives of Physical Medicine and Re-

habilitation 200384(8)1109ndash17 [MEDLINE 12917847]

Hu 1994 published data only

Hu MH Woollacott MH Multisensory training of standing balancein older adults I Postural stability and one-leg stance balance Jour-



nals of Gerontology Series A Biological Sciences and Medical Sciences

199449M52ndash61Hu MH Woollacott MH Multisensory training of standing bal-ance in older adults II Kinematic and electromyographic posturalresponses Journals of Gerontology Series A Biological Sciences and

Medical Sciences 199449M62ndash71

Inokuchi 2007 published data only

Inokuchi S Matsusaka N Hayashi T Shindo H Feasibility and ef-fectiveness of a nurse-led community exercise programme for pre-vention of falls among frail elderly people a multi-centre controlledtrial Journal of Rehabilitation Medicine 200739(6)479ndash85 [MED-LINE 17624483]

Iwamoto 2005 published data only

Iwamoto J Takeda T Sato Y Uzawa M Effect of whole-body vi-bration exercise on lumbar bone mineral density bone turnover andchronic back pain in post-menopausal osteoporotic women treatedwith alendronate Aging-Clinical amp Experimental Research 200517

(2)157ndash63 [MEDLINE 15977465]

Kempton 2000 published data only

Hahn A van Beurden E Kempton A Sladden T Garner E Meetingthe challenge of falls prevention at the population level a commu-nity-based intervention with older people in Australia Health Promo-

tion International 199611(3)203ndash11 [ EMBASE 1996287598]lowast Kempton A van Beurden E Sladden T Garner E Beard J Olderpeople can stay on their feet Final results of a community-based fallsprevention programme Health Promotion International 200015(1)27ndash33 [ EMBASE 2000091472]van Beurden E Kempton A Sladden T Garner E Designing an eval-uation for a multiple-strategy community intervention the NorthCoast Stay on Your Feet program Australian and New Zealand Jour-

nal of Public Health 199822(1)115ndash9

Kerschan-Schindl 2000 published data only

Kerschan-Schindl K Uher E Kainberger F Kaider A Ghanem AHPreisinger E Long-term home exercise program Effect in women athigh risk of fracture Archives of Physical Medicine and Rehabilitation

200081(3)319ndash23

Larsen 2005 published data only

Larsen ER Mosekilde L Foldspang A Determinants of acceptanceof a community-based program for the prevention of falls and frac-tures among the elderly Preventive Medicine 200133(2 Pt 1)115ndash9[MEDLINE 11493044]Larsen ER Mosekilde L Foldspang A Vitamin D and calcium sup-plementation prevents osteoporotic fractures in elderly communitydwelling residents a pragmatic population-based 3-year interven-tion study Journal of Bone and Mineral Research 200419(3)370ndash8[MEDLINE 15040824]lowast Larsen ER Mosekilde L Foldspang A Vitamin D and cal-cium supplementation prevents severe falls in elderly community-dwelling women A pragmatic population-based 3-year interventionstudy Aging-Clinical and Experimental Research 200517(2)125ndash32[MEDLINE 15977461]Larsen ER Mosekilde L Foldspang A Vitamin D and calcium treat-ment and environmental adjustment in the prevention of falls andosteoporotic fractures among elderly Danish community residents[abstract] Journal of Bone and Mineral Research 200217(Suppl 1)S157

Lee 2007 published data only

Lee JS Hurley MJ Carew D Fisher R Kiss A Drummond N Arandomized clinical trial to assess the impact on an emergency re-sponse system on anxiety and health care use among older emergencypatients after a fall Academic Emergency Medicine 200714(4)301ndash8 [MEDLINE 17331915]

Lehtola 2000 published data only

Lehtola S Hanninen L Paatalo M The incidence of falls during a six-month exercise trial and four-month followup among home dwellingpersons aged 70-75 years [Kaatumistapaturmien ilmaantuvuus 70ndash75ndashvuotiailla oululaisilla liikuntaintervention ja sen jaumllkeisen seuran-nan aikana] Liikuntatiede 2000641ndash6

Lin 2006 published data only

Lin MR Hwang H Wang Y Chang S Wolf SL Community-basedtai chi and its effect on injurious falls balance gait and fear of fallingin older people Physical Therapy 200686(9)1189ndash201 [MED-LINE 16959668]

Linnebur 2007 published and unpublished data

Linnebur S personal communication Sept 29 2007lowast Linnebur SA Vondracek SF Griend JP Ruscin JM McDermottMT Prevalence of vitamin D insufficiency in elderly ambulatory out-patients in Denver Colorado American Journal of Geriatric Pharma-

cotherapy 20075(1)1ndash8 [MEDLINE 17608242]

Mansfield 2007 published data only

Mansfield A Peters AL Liu BA Maki BE A perturbation-basedbalance training program for older adults study protocol for a ran-domised controlled trial BMC Geriatrics 2007712 [MEDLINE17540020]

Marigold 2005 published data only

Marigold DS Eng JJ Dawson AS Inglis JT Harris JE GylfadottirS Exercise leads to faster postural reflexes improved balance andmobility and fewer falls in older persons with chronic stroke Journalof the American Geriatrics Society 200553(3)416ndash23

Mead 2007 published data only

Mead GE Greig CA Cunningham I Lewis SJ Dinan S SaundersDH et alStroke a randomized trial of exercise or relaxation Journalof the American Geriatrics Society 200755892ndash9


Means KM Rodell DE OrsquoSullivan PS Cranford LA Rehabilitationof elderly fallers pilot study of a low to moderate intensity exerciseprogram Archives of Physical Medicine and Rehabilitation 1996771030ndash6

Ondo 2006 published data only

Ondo WG Almaguer M Cohen H Computerized posturographybalance assessment of patients with bilateral ventralis intermediusnuclei deep brain stimulation Movement Disorders 200621(12)2243ndash7

Peterson 2004 published and unpublished data

Allegrante JP personal communication November 26 2003Allegrante JP Improving functional recovery after hip fracture Clin-ical-



Trialsgov httpclinicaltrialsgovctshowNCT00000436order=1(accessed 010906)Allegrante JP Self-efficacy and strength training to improve postop-erative rehabilitation of hip fracture patients ClinicalTrialsgov httpclinicaltrialsgov (accessed 210401)lowast Peterson MGE Ganz SB Allegrante JP Cornell CN High-inten-sity exercise training following hip fracture Topics in Geriatric Reha-

bilitation 200420(4)273ndash84Ruchlin HS Elkin EB Allegrante JP The economic impact of amultifactorial intervention to improve postoperative rehabilitation ofhip fracture patients Arthritis amp Rheumatism 200145(5)446ndash52

Poulstrup 2000 published data only

Poulstrup A Jeune B Prevention of fall injuries requiring hospitaltreatment among community-dwelling elderly European Journal of

Public Health 200010(1)45ndash50

Protas 2005 published data only

Protas EJ Mitchell K Williams A Qureshy H Caroline K Lai ECGait and step training to reduce falls in Parkinsonrsquos disease Neurore-habilitation 200520(3)183ndash90 [PUBMED 16340099]


Resnick B personal communication October 14 2007Resnick B Testing the exercise plus program following hip fracture(PowerPoint presen-tation) httpww1odnihgovbehaviorchangeprojectsmaryland(accessed 25 August 2006)Resnick B Magaziner J Orwig D Yu-Yahiro J Hawkes W ShardellM et alTesting the effectiveness of the exercise plus program in olderwomen post-hip fracture Annals of Behavioral Medicine 200734(1)67ndash76lowast Resnick B Magaziner J Orwig D Zimmerman S Evaluating thecomponents of the Exercise Plus Program rationale theory andimplementation Health Education Research 200217(2)648ndash58Resnick B Orwig D Wehren L Zimmerman S Simpson M Maga-ziner J The Exercise Plus Program for older women post hip fractureparticipant perspectives Gerontologist 200545(4)539ndash44

Robertson 2001b published data only

Gardner MM Buchner DM Robertson MC Campbell AJ Practicalimplementation of an exercise-based falls prevention programmeAge and Ageing 200130(1)77ndash83Gardner MM Robertson MC McGee R Campbell AJ Applicationof a falls prevention program for older people to primary health carepractice Preventive Medicine 200234546ndash53lowast Robertson MC Gardner MM Devlin N McGee R CampbellAJ Effectiveness and economic evaluation of a nurse delivered homeexercise programme to prevent falls 2 Controlled trial in multiplecentres BMJ 2001322(7288)701ndash4

Rosie 2007 published data only

Rosie J Taylor D Sit-to-stand as home exercise for mobility-limitedadults over 80 years of age - GrandStand System may keep you stand-ing Age amp Ageing 200736(5)555ndash62 [MEDLINE 17646216]

Rucker 2006 published data only

Rucker D Rowe BH Johnson JA Steiner IP Russell AS HanleyDA et alEducational intervention to reduce falls and fear of fallingin patients after fragility fracture Results of a controlled pilot studyPreventive Medicine 200642(4)316ndash9 [MEDLINE 16488469]

Sakamoto 2006 published data only

Sakamoto K Nakamura T Hagino H Endo N Mori S Muto Yet alEffects of unipedal standing balance exercise on the preventionof falls and hip fracture among clinically defined high-risk elderlyindividuals A randomized controlled trial Journal of Orthopaedic

Science 200611(5)467ndash72 [MEDLINE 17013734]


Sato Y Honda Y Kaji M Asoh T Hosokawa K Kondo I etalAmelioration of osteoporosis by menatetrenone in elderly femaleParkinsonrsquos disease patients with vitamin D deficiency Bone 200231(1)114-8 Erratum in Bone 200843(1)217 [MEDLINE12110423]

Sato 2005a published data only

Sato Y Kanoko T Satoh K Iwamoto J The prevention of hip fracturewith risedronate and ergocalciferol plus calcium supplementation inelderly women with Alzheimer disease a randomized controlled trial[see comment] Archives of Internal Medicine 2005165(15)1737ndash42 [MEDLINE 16087821]


Sato Y Iwamoto J Kanoko T Satoh K Alendronate and vitamin D2for prevention of hip fracture in Parkinsonrsquos disease A randomizedcontrolled trial Movement Disorders 200621(7)924ndash9 [MED-LINE 16538619]

Schwab 1999 published and unpublished data

Klotz U personal communication March 29 2005Roder F Schwab M Aleker T Morike K Thon KP Klotz U Proximalfemur fracture in older patients - rehabilitation and clinical outcomeAge amp Ageing 200332(1)74ndash80 [MEDLINE 12540352]Schwab M Roder F Aleker T Ammon S Thon KP Eichelbaum Met alPsychotropic drug use falls and hip fracture in the elderly Aging-

Clinical and Experimental Research 200012(3)234ndash9 [MEDLINE10965382]lowast Schwab M Roder F Morike K Thon K Klotz U Prevention offalls in elderly people [letter] Lancet 1999353(9156)928

Shaw 2003 published data only

Dawson P Chapman KL Shaw FE Kenny RA Measuring the out-come of physiotherapy in cognitively impaired elderly patients whofall Physiotherapy 199783(7)352 [EMBASE 1997239545]ShawF Physiotherapy intervention for cognitively impaired elderly fallersattending casualty In National Research Register Oxford UpdateSoftware wwwnrrnhsukViewDocumentaspID=N0461021713(accessed 03 October 2006)Shaw F Risk modification of falls in cognitively impaired elderlypatients attending a casualty department A randomised controlledexplanatory study In National Research Register Oxford UpdateSoftware wwwnrrnhsukViewDocumentaspID=N0461044514(accessed 03 October 2006)lowast Shaw FE Bond J Richardson DA Dawson P Steen IN McKeithIG et alMultifactorial intervention after a fall in older people withcognitive impairment and dementia presenting to the accident andemergency department randomised controlled trial BMJ 2003326

(7380)73ndash5 [MEDLINE 12521968]Shaw FE Richardson DA Dawson P Steen IN McKeith IG Bond Jet alCan multidisciplinary intervention prevent falls in patients withcognitive impairment and dementia attending a casualty department[abstract] Age and Ageing 200029(Suppl 1)47



Shimada 2003 published and unpublished data

Shimada H personal communication July 29 2004Shimada H Uchiyama Y Kakurai S Specific effects of balance andgait exercises on physical function among the frail elderly ClinicalRehabilitation 200317(5)472ndash9 [EMBASE 2003345804]

Singh 2005 published data only

Singh NA Stavrinos TM Scarbek Y Galambos G Liber C FiataroneSingh MA A randomized controlled trial of high versus low intensityweight training versus general practitioner care for clinical depressionin older adults Journals of Gerontology Series A Biological Sciences

and Medical Sciences 200560(6)768ndash76 [MEDLINE 15983181]

Sohng 2003 published data only

Sohng K-Y Moon J-S Song H-H Lee K-S Kim Y-S Fall preventionexercise program for fall risk factor reduction of the community-dwelling elderly in Korea Yonsei Medical Journal 200344(5)883ndash91 [MEDLINE 14584107]

Sumukadas 2007 published data only

Sumukadas D Witham MD Struthers AD McMurdo ME Effect ofperindopril on physical function in elderly people with functional im-pairment a randomized controlled trial CMAJ Canadian MedicalAssociation Journal 2007177(8)867ndash74 [MEDLINE 17923654]

Tennstedt 1998 published data only

Tennstedt S Howland J Lachman M Peterson E Kasten L Jette AA randomized controlled trial of a group intervention to reduce fearof falling and associated activity restriction in older adults Journals ofGerontology Series B Psychological Sciences and Social Sciences 199853(6)P384ndash92

Thompson 1996 published data only

Cameron I Kurrle S Cumming R Preventing falls in the elderlyat home a community- based program [comment on Med J Aust1996164530-2] Medical Journal of Australia 1996165459ndash60lowast Thompson PG Preventing falls in the elderly at home a commu-nity-based program Medical Journal of Australia 1996164530ndash2

Tideiksaar 1992 published data only

Tideiksaar R Falls among the elderly a community prevention pro-gram American Journal of Public Health 199282892ndash3


Tinetti ME Baker DI Gottschalk M Williams CS Pollack D Gar-rett P et alHome-based multicomponent rehabilitation program forolder persons after hip fracture a randomized trial Archives of Phys-

ical Medicine and Rehabilitation 199980916ndash22

Von Koch 2001 published data only

Thorsen AM Holmqvist LW de Pedro-Cuesta J Von Koch L Arandomized controlled trial of early supported discharge and contin-ued rehabilitation at home after stroke five-year follow-up of patientoutcome Stroke 200536(2)297ndash303 [MEDLINE 15618441]Thorsen AM Widen Holmqvist L von Koch L Early supporteddischarge and continued rehabilitation at home after stroke 5-yearfollow-up of resource use Journal of Stroke and Cerebrovascular Dis-

eases 200615(4)139ndash43lowast Von Koch L de Pedro-Cuesta J Kostulas V Almazan J WidenHolmqvist L Randomized controlled trial of rehabilitation at homeafter stroke one-year follow-up of patient outcome resource use andcost Cerebrovascular Diseases 200112(2)131ndash8Von Koch L Widen Holmqvist L Kostulas V Almazan J de Pedro-Cuesta J A randomized controlled trial of rehabilitation at home

after stroke in Southwest Stockholm outcome at six months Scan-

dinavian Journal of Rehabilitation Medicine 200032(2)80ndash6Widen Holmqvist L Von Koch L Kostulas V Holm M Widsell G etalA randomized controlled trial of rehabilitation at home after strokein southwest Stockholm Stroke 199829(3)591ndash7 [MEDLINE9506598]

Ward 2004 published data only

Ward CD Turpin G Dewey ME Fleming S Hurwitz B RatibS et alEducation for people with progressive neurological condi-tions can have negative effects evidence from a randomized con-trolled trial Clinical Rehabilitation 200418(7)717ndash25 [MED-LINE 15573827]

Wolf-Klein 1988 published data only

Wolf-Klein GP Silverstone FA Basavaraju N Foley CJ Pascaru AMa PH Prevention of falls in the elderly population Archives ofPhysical Medicine and Rehabilitation 198869689ndash91

Wolfson 1996 published data only

Judge JO Whipple RH Wolfson LI Effects of resistive and balanceexercises on isokinetic strength in older persons Journal of the Amer-ican Geriatrics Society 199442(9)937ndash46Pacala JT Judge JO Boult C Factors affecting sample selection in arandomized trial of balance enhancement The FICSIT study Jour-

nal of the American Geriatrics Society 199644(4)377ndash82lowast Wolfson L Whipple R Derby C Judge J King M Amerman P etalBalance and strength training in older adults intervention gainsand Tai Chi maintenance Journal of the American Geriatrics Society

199644498ndash506Wolfson L Whipple R Judge J Amerman P Derby C King MTraining balance and strength in the elderly to improve functionJournal of the American Geriatrics Society 199341341ndash3

Yardley 2007 published data only

Yardley L Nyman SR Internet provision of tailored advice on fallsprevention activities for older people a randomized controlled eval-uation Health Promotion International 200722(2)122ndash8 [MED-LINE 17355994]

Yates 2001 published data only

Yates SM Dunnagan TA Evaluating the effectiveness of a home-based fall risk reduction program for rural community-dwelling olderadults Journals of Gerontology Series A Biological Sciences and Med-ical Sciences 200156(4)M226ndash30

Ytterstad 1996 published data only

Sattin RW Preventing injurious falls [comment on J EpidemiolCommun Health 199650551-8] Lancet 1997349150lowast Ytterstad B The Harstad injury prevention study communitybased prevention of fall-fractures in the elderly evaluated by meansof a hospital based injury recording system in Norway Journal of

Epidemiology and Community Health 199650(5)551ndash8

References to studies awaiting assessment

Beyer 2007 published data only

Beyer N Simonsen L Bulow J Lorenzen T Jensen DV Larsen Let alOld women with a recent fall history show improved mus-cle strength and function sustained for six months after finishingtraining Aging-Clinical amp Experimental Research 200719(4)300ndash9[MEDLINE 17726361]



Di Monaco 2008 published data only

Di Monaco M Vallero F De Toma E De Lauso L Tappero R Ca-vanna A A single home visit by an occupational therapist reduces therisk of falling after hip fracture in elderly women a quasi-random-ized controlled trial Journal of Rehabilitation Medicine 200840(6)446ndash50

Madureira 2007 published data only

Madureira MM Takayama L Gallinaro AL Caparbo VF Costa RAPereira RM Balance training program is highly effective in improv-ing functional status and reducing the risk of falls in elderly womenwith osteoporosis a randomized controlled trial Osteoporosis Inter-national 200718(4)419ndash25 [PUBMED 17089080 ]

Pfeifer 2004 published data only

Minne HW Dobnig H Pfeifer M Suppan K Effects of vitaminD and calcium supplementation on falls and parameters of musclefunction a prospective randomized double-blind multicenter study[abstract] Osteoporosis International 200617(Suppl 2)S212Minne HW Dobnig H Pfeifer M Suppan K Effects of vitaminD and calcium supplementation on falls and parameters of mus-cle-function - a prospective randomized double-blind multi-centerstudy [abstract] Osteoporosis International 200617(Suppl 1)S21Pfeifer M Dobnig H Begerow B Suppan K Effects of vitaminD and calcium supplementation on falls and parameters of musclefunction a prospective randomized double-blind multi-centre study[abstract] Journal of Bone and Mineral Research 200419(Suppl 1)S58Pfeifer M Dobnig H Minne HW Suppan K Effects of vitamin Dand calcium supplementation on falls and parameters of muscle func-tion - a prospective randomized double-blind multi-center study[abstract] Osteoporosis International 200516(Suppl 3)S45

Sato 2005b published data only

Sato Y Kanoko T Satoh K Iwamoto J Menatetrenone and vitaminD2 with calcium supplements prevent nonvertebral fracture in elderlywomen with Alzheimerrsquos disease Bone 200536(1)61ndash8 [MED-LINE 15664003]

Weber 2008 published data only

Weber V White A McIlvried R An electronic medical record(EMR)-based intervention to reduce polypharmacy and falls in anambulatory rural elderly population Journal of General Internal

Medicine 200823(4)399ndash404 [PUBMED 18373136]

References to ongoing studies

Behrman published data only

Behrman R personal communication September 12 2006Behrman R A study into the prediction and prevention of disabilityand falls in the over 75 year population National Research Regis-ter Archive httpsportalnihracuk (accessed 31 March 2008) [NRR publication ID N0105125155]Behrman R Prediction and prevention of falls in the el-derly National Research Register (NRR) Archive httpsportalnihracukPagesNRRArchiveSearchaspx (accessed 31 De-cember 2007) [ NRR Publication ID N0105009461]

Blalock published data only

Preventing falls through enhanced pharmaceutical care ClinicalTri-alsgov httpclinicaltrialsgov (accessed 31 March 2008)

Ciaschini published data only

Ciaschini FORCE (Falls Fracture and Osteoporosis Risk ControlEvaluation) study ClinicalTrialsgov httpclinicaltrialsgovct2showNCT00465387 accessed 25 Dec 2008Ciaschini PM Straus SE Dolovich LR Goeree RA Leung KMWoods CR et alCommunity-based randomised controlled trial eval-uating falls and osteoporosis risk management strategies Trials 2008Nov 49(1)62 [Epub ahead of print] [PUBMED 18983670]

Cryer published data only

Allen A Simpson JM A primary care based fall prevention pro-gramme Physiotherapy Theory and Practice 199915(2)121ndash33[EMBASE 1999232162 ]Cryer C personal communication August 27 2006Cryer C personal communication Dec 15 2008Cryer C Prevention of falls in older people in Canterbury NationalResearch Register (NRR) Archive httpsportalnihracuk (accessed26 March 2008) [ NRR Publication ID N0582105006]

Donaldson published data only

Donaldson M personal communication October 17 2007Donaldson M Trial of a home based strength and balance retrain-ing program in reducing falls risk factors ClinicalTrialsgov httpclinicaltrialsgov (accessed 31 March 2008)Donaldson MG Falls risk in frail seniors clinical and methodological

studies [thesis] Vancouver (CA) Univ of British Columbia 2007Donaldson MG Khan KM Sobolev B Janssen P Cook WL McKayHA Action Seniors An RCT of the Otago Home Exercise Programto ameliorate fall risk factor profile in patients at high risk of falls[abstract] Annual Meeting of the American Society for Bone andMineral Research 2007 Sept 16-20 Honolulu (Hawaii)Liu-Ambrose T Donaldson MG Ahamed Y Graf P Cook WL CloseJ et alOtago home-based strength and balance retraining improvesexecutive functioning in older fallers a randomized controlled trialJournal of the American Geriatrics Society 200856(10)1821ndash30

Edwards published data only

Edwards N Cere M Leblond D A community-based interventionto prevent falls among seniors Family and Community Health 199315(4)57ndash65

Grove published data only

Grove M Effects of Trsquoai Chi training on general wellbeing and mo-tor performance in patients with Parkinsonrsquos Disease National Re-search Register (NRR) Archive httpsportalnihracuk (accessed26 March 2008) [ NRR Publication ID N0202102542]

Haines published data only

Haines T Assessment and prevention of falls functional decline andhospital re-admission in older adults post-hospitalisation AustralianNew Zealand Clinical Trials Registry httpwwwanzctrorgau(accessed 31 March 2008)

Hill a published data only

Hill K Blackberry I A randomised controlled trial to reduce fur-ther falls and injuries for older fallers presenting to an EmergencyDepartment Australian New Zealand Clinical Trials Registry httpwwwanzctrorgau (accessed 31 March 2008)Hill K Blackberry I RCT to reduce further falls and in-juries for older fallers presenting to an emergency departmentwwwclinicaltrialsgov (accessed 26 March 2008)



Hill b published data only

Hill K Falls prevention for stroke patients following discharge homeA randomised trial intervention Australian New Zealand ClinicalTrials Registry httpwwwanzctrorgau (accessed 31 March 2008)

Jee published data only

Jee J Wang JJ Rose K Landau P Lindley R Mitchell P Incorpo-rating vision and hearing tests into aged care assessment methodsand the pilot study Ophthalmic Epidemiology 200411(5)427ndash36[MEDLINE 15590588]

Johnson published data only

Johnson J Community care and hospital based collaborative fallsprevention project Australian New Zealand Clinical Trials Registerwwwanzctrorgau (accessed 31 March 2008)

Kenny unpublished data only

Brooksby W SAFE PACE 2 trial Syncope and falls inthe elderly - pacing and carotid sinus evaluation randomisedcontrol trial of cardiac pacing in older patients with carotidsinus hypersensitivity National Research register (NRR)archive httpsportalnihracukProfilesNRRaspxPublication_ID=N0183041329 (accessed 09 January 2008) [ NRR PublicationID N0183041329]Doig JC SAFE PACE 2 Syncope and falls in the elderly - pacingand carotid sinus evaluation A randomised controlled trial of cardiacpacing in older patients with falls and carotid sinus hypersensitivity(SAFE PACE 2) In National Research Register Oxford UpdateSoftware 2007 Issue 3 [ Publication ID N0504077783]Fotherby M SAFE PACE 2 - Syncope and falls in the elderly - pacingand carotid sinus evaluation a randomised control trial of cardiacpacing in older patients with falls and carotid sinus hypersensitivityNational Research Register (NRR) Archive httpsportalnihracuk(accessed 31 March 2008) [ NRR Publication IDN0123090677]Gray R SAFE PACE 2 - Syncope and falls in the elderly - pacingand carotid sinus evaluation a randomised control trial of cardiacpacing in older patients with falls and carotid sinus hypersensitivityIn National Research Register Oxford Update Software 2003Issue 2 [ Publication ID N0277056223]Holdright D A randomised control trial of cardiac pacing in olderpatients with falls and carotid sinus hypersensitivity In NationalResearch Register Oxford Update Software 2000 Issue 2 [ Pub-lication ID N0263052736]Kenny RA SAFE PACE 2 Syncope and falls in the elderly - Pacingand carotid sinus evaluation - A randomized controlled trial of cardiacpacing in older patients with falls and carotid sinus hypersensitivityEuropace 19991(1)69ndash72 [PUBMED 11220545 ]lowast Kenny RA Seifer C SAFE PACE 2 Syncope and falls in theelderly pacing and carotid sinus evaluation A randomized controltrial of cardiac pacing in older patients with falls and carotid sinushypersensitivity American Journal of Geriatric Cardiology 19998(2)87 [EMBASE 1999111785]OrsquoBrien A Syncope and falls in the elderly - pacing and carotid sinusevaluation a randomised controlled trial of cardiac pacing in olderpatients with falls and carotid sinus hypersensitivity Safe Pace 2 InNational Research Register Oxford Update Software 2001 Issue1 [ Publication ID N0232077535]Pascaul J Syncope and falls in the elderly - Pacing and carotid si-nus evaluation a randomised control trial of cardiac pacing in olderpatients with falls and carotid sinus hypersensitivity In National

Research Register Oxford Update Software 2000 Issue 3 [ Pub-lication ID M0021042314]

Klaber Moffett published data only

Klaber Moffett J Prevention of falls and injuries in a communitysample A randomised trial of exercise for older women (PREFICS)National Research Register (NRR) Archive httpsportalnihracuk(accessed 26 March 2008) [ NRR Publication ID N0084162084]

Lesser published data only

Lesser T personal communication September 07 2006Lesser THJ Vestibular rehabilitation in prevention of falls due tovestibular disorders in adults National Research Register (NRR)Archive httpsportalnihracukProfilesNRRaspxPublication_ID=N0025078568 (accessed 26 March 2008) [ NRR PublicationID N0025078568]

Lips published data only

Lips P Prevention of fall incidents in patients with a high riskof falling a multidiciplinairy study on the effects of transmuralhealth care compared to usual care Current Controlled Trials httpcontrolled-trialscom (accessed 31 March 2008)Peeters GM de Vries OJ Elders PJ Pluijm SM Bouter LM LipsP Prevention of fall incidents in patients with a high risk of fallingdesign of a randomised controlled trial with an economic evaluationof the effect of multidisciplinary transmural care BMC Geriatrics2007715 [MEDLINE 17605771]

Lord published data only

Lord SR Haran MJ VISIBLE study (Visual Intervention Strategy In-corporating Bifocal amp Long-Distance Eyeware) ClinicalTrialsgovhttpclinicaltrialsgov (accessed 32 March 2008)

Maki published data only

Maki B Evaluation of a balance-recovery specific falls prevention ex-ercise program ClinicalTrialsgov httpclinicaltrialsgov (accessed31 March 2008)

Masud published data only

Conroy S Morris R Masud T Multifactorial day hospital interven-tion to reduce falls in high risk older people in primary care a multi-centre randomised controlled trial ProFaNE (Prevention of FallsNetwork Europe) meeting 2004 June 11-13 Manchester (UK)Masud T Multifactorial day hospital intervention to reduce falls inhigh risk older people in primary care a multi-centre randomisedcontrolled trial Current Controlled Trials httpcontrolled-tri-alscom (accessed 31 March 2008)lowast Masud T Coupland C Drummond A Gladman J Kendrick DSach T et alMultifactorial day hospital intervention to reduce fallsin high risk older people in primary care a multi-centre randomisedcontrolled trial [ISRCTN46584556] Trials 200675ndash10

Menz published data only

Menz H Podiatry treatment to improve balance and prevent falls inolder people Australian New Zealand Clinical Trials Register httpwwwanzctrorgau (accessed 31 March 2008)lowast Spink MJ Menz HB Lord SR Efficacy of a multifaceted podiatryintervention to improve balance and prevent falls in older peoplestudy protocol for a randomised trial BMC Geriatrics 20088(1)30[PUBMED 19025668]

Miller published data only

Thomas SK Humphreys KJ Miller MD Cameron ID WhiteheadC Kurrle et alIndividual nutrition therapy and exercise regime a



controlled trial of injured vulnerable elderly (INTERACTIVE trial)BMC Geriatrics 200884 [MEDLINE 18302787]

Olde Rikkert published data only

Olde Rikkert M Randomized controlled trial to reduce falls and fearof falling in frail elderly ClinicalTrialsgov httpclinicaltrialsgov(accessed 26 March 2008)

Palvanen published data only

Palvanen M The Chaos Clinic for prevention of falls and relatedinjuries a randomised controlled trial Current Controlled Trialshttpwwwcontrolled-trialscom (accessed 31 March 2008)

Pighills published data only

Pighills A personal communication April 3 2006

Press published data only

Press Y Comprehensive intervention for falls prevention in the el-derly ClinicalTrialsgov httpclinicaltrialsgov (accessed 31 March2008)

Sanders published data only

Sanders K personal communication November 29 2007Sanders K Vitamin D intervention to prevent falls and fracturesand to promote mental well-being Australian New Zealand ClinicalTrials Registry httpwwwanzctrorgau (accessed 31 March 2008)

Schumacher published data only

Schumacher J Fall prevention by alfacalcidol and training Clinical-Trialsgov httpclinicaltrialsgov (accessed 31 March 2008)

Snooks published data only

Logan P An evaluation of the Primary Care falls prevention servicesfor older fallers presenting to the ambulance service National Re-search Register (NRR) Archive httpsportalnihracuk (accessed26 March 2008) [ NRR Publication ID N0171168738]Snooks H Evaluation of the costs and benefits of computerised on-scene decision support for emergency ambulance personnel to as-sess and plan appropriate care for older people who have fallena randomised controlled trial Current Controlled Trials httpwwwcontrolled-trialscom (accessed 17 October 2007)

Stuck published data only

Iliffe S Kharicha K Harari D Swift C Gillmann G Stuck AEHealth risk appraisal in older people 2 the implications for clin-icians and commissioners of social isolation risk in older peopleBritish Journal of General Practice 200757(537)277ndash82 [MED-LINE 17394730]Kharicha K Iliffe S Harari D Swift C Gillmann G Stuck AEHealth risk appraisal in older people 1 are older people living alonean rdquoat-riskldquo group British Journal of General Practice 200757(537)271ndash6 [MEDLINE 17394729]Stuck A personal communication Sept 27 2007Stuck A Disability prevention in the older population use of infor-mation technology for health risk appraisal and prevention of func-tional decline Current Controlled Trials httpcontrolled-trialscom(accessed 31 March 2008) [ ISRCTN28458424]lowast Stuck AE Kharicha K Dapp U Anders J Von Renteln-Kruse WMeier-Baumgartner HP et alThe PRO-AGE study an internationalrandomised controlled study of health risk appraisal for older personsbased in general practice BMC Medical Research Methodology 200772 [MEDLINE 17217546]

Taylor published data only

Taylor D An evaluation of the Accident Compensation Cor-poration (ACC) Tai Chi programme in older adults does itreduce falls Australian New Zealand Clinical Trials Registryhttpwwwanzctrorgau (accessed 31 March 2008) [ AC-TRN12607000018415]

Tousignant published data only

Tousignant M Falls prevention for frail older adults Cost-effi-cacy analysis of balance training based on Tai Chi controlled-tri-alscomISRCTN11861569 (accessed 19 September 2008)

Vind published data only

Vind AB personal communication March 30 2006Vind AB Examination and treatment of elderly after a fall Clini-calTrialsgov httpclinicaltrialsgov (accessed 17 October 2007)

Zeeuwe published data only

Zeeuwe PE Verhagen AP Bierma-Zeinstra SM Van Rossum E FaberMJ Koes BW The effect of Tai Chi Chuan in reducing falls amongelderly people design of a randomized clinical trial in the Nether-lands [ISRCTN98840266] BMC Geriatrics 200666 [MED-LINE 16573825]

Zijlstra published data onlylowast Zijlstra G van Haastregt JC van Eijk JT Kempen GI Evaluatingan intervention to reduce fear of falling and associated activity re-striction in elderly persons design of a randomised controlled trial[ISRCTN43792817] BMC Public Health 20055(1)26 [MED-LINE 15780139]Zijlstra GAR Van Haastregt JCM Van Eijk JT Van Rossum EStalenhoef PA Kempen GIJM Prevalence and correlates of fear offalling and associated avoidance of activity in the general populationof community-living older people Age and Ageing 200736(3)304ndash9 [MEDLINE 17379605]

Additional references

AGSBGS 2001

Anonymous Guideline for the prevention of falls in older personsAmerican Geriatrics Society British Geriatrics Society and AmericanAcademy of Orthopaedic Surgeons Panel on Falls Prevention Journalof the American Geriatrics Society 200149(5)664ndash72 [MEDLINE11380764]

Beswick 2008

Beswick AD Rees K Dieppe P Ayis S Gooberman-Hill R Hor-wood J et alComplex interventions to improve physical functionand maintain independent living in elderly people a systematic re-view and meta-analysis Lancet 2008371(9614)725ndash35 [MED-LINE 18313501]

Bischoff 2003

Bischoff HA Stahelin HB Dick W Akos R Knecht M Salis Cet alEffects of vitamin D and calcium supplementation on falls Arandomized controlled trial Journal of Bone and Mineral Research200318(2)343ndash51 [MEDLINE 12568412]

Boutron 2008

Boutron I Moher D Altman DG Schulz KF Ravaud P CON-SORT Group Extending the CONSORT statement to randomizedtrials of nonpharmacologic treatment explanation and elaborationAnnals of Internal Medicine 2008148(4)295ndash309 [MEDLINE18283207]



Broe 2007

Broe KE Chen TC Weinberg J Bischoff-Ferrari HA Holick MFKiel DP A higher dose of vitamin D reduces the risk of falls innursing home residents A randomized multiple-dose study Journalof the American Geriatrics Society 200755(2)234ndash9 [MEDLINE17302660]

Buchner 1993

Buchner DM Hornbrook MC Kutner NG Tinetti ME Ory MGMulrow CD et alDevelopment of the common data base for theFICSIT trials Journal of the American Geriatrics Society 199341297ndash308

Cameron 2005

Cameron I Murray GR Gillespie LD Cumming RG Robert-son MC Hill K et alInterventions for preventing falls inolder people in residential care facilities and hospitals [Protocol]Cochrane Database of Systematic Reviews 2005 Issue 3 [DOI10100214651858CD005465]

Campbell 1990

Campbell AJ Borrie MJ Spears GF Jackson SL Brown JS Fitzger-ald JL Circumstances and consequences of falls experienced by acommunity population 70 years and over during a prospective studyAge and Ageing 199019136ndash41

Campbell 1999c

Campbell AJ Robertson MC Gardner MM Norton RN BuchnerD Falls prevention over 2 years a randomized controlled trial inwomen 80 years and older Age and Ageing 199928513ndash18

Campbell 2004

Campbell MK Elbourne DR Altman DG CONSORT GroupCONSORT statement extension to cluster randomised trials BMJ

2004328(7441)702ndash8 [PUBMED 15031246]

Campbell 2005

Campbell AJ Robertson MC La Grow SJ Kerse NM SandersonGF Jacobs RJ et alRandomised controlled trial of prevention of fallsin people aged gt or =75 with severe visual impairment the VIP trialBMJ 2005331(7520)817 [PUBMED 16183652]

Campbell 2006

Campbell AJ Robertson MC Implementation of multifactorial in-terventions for fall and fracture prevention Age and Ageing 200635

Suppl 2ii60ndash4

Campbell 2007

Campbell AJ Robertson MC Rethinking individual and communityfall prevention strategies a meta-regression comparing single andmultifactorial interventions Age and Ageing 200736(6)656ndash62[PUBMED 18056731]

Chapuy 2002


International 200213(3)257ndash64

Close 2000

Close JCT Patel A Hooper R Glucksman E Jackson SHD SwiftCG PROFET improved clinical outcomes at no additional cost[abstract] Age and Ageing 200029(Suppl 1)48

Cummings 1995

Cummings SR Nevitt MC Browner WS Stone K Fox KM EnsrudKE et alRisk factors for hip fracture in white women Study of Os-teoporotic Fractures Research Group [see comments] New EnglandJournal of Medicine 1995332(12)767ndash73

Excel

Microsoft Excel X for Mac 8 Microsoft 2001

Findorff 2007

Findorff MJ Wyman JF Nyman JA Croghan CF Measuring thedirect healthcare costs of a fall injury event Nursing Research 200756(4)283ndash7 [MEDLINE 17625468]

Flicker 2005

Flicker L MacInnis RJ Stein MS Scherer SC Mead KE NowsonCA et alShould older people in residential care receive vitamin D toprevent falls Results of a randomized trial Journal of the American

Geriatrics Society 200553(11)1881ndash8 [MEDLINE 16274368]

Gates 2008

Gates S Fisher JD Cooke MW Carter YH Lamb SE Multifac-torial assessment and targeted intervention for preventing falls andinjuries among older people in community and emergency care set-tings systematic review and meta-analysis BMJ 2008336(7636)130ndash3 [MEDLINE 18089892]

Gillespie 2003

LD Gillespie WJ Gillespie MC Robertson SE Lamb RG Cum-ming BH Rowe Interventions for preventing falls in elderly peo-ple Cochrane Database of Systematic Reviews 2003 Issue 4 [DOI10100214651858CD000340]

Goodwin 2008

Goodwin VA Richards SH Taylor RS Taylor AH Campbell JLThe effectiveness of exercise interventions for people with Parkinsonrsquosdisease a systematic review and meta-analysis Movement Disorders

200823(5)631ndash40 [MEDLINE 18181210]

Haas 2006

Haas M Economic analysis of tai chi as a means of pre-venting falls and falls related injuries among older adultsCHERE working paper 20064 Sydney Australia Centrefor Health Economics Research and Evaluation University ofTechnology httpdatasearchutseduauchereresearchworking_paperscfm (accessed 27 March 2008)

Hauer 2006

Hauer K Lamb SE Jorstad EC Todd C Becker C ProFaNE-GroupSystematic review of definitions and methods of measuring falls inrandomised controlled fall prevention trials Age and Ageing 200635(1)5ndash10 [MEDLINE 16364930]

Higgins 2003

Higgins JP Thompson SG Deeks JJ Altman DG Measuring incon-sistency in meta-analyses BMJ 2003327(7414)557ndash60 [MED-LINE 12958120]

Higgins 2008a

Higgins JPT Altman DG (editors) Chapter 8 Assessing risk of biasin included studies Table 85c In Higgins JPT Green S (editors)Cochrane Handbook of Systematic Reviews of Interventions Version500 (updated February 2008) The Cochrane Collaboration 2008Available from wwwcochrane-handbookorg



Higgins 2008b

Higgins JPT Deeks JJ Altman DG (editors) Chapter 1634 Ap-proximate analyses of cluster-randomized trials for meta-analysis ef-fective sample sizes In Higgins JPT Green S (editors) CochraneHandbook of Systematic Reviews of Interventions Version 500(updated February 2008) The Cochrane Collaboration 2008Available from wwwcochrane-handbookorg

Jackson 2007

Jackson C Gaugris S Sen SS Hosking D The effect of cholecalciferol(vitamin D3) on the risk of fall and fracture a meta-analysis QJM

2007100(4)185ndash92 [MEDLINE 17308327]

Keene 1993

Keene GS Parker MJ Pryor GA Mortality and morbidity after hipfractures BMJ 1993307(6914)1248ndash50 [MEDLINE 8166806]

Kellogg 1987

Anonymous The prevention of falls in later life A report of theKellogg International Work Group on the Prevention of Falls by theElderly Danish Medical Bulletin 198734 Suppl 41ndash24 [MED-LINE 3595217]

Lamb 2005

Lamb SE Jorstad-Stein EC Hauer K Becker C Prevention of FallsNetwork Europe and Outcomes Consensus Group Development ofa common outcome data set for fall injury prevention trials the Pre-vention of Falls Network Europe consensus Journal of the American


Lamb 2007

Lamb SE Hauer K Becker C Manual for the fall prevention clas-sification system wwwprofaneeuorgprofane_documentsFalls_Taxonomypdf (accessed 20 June 2008)

Lefebvre 2008

Lefebvre C Manheimer E Glanville J Chapter 6 Searching forstudies In Higgins JPT Green S (editors) Cochrane Handbook forSystematic Reviews of Interventions Version 500 (updated Febru-ary 2008) The Cochrane Collaboration 2008 Available fromwwwcochrane-handbookorg

Lord 2008

Lord SR Lack of effectiveness of a multidisciplinary fall-prevention program in elderly people at risk a random-ized controlled trial [Commentary] Falls Links (availablefrom wwwpowmrieduaufallsnetworkfalls_links_newsletterhtm)2008 Vol 3 issue 43ndash4

McAlister 2003

McAlister FA Straus SE Sackett DL Altman DG Analysis andreporting of factorial trials a systematic review JAMA 2003289

(19)2545ndash53 [MEDLINE 12759326]

RevMan 5

The Nordic Cochrane Centre The Cochrane Collaboration Re-view Manager (RevMan) 50 Copenhagen The Nordic CochraneCentre The Cochrane Collaboration 2008

Richy 2008

Richy F Dukas L Schacht E Differential effects of D-hormoneanalogs and native vitamin D on the risk of falls a comparative meta-analysis Calcified Tissue International 200882(2)102ndash7 [MED-LINE 18239843]

Rizzo 1996

Rizzo JA Baker DI McAvay G Tinetti ME The cost-effectivenessof a multifactorial targeted prevention program for falls among com-munity elderly persons Medical Care 199634954ndash69

Robertson 2001c

Robertson MC Devlin N Scuffham P Gardner MM Buchner DMCampbell AJ Economic evaluation of a community based exerciseprogramme to prevent falls Journal of Epidemiology and Community

Health 200155(8)600ndash6 [MEDLINE 11449021]

Robertson 2001d

Robertson MC Development of a falls prevention programme for elderlypeople evaluation of efficacy effectiveness and efficiency [PhD thesis]Dunedin New Zealand University of Otago 2001

Robertson 2007

Robertson MC Campbell AJ What type of exercise reduces falls inolder people In MacAuley D Best T editor(s) Evidence-based

sports medicine 2nd Edition Oxford UK Blackwell Publishing2007135ndash66

Sach 2007

Sach TH Foss AJ Gregson RM Zaman A Osborn F Masud T etalFalls and health status in elderly women following first eye cataractsurgery an economic evaluation conducted alongside a randomisedcontrolled trial British Journal of Ophthalmology 200791(12)1675ndash9 [MEDLINE 17585002]

Salkeld 2000

Salkeld G Cumming RG OrsquoNeill E Thomas M Szonyi G West-bury C The cost effectiveness of a home hazard reduction programto reduce falls among older persons Australian and New ZealandJournal of Public Health 200024(3)265ndash71

Sattin 1992

Sattin RW Falls among older persons a public health perspectiveAnnual Review of Public Health 199213489ndash508

Sherrington 2008

Sherrington C Whitney J Lord S Herbert R Cumming R CloseJ Effective exercise for the prevention of falls - a systematic reviewand meta-analysis Journal of the American Geriatrics Society 2008Vol 56 issue 122234ndash43

Smeeth 2002

Smeeth L Ng ES Intraclass correlation coefficients for cluster ran-domized trials in primary care data from the MRC Trial of the As-sessment and Management of Older People in the Community Con-trolled Clinical Trials 200223(4)409ndash21 [MEDLINE 15837446]

Stata

Statacorp Stata Statistical Software 80 Statacorp 2003

Tinetti 1988

Tinetti ME Speechley M Ginter SF Risk factors for falls amongelderly persons living in the community New England Journal ofMedicine 19883191701ndash7

Tinetti 1997

Tinetti ME Williams CS Falls injuries due to falls and the riskof admission to a nursing home New England Journal of Medicine1997337(18)1279ndash84 [MEDLINE 9345078]



Vellas 1997

Vellas BJ Wayne SJ Romero LJ Baumgartner RN Garry PJ Fearof falling and restriction of mobility in elderly fallers Age and Ageing

199726(3)189ndash93 [MEDLINE 9223714]

Zecevic 2006

Zecevic AA Salmoni AW Speechley M Vandervoort AA Defining afall and reasons for falling comparisons among the views of seniorshealth care providers and the research literature Gerontologist 200646(3)367ndash76 [MEDLINE 16731875]

References to other published versions of this review

Gillespie 2008

Gillespie LD Robertson MC Gillespie WJ Lamb S Gates S Cum-ming RG et alInterventions for preventing falls in older people liv-ing in the community Cochrane Database of Systematic Reviews 2008Issue 2 [DOI 10100214651858CD000340]

lowast Indicates the major publication for the study



C H A R A C T E R I S T I C S O F S T U D I E S

Characteristics of included studies [ordered by study ID]

Ashburn 2007

Methods RCTLosses 16 of 142 (11)

Participants Setting community UKN = 142Sample people with Parkinsonrsquos disease recruited from a specialist clinical database (39 women)Age range 44-91 mean 721 (SD 92)Inclusion criteria idiopathic PD living at home history of falls in previous yearExclusion criteria cognitively impaired

Interventions 1 Weekly 1 hour home-based exercise session for 6 weeks with physiotherapist (strengtheningflexibility balance training and walking) also taught fall prevention strategies Encouraged toexercise daily Monthly phone call after 6 weeks2 Control usual care

Outcomes 1 Number of people falling2 Number sustaining a fracture

Notes

Risk of bias

Item Authorsrsquo judgement Description

Adequate sequence generation Unclear Quote rdquorandomisation was stratified by NHS Trust using blocks of sizefourldquo

Allocation concealment Yes Quote rdquotreating physiotherapist obtained random allocation by telephon-ing Medical Statistics Group University of Southamptonldquo

BlindingFalls

Unclear Falls recorded by participants who were aware of their group allocation

BlindingFractures

Unclear Fractures recorded by participants who were aware of their group alloca-tion

Low risk of bias in recall of falls Yes Falls and fractures recorded prospectively by participants using diariessubmitted monthly



Assantachai 2002

Methods CCT (cluster randomised)Losses 156 of 1043 (15)

Participants Setting community Bangkok ThailandN = 1043Sample people living in 11 selected urban communities (64 women)Age mean 676 (SD 62)Inclusion criteria aged at least 60 living in one of the selected communities

Interventions 1 Educational leaflet and free access to geriatric clinic Leaflet about locally identified risk factorsfor falling (kyphoscoliosis nutritional status ADL hypertension special sense function cognitiveproblems) and ways of preventing correcting coping with them Assessed musculoskeletal defor-mity arthralgia hypertension ADL mobility gait hearing vision and presumably any problemsaddressed at geriatric clinic2 Control no intervention

Outcomes 1 Number of people falling

Notes

Risk of bias


Adequate sequence generation No Communities drawn from pool of 20 until 1043 subjects recruited Com-munities then allocated to intervention (odd number) or control (evennumber) using enrolment sequence (information provided by author)

Allocation concealment No Alternation

BlindingFalls


Low risk of bias in recall of falls Unclear Interval recall Falls ascertained by postcards every 2 months and phonecall if no card returned

Ballard 2004


Participants Setting community USAN = 40Sample volunteersAge mean 729 (SD 6)



Ballard 2004 (Continued)

Inclusion criteria aged 65 and over ambulatory community dwelling history of falling in previousyear or fear of future fall healthy enough to do moderate exerciseExclusion criteria cardiovascular disease or extreme vertigo that might prohibit moderate exerciserequiring walker for support

Interventions 1 Exercise sessions (warm up low impact aerobics exercise for strength and balance cool down)1 hour x3 per week for 15 weeks Plus 6 home safety education classes2 Control exercise sessions as above 1 hour x3 per week for 2 weeks + videotape so could continueat home Plus 6 home safety education classes as above

Outcomes 1 Rate of falls2 Number of people fallingFalls a secondary outcome of study Other outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquoassigned to exercise and control groups using stratified randomi-sationldquo

Allocation concealment Unclear Insufficient information to permit judgment

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocation

Low risk of bias in recall of falls No Falls identified retrospectively during intervention at each home safetyclass (every two months) and by telephone follow up one year after endof intervention

Barnett 2003


Participants Setting community AustraliaN = 163Sample elderly people identified (67 women) as at risk of falling by general practitioner orhospital physiotherapist using assessment toolAge mean 749 (SD 109)Inclusion criteria age over 65 years identified as rsquoat riskrsquo of falling (one or more of the followingrisk factors lower limb weakness poor balance slow reaction time)



Barnett 2003 (Continued)

Exclusion criteria cognitive impairment degenerative conditions eg Parkinsonrsquos disease or med-ical condition involving neuromuscular skeletal or cardiovascular system that precluded takingpart in exercise programme

Interventions 1 Exercise sessions (stretching and for strength balance coordination aerobic capacity) byaccredited exercise instructor in groups of 6 - 18 1 hour per week for 4 terms for 1 year (37classes)Home exercise programme based on class content + diaries to record participation2 Control no exercise interventionBoth groups received information on strategies for avoiding falls eg hand and foot placement ifloss of balance occurred

Outcomes 1 Rate of falls2 Number of people fallingOther outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquorandomised in matched blocksldquo (N = 6)

Allocation concealment Yes Consecutively numbered opaque envelopes

BlindingFalls


Low risk of bias in recall of falls Unclear Interval recall Falls identified by postal survey at the end of each calendarmonth Phoned if not returned within 2 weeks

Bischoff-Ferrari 2006


Participants Setting community Boston MA USAN = 445Sample men and women recruited by direct mailings and presentations (sample frame not given)(55 women)Age mean 71Inclusion criteria aged 65 and overExclusion criteria current cancer or hyperparathyroidism a kidney stone in last 5 years renaldisease bilateral hip surgery therapy with a bisphosphonate calcitonin oestrogen tamoxifen or



Bischoff-Ferrari 2006 (Continued)

testosterone in past 6 months or fluoride in past 2 years femoral neck bone mineral density morethan 2 SD below the mean for subjects of the same age and sex dietary calcium intake exceeding1500 mg per day laboratory evidence of kidney disease

Interventions 1 Cholecalciferol (700 IU vitamin D) and calcium citrate malate (500 mg elemental calcium)orally daily at bedtime for 3 years2 Control double placebo tablets

Outcomes 1 Rate of falls2 Number of people falling3 Number sustaining a fracture4 Number of people with adverse effectsOther outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquorandomly assignedldquo rdquorandom group assignment was performedwith stratification according to sex race and decade of ageldquo


BlindingFalls

Yes Falls reported by participants who were blind to their group allocation(placebo-controlled trial)

BlindingFractures

Yes Fractures reported at 6 monthly visit (placebo-controlled trial)

Low risk of bias in recall of falls Yes Asked to send a postcard after any fall Telephone call to verify circum-stances Subjects reported any additional falls at 6 monthly follow-upvisit Non-vertebral fractures reported at 6 monthly follow-up visit andverified by review of X-ray reports or hospital records

Brown 2002

Methods RCT Individually randomised but six clusters containing couples at same addressLosses 41 of 149 (28)



Brown 2002 (Continued)

Participants Setting community Perth Western AustraliaN = 149Sample men and women recruited by press releases in 11 newspapers and information brochuresdistributed to organisations GPs etc (79 women)Age N = 101 aged 75-84 N = 48 aged 85-94Inclusion criteria age 75 and over community living (house flat or retirement villa) independentin basic ADL able to walk 20 meters without personal assistanceExclusion criteria cognitive impairment (MMSE le24) various conditions eg angina claudica-tion cerebrovascular disease low or high blood pressure major systemic disease mental illness

Interventions 1 Exercise intervention to improve cardiovascular endurance general muscle performance bal-ance co-ordination and flexibility 2x per week for 60 minutes for 16 weeks (32 hours)2 Social intervention for 13 weeks involving presentations of travel slides and videos by partici-pants3 Control no intervention

Outcomes 1 Number of participants falling

Notes

Risk of bias


Adequate sequence generation Yes Quote rdquorandomised into one of three groups using a table of randomnumbersldquo

Allocation concealment Yes Randomised into one of three groups rdquoby a physiotherapist uninvolvedin the studyldquo

BlindingFalls


Low risk of bias in recall of falls Yes Participants provided details of falls in monthly report sheet returned inreply paid addressed envelopes

Buchner 1997a

Methods RCTLosses 15 of 105 (14) (14 from intervention groups)

Participants Setting community Seattle USAN = 105Sample HMO members (FICSIT intervention groups only)Age mean 75



Buchner 1997a (Continued)

Inclusion criteria aged 68 to 85 unable to do 8 step tandem gait test without errors below 50thpercentile in knee extensor strength for height and weightExclusion criteria active cardiovascular pulmonary vestibular and bone disease positive cardiacstress test body weight gt180 ideal major psychiatric illness active metabolic disease chronicanaemia amputation chronic neurological or muscle disease inability to walk dependency ineating dressing transfer or bathing terminal illness inability to speak English or complete writtenforms

Interventions Randomised into 7 groups 6 intervention groups (3 FICSIT trial 3 MoveIT trial) and 1 controlgroup Only FICSIT trial and control groups included in this reviewSupervised exercise classes 1 hour x 3 per week for 24-26 weeks followed by unsupervised exercise1 Six months endurance training (ET) (stationary cycles) with arms and legs propelling wheel2 Six months strength training (ST) classes (using weight machines for resistance exercises forupper and lower body)3 Six months ST plus ET4 Control usual activity levels but rsquoallowed to exercise after 6 monthsrsquoExercise sessions started with a 10 to 15 minute warm-up and ended with a 5 to 10 minute cooldown

Outcomes Fall outcomes reported for any exercise (all 3 groups combined) compared with control group(states rsquoa priori decisionrsquo)1 Rate of falls2 Number of people falling3 Number sustaining a fracture4 Number of people with adverse effects

Notes Seattle FICSIT trial [Province 1995]Only 13 of original sample randomisedFalls not primary outcomeOther outcomes assessed at end of intervention (6 months) then rdquocontrol group allowed to exerciseafter 6 monthsldquo 7 out of 30 subjects did

Risk of bias


Adequate sequence generation Yes Randomised rdquousing a variation of randomly permuted blocksldquo


BlindingFalls





Low risk of bias in recall of falls Yes Falls reported immediately by mail also monthly postcard return tele-phone follow up if no postcard received

Bunout 2005


Participants Setting community ChileN = 298Sample men and womenAge mean 75 (SD 5)Inclusion criteria rdquoelderly subjectsldquo consenting to participate able to reach community centreExclusion criteria severe disabling condition cognitive impairment (MMSE lt 20)

Interventions 1 Exercise class 1 hour 2x per week for 1 year moderate-intensity resistance exercise training(functional weight bearing exercises exercises with TheraBands and walking (see Appendix 2 ofsupplementary data on journal website for details)2 Control no intervention

Outcomes 1 Number of people fallingOther outcomes reported but not included in this review

Notes Journal website for supplementary data wwwageingoupjournalsorg Additional data obtainedfrom author

Risk of bias


Adequate sequence generation Yes Randomised using computer generated random number table


BlindingFalls


Low risk of bias in recall of falls Unclear Interval recall Falls ascertained at monthly outpatient clinic or by tele-phone



Campbell 1997


Participants Setting community Dunedin New ZealandN = 233Sample women identified from general practice registersAge mean 841 (SD 31)Inclusion criteria at least 80 years old community livingExclusion criteria cognitive impairment not ambulatory in own residence already receivingphysiotherapy

Interventions Baseline health and physical assessment for both groups1 1 hour visits by physiotherapist x 4 in first two months to prescribe home based individualisedexercise and walking programmeExercise 30 minutes x 3 per week plus walk outside home x 3 per week Encouraged to continuefor 1 yearRegular phone contact to maintain motivation after first 2 months2 Control social visit by research nurse x 4 in first two months Regular phone contact

Outcomes 1 Rate of falls2 Number of people falling

Notes Otago Exercise Programme manual can be ordered from http wwwaccconzotagoexerciseprogramme

Risk of bias


Adequate sequence generation Yes Allocation schedule developed using computer generated numbers

Allocation concealment Yes Assignment by independent person off site

BlindingFalls


Low risk of bias in recall of falls Yes Falls recorded daily on postcard calendars mail registration monthly bypostcard telephone follow up



Campbell 1999


Participants Setting community Dunedin New ZealandN = 93Sample identified from general practice registers (83 women)Age mean 747 (SD 72)Inclusion criteria at least 65 years old currently taking a benzodiazepine any other hypnotic orany antidepressant or major tranquillizer ambulatory in own residence not receiving physiother-apy thought by GP to benefit from psychotropic medication withdrawalExclusion criteria cognitive impairment

Interventions Baseline assessment1 Gradual withdrawal of psychotropic medication over 14 week period plus home based exerciseprogramme2 Psychotropic medication withdrawal with no exercise programme3 No change in psychotropic medication plus exercise programme4 No change in psychotropic medication no exercise programmeExercise programme 1 hour physiotherapist visits x 4 in first two months to prescribe home basedindividualised exercises (muscle strengthening and balance retraining exercises 30 min x 3 perweek) and walking x 2 per weekRegular phone contact to maintain motivationStudy capsules created by grinding tablets and packing into gelatin capsules Capsules containinginert and active ingredients looked and tasted the same

Outcomes 1 Rate of falls2 Number of people falling3 Number sustaining an adverse effect

Notes Only 19 randomisedPsychotropic medications recorded one month after completion of studyEight of the 17 who had taken the placebo for 30 weeks had restarted one month after end ofstudyOtago Exercise Programme manual can be ordered from http wwwaccconzotagoexerciseprogramme

Risk of bias


Adequate sequence generation Yes 2 by 2 factorial design Allocation schedule developed using computergenerated numbers




Campbell 1999 (Continued)

BlindingFalls



Campbell 2005

Methods RCT 2 by 2 factorial designLosses 30 of 391 (8)

Participants Setting community New ZealandN = 391Sample men and women with severe visual impairment (visual acuity 624 or worse) identifiedin blind register university and hospital outpatient clinics and private ophthalmology practice(68 women)Age mean (SD) 836 (48) years range 75-96Inclusion criteria vision worse than 624 in better eye age ge 75 yearsExclusion criteria unable to walk around home

Interventions 1 Home safety programme2 Otago Exercise Programme plus vitamin D supplements3 Both of the above4 Control x2 one-hour social visits during the first 6 months of the trial

Outcomes 1 Rate of falls2 Number of people falling3 Number of people with adverse effects


Risk of bias


Adequate sequence generation Yes 2 by 2 factorial design Computer generated random numbers

Allocation concealment Yes Schedule held by independent person at separate site telephone access

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationPhoned by independent assessor blind to allocation Person classifying fallevents also blind to allocation




Low risk of bias in recall of falls Yes Prospective Falls recorded on daily on monthly pre-paid postcard calen-dars telephone follow up

Carpenter 1990

Methods RCT (Individually randomised but small number of clusters as husbands allocated to same group)Losses 172 of 539 (32)

Participants Setting community Andover United Kingdom N = 539Sample women and men recruited from patient lists of two general medical practices The samplerepresents 895 of those in the age group in the participating practices (65 women)Age 75 years or over 23 men and 49 women were over 85 yearsInclusion criteria aged 75 and over living in Andover areaExclusion criteria living in residential care

Interventions 1 Visit by trained volunteers for dependency surveillance using Winchester disability rating scaleThe intervention was stratified by degree of disability on the entry evaluation For those with nodisability the visit was every six months for those with disability three months Scores comparedwith previous assessment and referral to GP if score increased by 5 or more2 Control no disability surveillance between initial and final evaluation

Outcomes 1 Rate of falls (in each group in the month before the final interview at 3 years)Other outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Yes Randomised by random number tables


BlindingFalls


Low risk of bias in recall of falls Unclear Retrospective recall but over one month period



Carter 1997


Participants Setting community Hunter Valley AustraliaN = 658Sample men and women identified by 37 general practitioners as meeting inclusion criteriaAge 70 or olderInclusion criteria aged 70 and over able to speak and understand English living independentlyat home in a hostel or in a retirement villageExclusion criteria psychiatric disturbance affecting comprehension of the aims of the study

Interventions 1 Brief feedback on home safety plus pamphlets on home safety and medication use (low intensityintervention)2 Action plan for home safety plus medication review (high intensity intervention)3 Control no intervention during study period but intervention after the end of the study period

Outcomes 1 Number of people falling (during previous month at 3 6 and 12 months)

Notes Unpublished study

Risk of bias


Adequate sequence generation Yes Random number generator


BlindingFalls


Low risk of bias in recall of falls No Retrospective recall at 3 6 and 12 months

Carter 2002


Participants Setting community Vancouver CanadaN = 93Subjects community dwelling osteoporotic womenAge mean 69 (SD 3)Inclusion criteria aged 65 to 75 years residents of greater Vancouver osteoporotic (based onBMD)



Carter 2002 (Continued)

Exclusion criteria lt 5 years post menopause weighed gt 130 ideal body weight other con-traindications to exercising already doing gt 8 hoursweek moderate to hard exercise planning tobe out of city gt 4 weeks during 20 week programme

Interventions 1 Exercise class (Osteofit) for 40 minutes 2 x per week for 20 weeks in community centresClasses of 12 per instructor 8 to 16 strengthening and stretching exercises using Theraband elasticbands and small free weights Bimonthly social seminar2 Control usual routine activities and bimonthly social seminar separate from intervention group

Outcomes 1 Rate of fallsOther outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Yes Randomised by computer generated programme


BlindingFalls


Low risk of bias in recall of falls Yes Falls recorded in falls calendars returned monthly

Cerny 1998

Methods RCTLosses none described

Participants Setting community California USAN = 28Sample community dwelling rdquowell elderlyldquo Age mean 71 (SD 4)Inclusion criteria none describedExclusion criteria none described

Interventions 1 Exercise programme of progressive resistance stretching aerobic and balance exercises and briskwalking over various terrains for 1 and a half hours 3 x weekly for 6 months2 Control no intervention

Outcomes 1 Number of people fallingOther outcomes reported but not included in this review Falls a secondary outcome



Cerny 1998 (Continued)

Notes Contact with lead author but no full paper or report prepared

Risk of bias


Adequate sequence generation Yes Randomised by coin toss Individually randomised but some clusters egcouples or two ladies where one was dependent on the other for transport(information from author)

Allocation concealment No Coin toss on site

BlindingFalls


Low risk of bias in recall of falls No Assume retrospective recall and 3 and 6 months assessment

Clemson 2004

Methods RCT Randomised in blocks of four stratified by sex and number of falls in previous 12 monthsLosses none described

Participants Setting community Sydney AustraliaN = 310Sample volunteer community dwelling men and women recruited by various strategies (74women)Age mean 78 (SD 5)Inclusion criteria aged 70 and over community dwelling fallen in past year or felt themselvesto be at risk of falling Exclusion criteria dementia (gt 3 errors on Short Portable Mental StatusQuestionnaire) homebound unable to independently leave home unable to speak English

Interventions Both groups received baseline assessment at home before randomisation1 Stepping On programme Multifaceted small-group (N =12) learning environment to encourageself efficacy behaviour change and reduce falls using decision making theory and a variety oflearning strategies Facilitated by OT Two hours weekly for 7 weeks taught exercises and practicedin classes OT home visit within 6 weeks of final programme session booster session 3 monthsafter final session2 Control at least 2 social visits from student OT with no discussion of falls or fall prevention




Clemson 2004 (Continued)

Notes Details of programme in Appendix A of Clemson 2004 risk appraisal exercise moving safelyhome hazards community safety footware vision and falls vitamin D hip protectors medicationmanagement mobility mastery review and plan

Risk of bias


Adequate sequence generation Unclear Quote rdquoRandomised by researcher not involved in subject screening orassessmentldquo Method not described


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Monthly falls postcard calendar

Close 1999


Participants Setting community London United KingdomN = 397Sample community dwelling individuals presenting at AampE after a fall Admitted patients notrecruited until dischargeAge mean 782 (SD 75)Inclusion criteria aged 65 and over history of fallingExclusion criteria cognitive impairment (AMT lt7) and no regular carer (for informed consentreasons) speaking little or no English not living locally

Interventions 1 Medical and occupational therapy assessments and interventionsMedical assessment to identify primary cause of fall and other risk factors present (general exam-ination and visual acuity balance cognition affect medications) Intervention and referral as re-quired Home visit by occupational therapist (functional assessment and environmental hazards)Advice equipment and referrals as required2 Control usual care only


Notes



Close 1999 (Continued)

Risk of bias


Adequate sequence generation Yes Randomised by random numbers table

Allocation concealment Yes List held independently of the investigators

BlindingFalls


Low risk of bias in recall of falls Yes Prospective Falls diary with 12 monthly sheets collected every 4 months

Coleman 1999

Methods RCT Cluster randomised Unit of randomisation physician practiceLosses 56 of 169 (33)

Participants Setting HMO members Washington USAN = 169Sample community dwelling men and women in 9 physician practices in an ambulatory clinicAge mean 77Inclusion criteria aged 65 and over high risk of being hospitalised or of developing functionaldecline community dwellingExclusion criteria living in nursing home terminal illness moderate to severe dementia or rdquotooillldquo (physicianrsquos judgment)

Interventions 1 Half-day Chronic Care Clinics every 3-4 months in 5 practices focusing on planning chronicdisease management (physician and nurse) reducing polypharmacy and high risk medications(pharmacist) patient self managementsupport group2 Control usual care (4 practices)


Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquorandomized using simple randomizationldquo

Allocation concealment No Cluster randomised



Coleman 1999 (Continued)

BlindingFalls


Low risk of bias in recall of falls No Falls recorded retrospectively by questionnaire at 12 and 24 months

Cornillon 2002


Participants Setting community St Eacutetienne FranceN = 303Subjects community dwelling and independent in ADL (83 women)Age mean 71Inclusion criteria aged over 65 living at home ADL independent consentedExclusion criteria cognitively impaired (MMSE lt20) obvious disorder of walking or balance

Interventions 1 Information on fall risk and balance and sensory training in groups of 10-16 One session perweek for 8 weeks Session started with foot and ankle warm-up (walking on tip toe and on heelsetc) walking following verbal orders walking bare foot on different surfaces standing on one legwith eyes open and shut practicing getting up from the floor2 Control normal activities


Notes

Risk of bias




BlindingFalls


Low risk of bias in recall of falls Yes Prospective Falls recorded on 6 monthly falls calenders



Cumming 1999

Methods RCT (randomised consent design)Losses 142 of 530 (27)

Participants Setting community Sydney AustraliaN = 530Sample community dwelling people recruited in hospital wards clinics and day care centresAge mean 77 (SD 72)Inclusion criteria aged 65 and over living in the community and within geographically definedstudy areaExclusion criteria cognitively impaired and not living with someone who could give informedconsent and report falls if OT home visit already planned as part of usual care

Interventions 1 One home visit by experienced occupational therapist assessing environmental hazards (stan-dardised form) and supervision of home modifications Telephone follow up after 2 weeks2 Control usual care


Notes

Risk of bias


Adequate sequence generation Yes Stratified block randomisation using random numbers table

Allocation concealment Yes Randomised off site by person not involved in recruitment

BlindingFalls


Low risk of bias in recall of falls Yes Prospective Falls ascertained using monthly falls calendar

Cumming 2007


Participants Setting community Sydney AustraliaN = 616Sample men and women from outpatient aged care services some volunteers recruited by adver-tisement (68 women)Age mean 806 (SD 6) years



Cumming 2007 (Continued)

Inclusion criteria age 70 and older living independently in the community no cataract surgeryor new eye glass prescription in previous 3 months participant or care giver able to completemonthly falls calendarExclusion criteria none noted

Interventions 1 Vision tests and eye examinations Dispensing of new spectacles if required Referral for expe-dited ophthalmology treatment if appropriate occular pathology identified Mobility training andcanes if required2 Control usual care

Outcomes 1 Rate of falls2 Number of people falling3 Number sustaining a fracture

Notes

Risk of bias


Adequate sequence generation Unclear Not described


BlindingFalls


BlindingFractures

Unclear Fractures reported by participants who were aware of their group alloca-tion

Low risk of bias in recall of falls Yes Monthly falls calendar

Davison 2005

Methods RCTLosses 31 if 313 (9)

Participants Setting AampE Newcastle UKN = 313Sample community-dwelling cognitively intact presenting at AampE with a fall or fall-relatedinjury ( women)Age mean 77 (SD 7)Inclusion criteria age gt 65 years presenting at AampE with a fall or fall related injury history of atleast one additional fall in previous year



Davison 2005 (Continued)

Exclusion criteria cognitively impaired (MMSE lt 24) gt 1 previous episode of syncope immobilelive gt 15 miles away from AampE registered blind aphasic clear medical explanation for their falleg acute myocardial infarction stroke epilepsy enrolled in another study

Interventions 1 Multifactorial post-fall assessment and intervention Hospital-based medical assessment and in-tervention fall history and examination including medications vision cardiovascular assessmentlaboratory blood tests ECG Home-based physiotherapist assessment and intervention gait bal-ance assistive devices footwear Home-based OT home hazard assessment and interventions2 Control usual care


Notes Only one participant in residentialnursing care More detailed description of intervention onjournal website (wwwageingoupjournalsorg)

Risk of bias


Adequate sequence generation Yes Randomised by computer-generated block randomisation


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Falls data collected using fall diaries returned 4 weekly

Day 2002

Methods RCT Factorial designLosses 17 of 1107 (15)

Participants Setting community Melbourne AustraliaN = 1107Sample community dwelling men and women identified from electoral roll (598 women)Age mean 761 (SD 50)Inclusion criteria aged 70 and over living in own home or apartment or leasing similar accom-modation and able to make modificationsExclusion criteria if not expected to remain in area for 2 years (except for short absences) hadparticipated in regular to moderate physical activity with a balance component in previous 2months unable to walk 10-20 m without rest or help or having angina had severe respiratoryor cardiac disease had a psychiatric illness prohibiting participation had dysphasia had recent



Day 2002 (Continued)

major home modifications had an education and language adjusted score gt4 on the short portablemental status questionnaire or did not have approval of their general practitioner

Interventions 1 Exercise weekly class of 1 hour for 15 weeks plus daily home exercises Designed by physio-therapist to improve flexibility leg strength and balance (or less demanding routine depending onsubjectrsquos capability)2 Home hazard management hazards removed or modified by participants or City of Whitehorsersquoshome maintenance programme Staff visited home provided quote for work including free labourand materials up to $A 1003 Vision improvement assessed at baseline using dual visual acuity chart Referred to usual eye careprovider general practitioner or local optometrist if not already receiving treatment for identifiedimpairment4 (1) + (2)5 (1) + (3)6 (3) + (2)7 (1) + (2) + (3)8 No intervention Received brochure on eye care for over 40 year olds


Notes

Risk of bias


Adequate sequence generation Yes Randomised by rdquoadaptive biased coinldquo technique to ensure balancedgroup numbers

Allocation concealment Yes Computer generated by an independent third party contacted by tele-phone

BlindingFalls


Low risk of bias in recall of falls Yes Falls reported using monthly postcard to record daily falls Telephonefollow-up if calendar not returned within 5 working days of the end ofeach month or reporting a fall



Dhesi 2004

Methods RCTLosses 16 of 139 (12) (see Notes)

Participants Setting community United KingdomN = 140Sample patients attending a falls clinic (77 women)Age mean 768 (SD 62)Inclusion criteria aged 65 and over living in own home fallen in previous 8 weeks normal bonechemistry 25 OHD le 12 mcglitreExclusion criteria AMT lt 710 taking vitamin D or calcium supplements history of chronicrenal failure alcohol abuse conditions or medications likely to impair postural stability or vitaminD metabolism

Interventions 1 One intramuscular injection (2 ml) of 600000 IU ergocalciferol2 Control one placebo injection of 2 ml normal saline


Notes Flowchart in Figure 1 shows N = 139 randomised with 70 in intervention group but Table 1(baseline characteristics) shows N = 138 randomised with 69 in intervention group

Risk of bias


Adequate sequence generation Yes Randomised in blocks of 20 by computer programme

Allocation concealment Yes Randomised independently of the investigators

BlindingFalls


Low risk of bias in recall of falls Yes Falls recorded in falls diary which was reviewed at follow-up assessment

Dukas 2004


Participants Setting community Basel SwitzerlandN = 378Sample volunteers recruited from long term cohort study and newspaper advertisements (52women)Age mean 75 (SD 42)



Dukas 2004 (Continued)

Inclusion criteria aged over 70 mobile independent lifestyleExclusion criteria primary hyperparathyroidism polyarthritis or inability to walk calcium sup-plementation gt 500 mgd vitamin D intake gt 200 IUday active kidney stone disease history ofhypercalcuria cancer or other incurable diseases dementia elective surgery planned within next3 months severe renal insufficiency fracture or stroke within last 3 months

Interventions 1 Alfacalcidol (Alpha D3 TEVA) 1 mcg per day for 36 weeks2 Placebo daily for 36 weeks

Outcomes 1 Rate of falls2 Number of people falling3 Number of people with adverse effectsOther outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Yes Randomised using rdquonumbered containersldquo numbered and blinded byindependent statistical group

Allocation concealment Yes Numbered and blinded by independent statistical group

BlindingFalls


Low risk of bias in recall of falls Unclear Questionnaire about incidence of falls at clinic visits (4 weeks 12 weeksand every 12 weeks subsequently to 36 weeks) Subjects asked to recordfalls in a diary and to telephone within 48 hours of a fall

Elley 2008


Participants Setting Hutt Valley New ZealandN = 312Sample patients from 19 primary care practices (69 women)Age mean 808 (SD 5)Inclusion criteria aged 75 and over (gt 50 years for Maori and Pacific people) fallen in last yearliving independently



Elley 2008 (Continued)

Exclusion criteria unable to understand study information and consent processes unstable orprogressive medical condition severe physical disability dementia (lt 7 on Abbreviated MentalTest Score)

Interventions 1 Community-based nurse assessment of falls and fracture risk factors home hazards referral toappropriate community interventions and strength and balance exercise programme2 Control usual care and social visits


Notes

Risk of bias


Adequate sequence generation Yes Quote rdquocomputer randomisationldquo

Allocation concealment Yes Quote rdquoindependent researcher at a distant siteldquo

BlindingFalls

Unclear Participants not blind to allocation Assessors blind to allocation

Low risk of bias in recall of falls Yes Quote rdquoPostcard calendars completed daily and posted monthlyldquo

Fabacher 1994


Participants Setting community California USAN = 254Sample men and women aged over 70 years and eligible for veterans medical care Identified fromvoter registration lists and membership lists of service organisations (2 women)Age mean 73 yearsInclusion criteria aged 70 and over not receiving health care at Veterans Administration MedicalCentreExclusion criteria known terminal disease dementia

Interventions 1 Home visit by health professional to screen for medical functional and psychosocial problemsfollowed by a letter for participants to show to their personal physician Targeted recommendationsfor individual disease states preventive health practices2 Control follow-up telephone calls for outcome data only



Fabacher 1994 (Continued)


Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquorandomly assigned using randomly generated assignmentcards in sealed envelopesldquo Judged to be unclear

Allocation concealment Unclear Quote rdquorandomly assigned using randomly generated assignmentcards in sealed envelopesldquo Judged to be unclear

BlindingFalls


Low risk of bias in recall of falls No Falls identified at 4 monthly intervals by structured interview for activearm and by telephone for controls

Fiatarone 1997


Participants Setting community USAN = 34Sample frail older people (94 women)Age mean 82 (SD 1)Inclusion criteria community dwelling older people moderate to severe functional impairmentExclusion criteria none given

Interventions 1 High intensity progressive resistance training exercises in own home Two weeks of instructionand then weekly phone calls 11 different upper and lower limb exercises with arm and leg weights3 days per week for 16 weeks2 Control wait list control Weekly phone calls


Notes Abstract only

Risk of bias



Fiatarone 1997 (Continued)


Adequate sequence generation Unclear Method of randomisation not described


BlindingFalls


Low risk of bias in recall of falls Unclear Interval recall Falls identified weekly by phone call

Foss 2006


Participants Setting community Nottingham United KingdomN = 239Sample referred to ophthalmology outpatient clinic (100 women)Age mean 795 (range 70 to 92)Inclusion criteria over 70 years of age following successful cataract operation and with operablesecond cataractExclusion criteria having complex cataracts visual field defects or severe comorbid eye diseaseaffecting visual acuity memory problems preventing completion of questionnaires or reliablerecall of falls

Interventions 1 Small incision cataract surgery with insertion of intraocular lens under local anaesthetic2 Control waiting list


Notes

Risk of bias


Adequate sequence generation Yes Quote rdquolists prepared from random numbers in variably sized permutedblocks to maintain approximate equality in the size of the groupsldquo

Allocation concealment Yes Sequentially numbered opaque envelopes



Foss 2006 (Continued)

BlindingFalls


BlindingFractures


Low risk of bias in recall of falls Yes Prospective Falls recorded on daily diary Data collected by phone at 3and 9 months and by interview at 6 and 12 months

Gallagher 1996


Participants Setting community Victoria British Columbia CanadaN = 100Sample community dwelling volunteers (80 women)Age mean 746Inclusion criteria aged 60 and over fallen in previous 3 monthsExclusion criteria none described

Interventions 1 Two risk assessment interviews of 45 minutes each One counselling interview of 60 minutesshowing video and booklet and results of risk assessment2 Control baseline interview and follow up only No intervention


Notes

Risk of bias



Allocation concealment Unclear Method of randomisation not described

BlindingFalls


Low risk of bias in recall of falls Yes Calendar postcards completed and returned every two weeks for sixmonths Telephone follow up of reported falls



Gallagher 2001


Participants Setting presumed community Omaha USAN = 489Sample mailing lists used to contact women aged 65-77 years in Omaha and surrounding district(100 women)Age range 65-77 mean 71 (SD 4)Inclusion criteria 65 - 77 years not osteoporotic (femoral neck density in normal range for age)Exclusion criteria severe chronic illness primary hyperparathyroidism or active renal stone diseaseon certain medications in last 6 months eg bisphosphonates anticonvulsants estrogen fluoridethiazide diuretics

Interventions 1 Calcitriol (Rocaltrol) 025 mcg twice daily for 3 years2 HRTERT (conjugate estrogens (Premarin) 0625 mg daily + medroxyprogesterone (Provera)25 mg daily3 Calcitriol plus HRTERT as above4 Control placebo(ERT given to hysterectomised women N = 290 ie not given progestin)All groups advised to increase dietary calcium if daily intake lt 500 mgd and to decrease dietarycalcium if intake gt 1000 mgd


Notes

Risk of bias


Adequate sequence generation Unclear rdquoSimple randomisationldquo stratified on presence or absence of uterus Nofurther details

Allocation concealment Unclear Quote rdquorandomly assignedldquo No methods described

BlindingFalls


Low risk of bias in recall of falls No Falls retrospectively monitored by interview questionnaire at 6 weeks 12weeks and 6 monthly thereafter



Grant 2005

Methods RCT (multicentre) 2x2 factorial designLosses

Participants Setting United KingdomN = 5292Sample 21 centres in England and Scotland (85 women)Age mean 77 (SD 6)Inclusion criteria aged 70 and over recent previous osteoporotic fracture (defined as caused by afall)Exclusion criteria bed or chair bound prior to fracture abbreviated mental test score 6 or lesscancer likely to metastasise to bone within previous 10 years fracture associated with pre-existingbone abnormality known hypercalcaemia renal stone in last 10 years life expectancy lt 6 m knownto be leaving the UK taking gt 200 IU (5 mcg) vitamin D or gt 500 mg calcium supplements dailyhad fluoride calcitonin tibolone HRT selective estrogen receptor modulators or any vitamin Dmetabolite (such as calcitriol) in the last 5 years vitamin D by injection in preceding year

Interventions Two tablets daily with meals for two years Tablets delivered every four months by post Ran-domised to tablets containing a total of either1 800 IU (20 mcg) vitamin D3 plus placebo calcium2 800 IU vitamin D3 + 1000 mg calcium3 1000 mg elemental calcium (calcium carbonate) plus placebo vitamin D4 Double placebo

Outcomes 1 Number of people falling2 Number sustaining a fracture3 Number of people with adverse effects

Notes

Risk of bias


Adequate sequence generation Yes Computer-generated centralised randomisation stratified by centre

Allocation concealment Yes Centralised randomisation

BlindingFalls


BlindingFractures

Yes Fractures reported by participants who were blind to their group alloca-tion and identified from other sources (placebo-controlled trial)

Low risk of bias in recall of falls Unclear Interval recall Falls ascertained in 4 monthly postal questionnaire (rdquoHaveyou fallen during the last weekldquo) with telephone follow up if requiredalso from hospital and GP staff annotating notes



Gray-Donald 1995


Participants Setting community Quebec CanadaN = 50Subjects men and women recruited from those receiving long term home help services (71women)Age mean 775 (SD 8)Inclusion criteria aged over 60 requiring community services elevated risk of under-nutrition(excessive weight loss or BMI lt24 kgm2)Exclusion criteria alcoholic terminal illness

Interventions 1 12 week intervention of high energy nutrient dense supplements provided by dietitian Two235 ml cans per day (1045-1480 kj per can) for 12 weeks2 Control visits only (encouragement and suggestions about improving diets)


Notes

Risk of bias


Adequate sequence generation Unclear Method of randomisation not described Stratified by gender and nutri-tional risk criteria


BlindingFalls


Low risk of bias in recall of falls No Retrospectively monitored at 6 and 12 weeks

Green 2002


Participants Setting Bradford United KingdomN = 170Sample patients on hospital and community therapy stroke registers (44 women)Age mean 725 (SD 85) yearsInclusion criteria gt 50 years old stroke at least 1 year previously persisting stroke-related mobilityproblems



Green 2002 (Continued)

Exclusion criteria dementia severe comorbidity confined to bed physiotherapy treatment withinprevious 6 months

Interventions 1 Community physiotherapy programme at home or in outpatient rehabilitation centres Maxi-mum contact period usually 13 weeks with a minimum of three contacts per patient2 Control usual care


Notes

Risk of bias


Adequate sequence generation Yes rdquoRandom number tables and used four length permuted blocksldquo

Allocation concealment Yes Numbered sealed opaque envelopes

BlindingFalls


Low risk of bias in recall of falls Unclear Retrospective interval recall at 3 monthly assessments

Greenspan 2005

Methods RCT 2x2 factorial designLosses 36 of 373 (10)

Participants Setting community Boston USAN = 373Sample identified from newspaper advertisements targeted mailings presentations to seniorsgroups and physician referrals (100 women)Age mean 713 (SD 52)Inclusion criteria community-dwelling women including women with hysterectomy aged 65and olderExclusion criteria illness that could affect bone mineral metabolism current use of medicationsknown to alter bone mineral metabolism known contraindication to HRT use

Interventions 1 HRTERT plus placebo alendronate2 HRTERT plus alendronate3 Alendronate plus placebo HRTERT4 Placebo HRTERT plus placebo alendronateAll participants received calcium and vitamin D supplementation throughout the study



Greenspan 2005 (Continued)

(ERT given to hysterectomised women ie not given progestin)

Outcomes 1 Number of people fallingFalls a secondary outcome of study Other outcomes reported but not included in this review

Notes In the 2005 report the data presented are for all women receiving HRT This includes womenwho received HRT + alendronate Although there is no evidence of an interaction between theseagents which might plausibly affect falls this cannot be absolutely ruled out Therefore in thisreview we have taken a conservative approach and not used data the group who received HRT +alendronate

Risk of bias


Adequate sequence generation Yes Computer random number generation

Allocation concealment Yes Sequentially numbered opaque sealed envelopes

BlindingFalls


Low risk of bias in recall of falls No Interval recall but at six months and one year

Harwood 2004


Participants Setting Nottingham UKN = 150Sample women admitted to orthogeriatric rehabilitation ward within 7 days of surgery for hipfracture (100 women)Age mean 812 (range 67-92) yearsInclusion criteria recent surgery for hip fracture previous community residence previous inde-pendence in ADLExclusion criteria previously institutionalised disease or medication known to affect bonemetabolism lt 7 on 10 point mental state score

Interventions 1 Single injection of vitamin D2 (ergocalciferol) 300000 units2 Single injection of vitamin D2 (ergocalciferol) 300000 units plus oral calcium carbonate(calcichew) 1 tablet x 2 per day (1 g elemental calcium daily)3 Oral vitamin D3 + calcium carbonate (Calceos) 1 tablet x 2 per day (cholecalciferol 800unitsday + calcium 1 gday)



Harwood 2004 (Continued)

4 Control no treatment

Outcomes 1 Number of people falling2 Number sustaining a fracture3 Number of people with adverse effectsOther outcomes reported but not included in this review

Notes Recruited in hospital but meets the inclusion criteria as participants were all community-dwellingand intervention was designed to prevent falls in the community

Risk of bias


Adequate sequence generation Yes Randomised to four groups by computer generated random number lists

Allocation concealment Unclear Quote rdquousing sealed opaque envelopesldquo

BlindingFalls

No Falls reported by participants to researchers who were aware of their groupallocation

BlindingFractures

No Fractures reported by participants to researchers who were aware of theirgroup allocation

Low risk of bias in recall of falls No Falls not recorded in diaries Presume falls and fractures ascertained atdedicated clinic at 3 6 and 12 months

Harwood 2005


Participants Setting Nottingham UKN = 306Sample women referred to one of three consultant ophthalmologists (or to an optometrist-ledcataract clinic)Age median 785 (range 70 - 95) yearsInclusion criteria women aged gt 70 years with cataract no previous ocular surgeryExclusion criteria cataract not suitable for surgery by phacoemulsification severe refraction errorin 2nd eye visual field deficits severe co-morbid eye disease affecting visual acuity registrablepartially sighted as a result of cataract memory problems




Interventions 1 Expedited cataract surgery (target within 1 month)2 Routine waiting list for surgery (within 13 months) plus up-to-date spectacle prescription


Notes

Risk of bias


Adequate sequence generation Yes Random numbers in variably sized permuted blocks rdquoBlock randomisedconsecutively to groupsldquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationUnclear whether the assessors were aware of group allocation

BlindingFractures

Unclear Presume fractures reported by participants who were aware of their groupallocation Unclear whether the assessors were aware of group allocation

Low risk of bias in recall of falls Yes Prospective Falls recorded in diaries telephoned at 3 and 9 monthsinterviewed at 6 and 12 months for data

Hauer 2001


Participants Setting community GermanyN = 57Sample recruited at the end of ward rehabilitation from a geriatric hospital (100 women)Age mean 82 (SD 48) range 75-90 yearsInclusion criteria ge75 years fall(s) as reason for admission to hospital or recent history of injuriousfall leading to medical treatment residing within study communityExclusion criteria acute neurological impairment severe cardiovascular disease unstable chronicor terminal illness major depression severe cognitive impairment musculoskeletal impairmentpreventing participation in training regimen falls known to be due to a single identifiable diseaseeg stroke or hypoglycaemia



Hauer 2001 (Continued)

Interventions 1 Exercise group resistance training and progressive functional balance training x3 days per weekfor 12 weeks2 Control rdquomotor placeboldquo ie flexibility calisthenics ball games and memory tasks while seatedx3 days per week


Notes

Risk of bias


Adequate sequence generation Unclear Stratified randomisation


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Daily diaries collected every two weeks

Helbostad 2004


Participants Setting 6 local districts in Trondheim NorwayN = 77Sample volunteers recruited by announcement in local newspapers and invitations distributed bylocal health workers (81 women)Age mean 81 (SD 45)Inclusion criteria aged 75 and over one or more falls in last year using walking aid indoor oroutdoorExclusion criteria exercising one or more times weekly terminal illness cognitive impairment(MMSE lt22) stroke during previous 6 months geriatric assessment showed not able to tolerateexercise

Interventions 1 Combined training home visit by physical therapist for assessment group classes 5-8 people(individually tailored progressive resistance exercises functional balance training) 1 hour 2x perweek for 12 weeks + home exercises as below (2)2 Home training four non-progressive exercises (functional balance and strength exercises) 2xdaily for 12 weeks + 3 group meetings



Helbostad 2004 (Continued)


Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquorandomised into one of two exercise programsldquo

Allocation concealment Yes Randomised by independent research office using sealed envelopes

BlindingFalls

Yes Falls reported by participants Both groups received an exercise interven-tion Assessors blind to subjectsrsquo assignment

Low risk of bias in recall of falls Yes Monthly falls diary (pre-paid post card) telephone call if no response orfall reported

Hendriks 2008

Methods RCT with economic evaluationLosses 83 of 333 (25)

Participants Setting Maastricht The NetherlandsN = 333Sample people aged who have visited an AampE department or a GP because of a fall (70 women)Age mean 748 (SD 64) yearsInclusion criteria community-dwelling ge 65 years history of a fall requiring visit to AampE orGP living in Maastricht areaExclusion criteria not able to speak or understand Dutch not able to complete questionnaires orinterviews by telephone cognitive impairment (lt 4 on AMT4) long-term admission to hospitalor other institution (gt 4 weeks from date of inclusion) permanently bedridden fully dependenton a wheelchair

Interventions 1 Multifactorial intervention detailed assessment by geriatrician rehabilitation physician geri-atric nurse recommendations and indications for referral sent to participantsrsquo GPs GPs could thentake action if they agreed with the recommendations andor referrals Home assessment by OTrecommendations sent to participants and their GPs and direct referral to social or communityservices for provision of technical aids and adaptations or additional support2 Control usual care




Hendriks 2008 (Continued)

Notes

Risk of bias


Adequate sequence generation Unclear Quote rdquoRandomisation was achieved by means of computerised alterna-tive allocation and performed by an external agencyldquo

Allocation concealment Unclear Quote rdquoRandomisation was achieved by means of computerised alterna-tive allocation and performed by an external agencyldquo

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationQuote rdquoTo ensure blinding during data collection measurements byphone were contracted out to an independent call centre () whoseoperators were unaware of group allocationldquo

Low risk of bias in recall of falls Yes Quote rdquoParticipants recorded their falls continuously on a fall calendarduring twelve months after baseline They were contacted monthly bytelephone by an independent call centre (MEMIC) to report the fallsnoted on the calendarldquo

Hill 2000


Participants Setting community Staffordshire United KingdomN = 100Sample people referred to falls assessment clinic (73 women)Age mean 785 yearsInclusion criteria history of recurrent falls referred to falls clinicExclusion criteria cognitive impairment

Interventions 1 Daily exercise twice weekly supervised group balance exercise and individualised fall preventionadvice2 Control standard fall prevention advice

Outcomes 1 Rate of falls2 Number of people falling3 Number sustaining a fracture4 Number of people with adverse effects

Notes



Hill 2000 (Continued)

Risk of bias




BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationUnclear whether assessors collecting data did

Low risk of bias in recall of falls No Recall at end of study period (6 months)

Hogan 2001


Participants Setting community Calgary CanadaN = 163Sample high risk community dwelling men and women (71 women)Age mean 776 (SD 68)Inclusion criteria aged 65 and over fall in previous 3 months living in the community ambulatory(with or without aid) mentally intact (able to give consent)Exclusion criteria qualifying fall resulted in lower extremity fracture resulted from vigorous orhigh-risk activities because of syncope or acute stroke or while undergoing active treatment inhospital

Interventions 1 One in-home assessment by a geriatric specialist (doctor nurse physiotherapist or OT) lasting1-2 hours Intrinsic and environmental risk factors assessed Multidisciplinary case conference (20minutes) Recommendations sent to patients and patientsrsquo doctor for implementation Subjectsreferred to exercise class if problems with balance or gait and not already attending an exerciseprogramme Given instructions about exercises to do at home2 Control one home visit by recreational therapist


Notes

Risk of bias




Hogan 2001 (Continued)

Adequate sequence generation Yes Computer generated Stratified by number of falls in previous year 1 orgt1

Allocation concealment Unclear Sequence concealed in locked cabinet prior to randomisation

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationrdquoThe RA (research assistant) remained blinded throughout the study asto each subjectrsquos group assignmentldquo

BlindingFractures

Unclear Unclear if self-reported first Research assistant collecting data remainedblinded throughout the study as to each participantrsquos group assignment

Low risk of bias in recall of falls Unclear Falls recorded on monthly calenders (478 returned) Also retrospectiverecall at 3 6 months (at visit) and 12 months (by phone)

Hornbrook 1994

Methods RCT (cluster randomised by household)Losses 156 of 3182 (5) in the intervention group

Participants Setting community USAN = 3182 (N = 2509 households)Sample independently living members of HMO recruited by mail (38 women)Age mean 73 (SD 6)Inclusion criteria aged over 65 ambulatory living within 20 miles of investigation site consent-ingExclusion criteria blind deaf institutionalised housebound non-English speaking severely men-tally ill terminally ill unwilling to travel to research centre

Interventions 1 Home visit safety inspection (prior to randomisation) hazards booklet repair advice fallprevention classes (addressing environmental behavioural and physical risk factors) financial andtechnical assistance2 Control home visit safety inspection (prior to randomisation) hazards booklet


Notes

Risk of bias




Hornbrook 1994 (Continued)

Adequate sequence generation Unclear Quote rdquorandomly assignedldquo


BlindingFalls


BlindingFractures


Low risk of bias in recall of falls Yes Prospective Returned a postcard after each fall Also recorded falls onmonthly diaries and received quarterly mailtelephone contacts

Huang 2004


Participants Setting community Hsin-Chu County Northwest TaiwanN = 120Sample persons in registered households (46 women)Age mean 72 (SD 57)Inclusion criteria aged 65 and over community living cognitively intactExclusion criteria none stated

Interventions 1 3 home visits over 4 months (HV1 HV2 and HV3) by nurseHV1 risk assessment (medications and environmental hazards)HV2 two months later Standard fall prevention brochure plus individualised verbal teaching andbrochure relating to fall risk factors identified at HV1HV3 assessment and collection of falls data2 Control HV1 risk assessmentHV2 standard fall prevention brochureHV3 assessment and collection of falls data


Notes

Risk of bias




Huang 2004 (Continued)

Adequate sequence generation Unclear Method of randomisation not described Quote rdquoIn applying clustersampling half of the sample was randomly assigned to the experimentalgroup and the other half as the comparison groupldquo


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Self reported falls recorded on a calender in a Falls RecordChecklist for the two months after the intervention visit

Huang 2005


Participants Setting hospital northern TaiwanN = 141Sample people in hospital with a fall-related hip fracture (69 women)Age mean 77 (SD 76) yearsInclusion criteria in hospital with hip fracture resulting from a fall aged 65 and over dischargedwithin medical centre catchment areaExclusion criteria cognitively impaired too ill (comorbidities unable to communicate or inintensive care unit)

Interventions 1 Discharge planning intervention by masters-level gerontological nurse from hospital admissionuntil 3 month after discharge (first visit within 48 hours of admission seen every 48 hours whilein hospital one home visit 3-7 days after discharge available by phone 8am - 8pm seven days aweek phoned participant or care-giver once a week) Nurse created individualised discharge planand facilitated set up of home care services etc Participants provided with brochures on self-carefor hip fracture patients and fall prevention (environmental safety and medication issues) Nurseprovided direct care and education on correct use of assistive devices and assessed rehabilitationneeds Collaborated with physicians to modify therapies2 Control usual discharge planning also by nurses but not specialists No brochures writtendischarge summaries home visits phone calls


Notes Majority were community-dwelling as states rdquothe majority of older people with hip fracture whoare discharged from hospital are at homeldquo Intervention included a home visit 91 living withfamilyrdquo




Risk of bias


Adequate sequence generation Yes Randomly assigned using a computer generated table


BlindingFalls

No Falls reported by participants who were aware of their group allocationResearch assistant did assigning to groups and assessments (not blind)

Low risk of bias in recall of falls Unclear Falls data collected using falls diary Appear to have been interviewed at2 weeks and 3 months No mention of diaries being returned by post

Jitapunkul 1998


Participants Setting community ThailandN = 160Sample community dwelling men and women recruited from a sample for a previous study (66women)Age mean 756 (SD 58)Inclusion criteria aged 70 and over living at homeExclusion criteria none stated

Interventions 1 Home visit from non health professional with structured questionnaire 3 monthly visits for3 years Referred to nursegeriatrician (community based) if Barthel ADL index andor ChulaADL index declined 2 or more points or subject fell more than once during previous 3 monthsNursegeriatrician would visit assess educate prescribe drugsaids provide rehabilitation pro-gramme make referrals to social services and other agencies2 Control no intervention Visit at the end of 3 years


Notes

Risk of bias





Jitapunkul 1998 (Continued)


BlindingFalls

No Falls reported by participants who were aware of their group allocationPossible bias Intervention group provided falls data every three monthsfor three years but control group received no other visits in which fallsdata were collected

Low risk of bias in recall of falls No Retrospective Falls data for preceding three months collected at exit as-sessment at 3 years

Kenny 2001


Participants Setting Cardiovascular Investigation Unit Newcastle UKN = 175Sample individuals presenting at AampE with non-accidental fall (60 women)Age mean 73 (SD 10)Inclusion criteria aged 50 and over history of a non-accidental fall diagnosed as having cardioin-hibitory CSH by carotid sinus massageExclusion criteria cognitive impairment medical explanation of fall within 10 days of presenta-tion an accidental fall blind lived gt15 miles from AampE had contraindication to CSM receivingmedications known to cause a hypersensitive response to CSM

Interventions 1 Pacemaker (rate drop response physiologic dual-chamber pacemaker Thera RDR MedtronicMinneapolis Minnesota)2 Control no pacemaker

Outcomes 1 Rate of falls2 Number of people falling3 Number sustaining a fractureOther outcomes reported but not included in this review

Notes Out of 3384 AampE attendees with non-accidental falls 257 were diagnosed as having carotid sinushypersensitivity 175 of these were randomised ie 5 of non-accidental falls

Risk of bias


Adequate sequence generation Unclear Quote ldquo Randomisedby block randomisation in blocks of eightrdquoMethod of sequence generation not described



Kenny 2001 (Continued)


BlindingFalls


BlindingFractures

Unclear Presume fractures reported by participants who were aware of their groupallocation

Low risk of bias in recall of falls Yes Prospective Falls recorded daily on self-completion diary cards whichwere returned at the end of each week for one year

Kingston 2001


Participants Setting AampE Staffordshire UKN = 109Sample community-dwelling women attending AampE with a fallAge mean 719Inclusion criteria female aged 65-79 history of a fall discharged directly to own homeExclusion criteria admitted from AampE to hospital or any form of institutional care

Interventions 1 Rapid Health Visitor intervention within 5 working days of index fall pain control and medi-cation how to get up after a fall education about risk factors (environmental and drugs alcoholetc) advice on diet and exercise to strengthen muscles and joints Also care managed on individualbasis for 12 months post index fall2 Control usual post fall treatment ie letter to GP from AampE detailing the clinical event anyinterventions carried out in hospital and recommendations about follow up

Outcomes 1 Number of people fallingFalls not primary outcome of study Other outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Unclear Quote ldquorandomly allocatedrdquo

Allocation concealment Unclear Quote ldquorandomly allocatedrdquo Insufficient information to permit judg-ment



Kingston 2001 (Continued)

BlindingFalls


Low risk of bias in recall of falls No Quote ldquoFalls were recorded at week twelve assessmentrdquo (information fromauthor)

Korpelainen 2006


Participants Setting community Oulu FinlandN = 160Sample birth cohort of womenAge mean 73 (SD 12) yearsInclusion criteria hip BMD gt 2 less than the reference valueExclusion criteria ldquomedical reasonsrdquo use of a walking aid other than a stick bilateral total hipjoint replacement unstable chronic illness malignancy medication known to affect bone densitysevere cognitive impairment involvement in other interventions

Interventions 1 Supervised exercise programme (physiotherapist led) Mixed home and supervised group pro-gramme plus twice yearly seminars on nutrition health medical treatment and fall prevention2 Control twice yearly seminars on nutrition health medical treatment and fall prevention


Notes

Risk of bias


Adequate sequence generation Yes Quote ldquoEach participant received sequentially according to the originalidentification numbers the next random assignment in the computerlistrdquo

Allocation concealment Yes The randomisation was ldquoprovided by a technical assistant not involved inthe conduction of the trialrdquo

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessors blind to allocation



Korpelainen 2006 (Continued)

BlindingFractures

Unclear Fractures reported by participants who were aware of their group alloca-tion Assessors blind to allocation

Low risk of bias in recall of falls No Three monthly retrospective recall

Lannin 2007


Participants Setting community Sydney AustraliaN = 10Sample patients admitted to a rehabilitation facility and referred to OT (80 women)Age mean 81 (SD 7)Inclusion criteria mild or no cognitive impairment community dwelling (non institutional)aged 65 or older no medical contraindications that would require strict adherence to equipmentrecommendationsExclusion criteria none

Interventions 1 Best practice occupational therapy home visit intervention2 Control standard practice in-hospital assessment and education


Notes Pilot study

Risk of bias


Adequate sequence generation Yes Allocation schedule computer generated

Allocation concealment Yes Quote ldquoConcealed in opaque consecutively numbered envelopes by aperson not involved in the studyrdquo

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessor blind to group allocation

Low risk of bias in recall of falls Unclear Interval recall Falls ascertained by assessor at home visit at 2 weeks andone two and three months after discharge



Latham 2003

Methods RCT (factorial design)Losses none described

Participants Setting Five hospitals in Auckland New Zealand and Sydney AustraliaN = 243Sample frail older people recently discharged from hospital (53 women)Age mean 79 yearsInclusion criteria aged 65 and over considered frail (one or more health problems eg depen-dency in an ADL prolonged bed rest impaired mobility or a recent fall) no clear indication orcontraindication to either of the study treatmentsExclusion criteria poor prognosis and unlikely to survive 6 months severe cognitive impairmentphysical limitations that would limit adherence to exercise programme unstable cardiac statuslarge ulcers around ankles that would preclude use of ankle weights living outside hospitalsrsquogeographical zone not fluent in English

Interventions 1 Exercise quadriceps exercises using adjustable ankle cuff weights 3 x per week for 10 weeksFirst 2 sessions in hospital remainder at home Monitored weekly by physiotherapist alternatinghome visit with telephone calls2 Exercise control frequency matched telephone calls and home visits from research physicaltherapist including general enquiry about recovery general advice on problems support3 Vitamin D single oral dose of six 125 mg calciferol (300000 IU)4 Vitamin D control placebo tablets


Notes Detailed description of exercise regimen given in paper

Risk of bias


Adequate sequence generation Yes Study biostatistician generated random sequence Block randomisationtechnique

Allocation concealment Yes Computerised centralised randomisation scheme

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessors blind to allocation group

Low risk of bias in recall of falls Yes Prospective Falls recorded in fall diary with weekly reminders for first 10weeks Nurses examined fall diaries and sought further details about eachfall at 3 and 6 month visits Reminder phone call between visits



Li 2005


Participants Setting community Legacy Health System Portland Oregon USAN = 256Sample enrolled in health maintenance organisation recruited from (70 women)Age mean 775 (SD 5) range 70 - 92 yearsInclusion criteria age ge 70 physician clearance to participate inactive (no moderate to strenuousactivity in last 3 months) walks independentlyExclusion criteria chronic medical problems that would limit participation cognitive impairment

Interventions 1 Exercise intervention Tai Chi 1 hour x3 per week for 26 weeks2 Control low level stretching 1 hour x3 per week for 26 weeks


Notes

Risk of bias


Adequate sequence generation Yes Computer generated random numbers


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Daily fall calendar

Lightbody 2002

Methods RCT Cluster randomised Randomisation of 16 treating physicians matched in 4 groups of 42 control and 2 intervention in each group enrolled subjects assigned to same group as theirphysicianLosses 10 of 301 (3)

Participants Setting hospital Liverpool UKN = 348Subjects consecutive patients attending AampE with a fall (74 women)Age median 75 IQR 70-81Inclusion criteria aged gt 65 patients attending AampE with a fall



Lightbody 2002 (Continued)

Exclusion criteria admitted to hospital as result of index fall living in institutional care refusedor unable to consent lived out of the area

Interventions 1 Multifactorial assessment by falls nurse at one home visit (medication ECG blood pressurecognition visual acuity hearing vestibular dysfunction balance mobility feet and footwear en-vironmental assessment) Referral for specialist assessment or further action (relatives communitytherapy services social services primary care team No referrals to day hospital or hospital outpa-tients) Advice and education about home safety and simple modifications eg mat removal2 Control usual care


Notes Assessment of risk factors medication ECG blood pressure cognition visual acuity hearingvestibular dysfunction balance mobility feet and footwear Environmental assessmentFalls reported in diary and by questionnaire different

Risk of bias


Adequate sequence generation Unclear Insufficient information to permit judgment


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Falls injury and treatment recorded in diary Postal ques-tionnaire at 6 months to collect data GP records and hospital databasessearched

Lin 2007


Participants Setting community TaiwanN = 150Sample residents of rural agricultural area ( women not known)Age mean 765 yearsInclusion criteria medical attention for a fall in previous 4 weeks ge 65 yearsExclusion criteria none described



Lin 2007 (Continued)

Interventions 1 Home-based exercise training2 Home safety assessment and modification3 Control ldquoeducationrdquo 1 social visit 30-40 minutes every 2 weeks for 4 months with fall preven-tion pamphlets provided


Notes

Risk of bias


Adequate sequence generation Unclear Block randomised Insufficient information to permit judgment


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Reported falls by telephone or postcard when they occurredPhoned every 2 weeks to ascertain occurrence of falls

Liu-Ambrose 2004


Participants Setting community British Colombia CanadaN = 104Sample all women residents of greater Vancouver aged 75-85 with osteoporosis or osteopeniadiagnosed at British Colombia Womenrsquos Hospital and Health Centre Also list of individualswith low bone mass provided by Osteoporosis Society of Canada British Colombia section andnewspaper radio and poster advertisements (100 women)Age mean 79 (SD 3) range 75-85Inclusion criteria women aged 75-85 osteoporosis or osteopenia (BMD total hip or spine T scoreat least 1 SD below young normal sex matched area BMD of the Lunar reference database)Exclusion criteria living in care facility non-Caucasian race regularly exercising 2 x weekly ormore history of illness or a condition affecting balance (stroke Parkinsonrsquos disease) unable tosafely participate in exercise programme MMSE 23 or less



Liu-Ambrose 2004 (Continued)

Interventions 1 High intensity resistance training 50 minutes 2x weekly for 25 weeks using Keiser PressurizedAir system and free weights Instructorparticipant ratio 122 Agility training 50 minutes 2x weekly for 25 weeks Training (ball games relay races dance move-ments obstacle courses wearing hip protectors) designed to challenge hand-eye and foot-eye co-ordination and dynamic standing and leaning balance and reaction time Instructorparticipantratio 133 Control sham exercises 50 minutes 2x weekly for 25 weeks Stretching deep breathing relax-ation general posture Instructorparticipant ratio 14


Notes

Risk of bias


Adequate sequence generation Unclear Method of randomisation not described but stratified by baseline perfor-mance in postural sway


BlindingFalls


Low risk of bias in recall of falls Yes Prospective ldquoFalls documented using monthly falls calendarsrdquo

Lord 1995

Methods RCT Pre-randomisation prior to consent from a schedule of participants in a previous studyLosses 19 of 194 (10) all from intervention group

Participants Setting community AustraliaN = 194Sample women recruited from a schedule from a previous epidemiologic study Fitness level notdefinedAge mean 716 (SD 54) range 60-85Inclusion criteria living independently in the communityExclusion criteria unable to use English



Lord 1995 (Continued)

Interventions 1 Twice weekly exercise classes (warm-up conditioning stretching relaxation) lasting 1 hourover a 12 month period2 Control no intervention


Notes

Risk of bias


Adequate sequence generation Unclear Quote ldquorandomly assignedrdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessors not blind to treatment status

Low risk of bias in recall of falls Unclear Interval recall Fall ascertainment questionnaires sent out every 2 monthsTelephone call if questionnaire not returned

Lord 2003

Methods RCT Cluster randomised by village Stratified by accommodation (self care or intermediate care)and by cluster size (lt75 or at least 75 residents)Losses 47 of 551 (9)

Participants Setting retirement villages Sydney AustraliaN = 551 (N = 20 clusters)Sample recruited from self-care apartment villages (78) and intermediate-care hostels (22)(86 women)Age mean 795 (SD 64) range 62-95Inclusion criteria resident in one of 20 retirement villagesExclusion criteria MMSE lt 20 already attending exercise classes of equivalent intensity medicalconditions that precluded participation as determined by nurse or physician (neuromuscularskeletal cardiovascular) in hospital or away at recruitment time

Interventions 1 Group exercise classes for 1 hour 2x weekly for 1 year Designed to improve strength speedcoordination balance and gait and to improve performance in ADLs (turning and reachingrising from chair stair climbing standing and walking balance) 35-40 minute conditioningperiod Aerobic exercises strengthening exercises activities for balance and hand-eye and foot-eyecoordination and flexibility (mostly weight bearing)




2 Control seated flexibility and relaxation activities by yoga instructors (4 village sites) 1 hour2x weekly for 1 year3 Control no group activity

Outcomes 1 Rate of falls

Notes Detailed description of exercise interventions in Lord 2004

Risk of bias




BlindingFalls


Low risk of bias in recall of falls Unclear Falls ascertained by monthly questionnaires and follow-up phone calls orhome visit for none responders Nurses recorded falls in falls record bookin intermediate-care hostels

Lord 2005


Participants Setting community Sydney AustraliaN = 620Sample health insurance membership database (66 women)Age mean 804 (SD 45) yearsInclusion criteria low score on PPA test community dwelling ge 75 yearsExclusion criteria minimal English language skills blind PD cognitive impairment

Interventions 1 Extensive intervention comprising individualised exercise intervention (2x per week for 12months) visual intervention peripheral sensation counselling intervention2 Minimal intervention Participants received a report outlining their falls risk a profile of theirtest results and specific recommendations on preventing falls based on their test performances3 Control no intervention (received minimal intervention after 12 month follow up)





Notes

Risk of bias


Adequate sequence generation Yes Quote ldquorandomised in matched blocks N = 20 using concealed alloca-tion (drawing lots)rdquo

Allocation concealment Yes Quote ldquoconcealed allocationrdquo

BlindingFalls


Low risk of bias in recall of falls Yes Prospective Monthly fall calendars Telephoned at end of month if notreturned

Luukinen 2007


Participants Setting community Oulu FinlandN = 486Sample identified from population and geriatric registers of Oulu (79 women)Age mean 88 (SD 3)Inclusion criteria age ge 85 home dwelling ge 1 risk factor for falling (ge2 falls in previous yearloneliness poor self-rated health poor visual acuityhearing depression poor cognition impairedbalance chair rise slow walking speed difficulty with at least 1 ADL able to walk outdoors upor down stairs)Exclusion criteria none described

Interventions 1 Intervention plans developed by OT and physiotherapist at home visit based on nursersquos assess-ment pre-randomisation Feasibility of plan assessed by GP Plan included home exercise or groupexercise walking exercises self-care exercises (duration and frequency not described) Interven-tions carried out by OT andor physiotherapist2 Control asked to visit GP without written intervention form


Notes

Risk of bias



Luukinen 2007 (Continued)


Adequate sequence generation Yes Quote ldquoRandomization was done by the study statistician using a randomnumbers tablerdquo


BlindingFalls

Yes Falls reported by participants who might not have been aware of theirgroup allocation ldquoFalls recorded by a research nurse unaware of ran-domisation or the interventionrdquo

Low risk of bias in recall of falls No Interval recall Quote ldquoFalls recorded every second month by telephoneby a research nurse unaware of randomisation or the interventionrdquo

Mahoney 2007

Methods RCTLosses 5 of 349 (1) but all included in analysis

Participants Setting community USAN = 349Sample recruited from seniors centres meal sites senior apartment buildings other senior con-gregate sites by referral from caseworkers and healthcare providers (79 women)Age mean 80 (SD 75)Inclusion criteria aged 65 and over living independently 2 or more falls in previous year or 1injurious fall in previous 2 years or gait and balance problemsExclusion criteria unable to give informed consent and no related caregiver in hospice or assisted-living facility expected to move away from area

Interventions 1 Fall risk assessment by nurse or physiotherapist (two home visits) followed by recommenda-tions and referrals to primary physician physiotherapist OT ophthalmologist podiatrist etcAll participants given exercise plan for long-term exercise (walking programme standing balanceexercises in group setting etc) monthly exercise calendar and 11 monthly phone calls to promoteadherence to exercises and other recommendations2 Control one in-home assessment by OT ldquolimited to home safety recommendations and adviceto see their doctor about fallsrdquo


Notes

Risk of bias




Mahoney 2007 (Continued)

Adequate sequence generation Yes Randomised using computer-generated randomisation table

Allocation concealment Unclear Sealed envelopes used but no mention of numbering or how they wereused

BlindingFalls


Low risk of bias in recall of falls Yes Falls ascertained using monthly calendars telephone call if calendar notreturned or if fall reported

McKiernan 2005


Participants Setting community Wisconsin USAN = 113Sample (60 women)Age mean 742 range 65-96Inclusion criteria aged ge 65 years community dwelling ge1 falls in previous year independentlyambulatoryExclusion criteria not capable of applying Yaktrax walker correctly or discerning correct outdoorconditions to wear them

Interventions 1 Yaktrax walker (netting applied over usual footwear with wire coils to increase grip in winteroutdoor conditions)2 Control usual winter footwear


Notes

Risk of bias


Adequate sequence generation Unclear Quote ldquorandomizedrdquo


BlindingFalls

No Falls reported by participants who were aware of their group allocation20 of control group had also used this or a similar intervention becausethey were not blinded This might have influenced the outcome



McKiernan 2005 (Continued)

Low risk of bias in recall of falls Yes Prospective Fall diary returned by post

McMurdo 1997

Methods RCTLosses 26 of 118 (22) over 2 years

Participants Setting community Dundee United KingdomN = 118Sample community dwelling post menopausal women recruited by advertisementAge mean 645 range 60-73Exclusion criteria conditions or drug treatment likely to affect bone

Interventions 1 Exercise programme of weight bearing exercise to music 45 minutes 3 x weekly 30 weeks peryear over 2 years plus 1000 mg calcium carbonate daily2 Control 1000 mg calcium carbonate daily


Notes

Risk of bias




BlindingFalls


BlindingFractures

Unclear No description about ascertainment

Low risk of bias in recall of falls Unclear No description about ascertainment



Means 2005

Methods RCTLosses 100 (for falls data) of 338 (30)

Participants Setting community Arkansaw USAN = 338Sample from 17 senior citizenrsquos centres (57 women)Age mean 735 yearsInclusion criteria aged ge 65 years able to walk at least 30 feet without assistance from othersable to follow instructions and give consentExclusion criteria resident in a nursing home acute medical problems cognitive impairment

Interventions 1 Balance rehabilitation intervention Active stretching postural control endurance walking andrepetitive muscle coordination exercises Group sessions 90 minutes x3 per week for 6 weeks2 Control group seminars on non health-related topics of interest to senior citizens Same timeand frequency as intervention group


Notes

Risk of bias


Adequate sequence generation Yes Randomised by coin flip


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Recorded on pre-printed postcards weekly with telephonecalls to non correspondents to optimise compliance

Meredith 2002


Participants Setting community New York and Los Angeles USAN = 317Sample participants enrolled from home health care agencies client lists if agency office agreed toparticipate (75 women)Age mean 80 (SD 8)



Meredith 2002 (Continued)

Inclusion criteria Medicare patients aged 65 and older registered with home health care officesin defined period for medical or surgical services having one of four study medication problemshaving an identifiable physician expected home health care for at least 4 weeksExclusion criteria not expected to survive through follow up unable to understand spoken Englishresident in an unsafe area that requires an escort for visits

Interventions 1 Medication review by pharmacist and participantrsquos nurse based on reported problems (includingfalls) relating to medication use Targetted therapeutic duplication cardiovascular psychotropicand NSAID use Plan to reduce medication problem presented to physician in person by nurseor pharmacist Nurse assisted participant with the medication changes and monitored effect2 Control usual care which might include review of medications and adverse effects if relevant


Notes

Risk of bias


Adequate sequence generation Yes Assigment generated by computer random number generator (SAS v610) Balanced block randomisation stratified by the two areas

Allocation concealment Unclear Randomised off site but insufficient information to permit judgment

BlindingFalls


Low risk of bias in recall of falls No No description of how falls ascertained presumably retrospectively atfollow up interview

Morgan 2004


Participants Setting community and assisted-living facilities Florida USAN = 294Sample men and women recruited from Miami Department of Veterans Affairs Medical Centre9 assisted-living facilities private physical therapy clinic (71 women)Age mean 805 (SD 75)Inclusion criteria aged 60 and over hospital admission or bedrest for 2 or more days in previousmonth



Morgan 2004 (Continued)

Exclusion criteria medical conditions precluding exercise programme (angina severe osteoporosisetc) MMSE lt23 (unable to follow instructions) using oxygen therapy at home planned inpa-tient treatment or evaluation in 2 months following recruitment requiring human assistancewheelchair or artificial limbs to walk

Interventions 1 Low-intensity group exercise seated and standing exercises to improve muscle strength jointflexibility balance and gait 5 people per group 45 minutes 3 x per week for 8 weeks2 Control usual activities


Notes SAFE-GRIP (Study to Assess Falls among Elderly Geriatric Rehabilitation Intensive Program)

Risk of bias


Adequate sequence generation Unclear Randomisation stratified by sex age (lt75 and 75 and over) falls historyin previous month (fallno fall) Method of randomisation not described


BlindingFalls


Low risk of bias in recall of falls Yes Prospective Pre-dated postcard diaries returned every 2 weeks

Newbury 2001


Participants Setting community Adelaide AustraliaN = 100Sample every 20th name in an age-sex register of community dwelling patients registered with 6general practices (63 women)Age range 75 - 91 years median age in intervention group 785 control group 80 yearsInclusion criteria aged 75 and over living independently in the communityExclusion criteria none

Interventions 1 Health assessment of people aged 75 years or older by nurse (75+HA) Problems identified werecounted and reported to patientrsquos GP No reminders or other intervention for 12 months2 No 75+HA until 12 months



Newbury 2001 (Continued)


Notes 75+HA introduced in Australia November 1999 as part of Enhanced Primary Care packageSimilar to ldquohealth checkrdquo for patients in this age group in the United Kingdom

Risk of bias


Adequate sequence generation Yes Randomisation by random numbers

Allocation concealment Yes Sequentially numbered sealed envelopes

BlindingFalls


Low risk of bias in recall of falls No Falls identified retrospectively at follow up visit at 1 year

Nikolaus 2003


Participants Setting enrolled in hospital but community based intervention GermanyN = 360Sample frail ldquoolder peoplerdquo admitted to a geriatric clinic who normally lived at home (733female)Age mean 815 (SD 64)Inclusion criteria lived at home before admission and able to be discharged home with at least twochronic conditions (eg osteoarthritis or chronic cardiac failure stroke hip fracture parkinsonismchronic pain urinary incontinence malnutrition) or functional decline (unable to reach normalrange on at least one assessment test of ADL or mobility)Exclusion criteria terminal illness severe cognitive decline living gt15 km from clinic

Interventions 1 Comprehensive geriatric assessment + at least 2 home visits (from interdisciplinary homeintervention team (HIT) One home visit prior to discharge to identify home hazards and prescribetechnical aids if necessary At least one more visit (mean 26 range 1-8) to inform about possiblefall risks in home advice on changes to home environment facilitate changes and teach use oftechnical and mobility aids2 Control comprehensive geriatric assessment + recommendations alone No home visit untilfinal assessment at one year Usual post discharge management by GPs



Nikolaus 2003 (Continued)


Notes Home intervention team consisted of 3 nurses physiotherapist occupational therapist socialworker and secretary Usually two members at first home visit (OT + nurse or OT + physiotherapistdepending on anticipated needs and functional limitations)Methods paper described a third arm receiving usual hospital and home care

Risk of bias


Adequate sequence generation Yes Quote ldquosealed envelopes containing group assignments using a randomnumber sequencerdquo

Allocation concealment Unclear Quote ldquosealed envelopes containing group assignmentsrdquo

BlindingFalls


Low risk of bias in recall of falls Yes Falls recorded in falls diary and by monthly telephone calls

Nitz 2004


Participants Setting community Queensland AustraliaN = 73Sample volunteers recruited through newspaper adverts fliers sent to medical practitioners seniorsgroups and physiotherapists in local community (92 women)Age mean 758 (SD 78)Inclusion criteria aged over 60 living independently in the community at least 1 fall in previousyearExclusion criteria unstable cardiac condition living too far from exercise class site unable toguarantee regular attendance

Interventions 1 Balance training in small groups using workstation (circuit training) format 1 hour per weekfor 10 weeks Up to 6 people per group with physiotherapist instructor2 Control gentle exercise and stretching 1 hour per week for 10 weeks




Nitz 2004 (Continued)

Notes

Risk of bias




BlindingFalls


Low risk of bias in recall of falls Yes Falls ascertained by marked calendar returned monthly

Pardessus 2002


Participants Setting recruited in hospital community dwelling FranceN = 60Sample individuals hospitalised for a fallAge mean 832 (SD 77)Inclusion criteria aged 65 and over hospitalised for falling able to return home able to giveconsentExclusion criteria cognitive impairment (MMSE lt24) falls due to cardiac neurologic vascularor therapeutic problems without a phone lived gt 30 km from hospital

Interventions 1 Comprehensive 2 hour home visit prior to discharge with rsquophysical medicine and rehabilitationdoctorrsquo and OT Assessment of ADLs IADLs transfers mobility inside and outside use of stairsEnvironmental hazards identified and modified where possible If not advice given Discussionof social support Referrals for social assistance2 Control usual care


Notes

Risk of bias




Pardessus 2002 (Continued)

Adequate sequence generation Yes Randomised using random numbers table


BlindingFalls


Low risk of bias in recall of falls Unclear Interval recall but short interval Falls identified by monthly telephonecalls

Pereira 1998

Methods RCT in 1982-85 Reporting 10 year follow upLosses 31 of 229 (14)

Participants Setting community Pittsburgh USAN = 229 randomised 198 available for 10 year follow upSample healthy post-menopausal women (volunteers)Age at randomisation mean 57 at follow up mean 70 (SD 4)Inclusion criteria 1 year post menopause aged 50 and 65Exclusion criteria on HRT unable to walk

Interventions 1 8 week training period with organised group walking scheme 2 x weekly Also encouraged towalk once weekly on their own Building up to 7 miles per week total2 Control no intervention


Notes

Risk of bias




BlindingFalls


Low risk of bias in recall of falls No Falls in the previous 12 months ascertained by telephone interview



Pfeifer 2000


Participants Setting community GermanyN = 148Sample healthy ambulatory community living women recruited through advertisementAge 70 years or olderInclusion criterion 25-hydroxycholecalciferol serum level below 50 nmollitreExclusion criteria hypercalcaemia primary hyperparathyroidism osteoporotic extremity fracturetreatment with bisphosphonate calcitonin vitamin D or metabolites oestrogen tamoxifen inpast 6 months fluoride in last 2 years anticonvulsants or medications possibly interfering withpostural stability or balance intolerance to vitamin D or calcium chronic renal failure drugalcohol caffeine or nicotine abuse diabetes mellitus holiday at different latitude

Interventions An 8 week supplementation at the end of winter1 400 IU vitamin D plus 600 mg elemental calcium (calcium carbonate)2 Control 600 mg calcium carbonate


Notes

Risk of bias




BlindingFalls

Unclear Falls reported by participants who were unlikely to be aware of their groupallocation although the study was not placebo controlled Blinding ofassessor not described

Low risk of bias in recall of falls No Retrospective Falls and fractures monitored retrospectively by question-naire at 1 year



Pit 2007

Methods RCT Cluster randomised by general practiceLosses one GP and 190 of 849 (22) participants

Participants Setting general practices in Hunter Region New South Wales AustraliaN = 849 participants (17 practices 23 GPs)Sample 59 womenAge 65 and over No distribution givenInclusion criteria GPs based at their current practice for at least 12 months working 10 or morehours per week member of a randomly selected network of practices Patients aged 65 and overliving in the communityExclusion criterion confused patients not accompanied by a caregiver

Interventions 1 GPs education (academic detailing (x2 visits from pharmacist) provision of prescribing in-formation and feedback) completion of medication review checklist financial rewards Patientscompleted medication risk assessment form2 Control GPs no academic detailing but received feedback on number of medication reviewscompleted and medication risk factors Patients completed medication risk assessment form butnot passed on to GP for action


Notes

Risk of bias


Adequate sequence generation Yes Assignment undertaken ldquousing computer-generated random number al-location in SAS softwarerdquo

Allocation concealment Yes Randomisation carried out by off-site statistician

BlindingFalls

Yes Falls reported by participants who were unaware of their group allocationData collectors also blind to allocation

Low risk of bias in recall of falls No Retrospecitive interval recall Falls ascertained by phone at 4 and 12months

Porthouse 2005

Methods RCT (multicentre)Losses 312 of 3314 (9)



Porthouse 2005 (Continued)

Participants Setting community United KingdomN = 3314Sample community-dwelling women registered with 107 general practices in EnglandAge mean 769 (SD 51)Inclusion criteria aged 70 and over female community-dwelling one or more risk factors forfracture (prior fracture body weight 58 kg or less smoker family history of hip fracture poor orfair health)Exclusion criteria cognitive impairment life expectancy lt 6 months unable to give writtenconsent taking more than 500 mg calcium supplementation per day past history of kidney orbladder stones renal failure or hypercalcaemia

Interventions 1 Oral vitamin D3 800 IU (Calcichew D3 Forte) + oral 1000 mg calcium (calcium carbonate)daily for 6 months plus session with practice nurse life-style advice on how to reduce risk offracture + leaflet on dietary sources of vitamin D2 Control sent same leaflet as intervention group received

Outcomes 1 Rate of falls2 Number of people falling3 Number sustaining a fractureFalls are a secondary outcome in this study Other outcomes reported but not included in thisreview

Notes

Risk of bias


Adequate sequence generation Yes Randomised (stratified by GP practice) by computer Initially 21 ratioin favour of the control group to achieve most statistical power withinbudget Changed to 11 towards end of study after re-analysis of trialrsquoscost profile

Allocation concealment Yes Quote ldquoRandomised at the York Trials Unit by an independent personwho had no knowledge of the baseline characteristics of participantsrdquo

BlindingFalls


Low risk of bias in recall of falls No Retrospective Falls reported in six monthly postal questionnaires



Prince 2008


Participants Setting Perth AustraliaN = 302Sample women attending AampE receiving home nursing management of falls electoral roleAge mean 772 (SD 36)Inclusion criteria aged 70 - 90 years history of falling in last 12 months plasma 25OHD lt 24ngmLExclusion criteria current consumption of vitamin D or bone or mineral active agents other thancalcium BMD z score at total hip site lt -20 medical conditions or disorders affecting bonemetabolism fracture in last 6 months MMSE lt 24 neurological conditions affecting balance egstroke or Parkinsonrsquos disease

Interventions 1 1000 IUd ergocalciferol (vitamin D2) with evening meal + 1000 mgd calcium citrate (250mgtablets x2 with breakfast and evening meal) for 1 year2 Control placebo + 1000 mgd calcium citrate (250 mg tablets x2 with breakfast and eveningmeal) for 1 year

Outcomes 1 Number of people falling2 Number of people with adverse effects

Notes

Risk of bias


Adequate sequence generation Yes Used random number generator with block size of 10 to randomise in aratio of 11

Allocation concealment Yes Randomisation schedule generated by ldquoindependent research scientistrdquoSchedule kept in pharmacy department of hospital where bottles werelabelled and dispensed to participants

BlindingFalls


Low risk of bias in recall of falls No Retrospective Interviewed by study staff every 6 weeks by phone or at aclinic visit



Reinsch 1992

Methods RCT 2x2 factorial design Cluster randomised by senior centre rather than by individual partici-pantLosses 46 of 230 (20)

Participants Setting community Los Angeles County and Orange County California USAN = 230Sample men and women recruited from 16 senior centres ( women)Age mean 742 (SD 60)Inclusion criteria aged over 60Exclusion criteria none listed

Interventions 1 ldquoStand upstep uprdquo exercise programme with preliminary stretching exercise 1 hour x 3 daysper week for 1 year2 Cognitive-behavioural intervention consisting of relaxation training reaction time training andhealth and safety curriculum 1 hour x 1 day per week for 1 year3 Exercise (2 meetings per week) and cognitive intervention (x 1 meeting per week) for 1 year4 Discussion control group 1 hour x 1 day per week for 1 year


Notes MacRae paper includes a subset of results for only two arms of the study in Los Angeles countyonly

Risk of bias


Adequate sequence generation Unclear Quote ldquorandomly assigned to treatmentsrdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationBlinding of research assistant not described

Low risk of bias in recall of falls Yes Prospective Monthly diaries plus weekly phone calls or visits

Resnick 2002


Participants Setting community Baltimore Maryland USAN = 20Sample women in a continuing care retirement communityAge mean 88 (SD 37) years



Resnick 2002 (Continued)

Inclusion criteria able to walk 50 feet with or without assistive device sedentary lifestyleExclusion criteria cognitive impairment (MMSE gt20) terminal illness medical condition pre-cluding participation in aerobic exercise

Interventions 1 WALK intervention walk (join group or walk alone 20 min per week) address pain fear fatigueduring exercise learn about exercise cue by self modelling2 Control no intervention

Outcomes 1 Number of falls (mean) but not rate Insufficient data to include in analysis

Notes Participants lived independently in apartments and could ambulate independently (Personalcorrespondence) Pilot study with no usable data

Risk of bias


Adequate sequence generation Yes Randomised by coin flip (personal communication)


BlindingFalls


Low risk of bias in recall of falls Unclear Quote ldquobased on self-reportrdquo No additional information

Robertson 2001a


Participants Setting community West Auckland New ZealandN = 240Sample men and women living at home (68 women) identified from computerised registersat 17 general practices (30 doctors)Age mean 809 (SD 42) range 75-95Inclusion criteria aged 75 and overExclusion criteria inability to walk around own residence receiving physiotherapy at the time ofrecruitment not able to understand trial requirements

Interventions 1 Home exercise programme individually prescribed by district nurse in conjunction with herdistrict nursing duties (see Notes)Visit from nurse at 1 week (1 hour) and at 2 4 and 8 weeks and 6 months (half hour) plus monthlytelephone call to maintain motivation



Robertson 2001a (Continued)

Progressively difficult strength and balance retraining exercises plus walking plan Participantsexpected to exercise 3 x weekly and walk 2 x weekly for 1 year2 Control usual care


Notes District nurse had no previous experience in exercise prescription Received 1 weeksrsquo training fromresearch grouprsquos physiotherapist who also made site visits and phone calls to monitor qualityOtago Exercise Programme manual can be ordered from http wwwaccconzotagoexerciseprogramme

Risk of bias


Adequate sequence generation Yes Randomised using allocation schedule developed using computer gener-ated numbers


BlindingFalls


BlindingFractures

Yes Injuries reported by participants who were aware of their group allocationPhoned by independent assessor blind to allocation Person classifying fallevents also blind to allocation

Low risk of bias in recall of falls Yes Active fall registration with daily postcard calendars returned monthly +telephone calls

Robson 2003


Participants Setting community Alberta CanadaN = 660Sample healthy volunteers living in Edmonton area and two rural communities in AlbertaRecruited by newspaper adverts radio public notices and word of mouth (81 women)Age mean 730 (SD 67)



Robson 2003 (Continued)

Inclusion criteria able to walk unassisted for 20 minutes to get down and up off the floorunassistedExclusion criteria dizzy spells or ldquoother health problems that made it difficult for them to functionrdquo

Interventions 1 Two 90 minute group sessions one month apart taken by lay senior facilitatorsSession 1) Given Client Handbook (self assessed risk and risk reduction strategies relating tobalance strength shoes vision medications environmental hazards paying attention) Instructedto complete assessment and implement strategies to reduce risk by session 2 Given fitness video(Tai Chi movements for balance and leg strength) Used video in Session 1 and instructed touse daily for 20 minutes or get involved in community exercise programme for 45 minutes 3xper week Asked to identify and report community hazards Session 2) no details of this sessionprovided in paper2 Control received no intervention until after 4 months


Notes SAYGO (Steady As You Go) program

Risk of bias


Adequate sequence generation Unclear Quote ldquoRandomly assigned by phonerdquo Insufficient information to per-mit judgment

Allocation concealment Unclear Quote ldquoRandomly assigned by phonerdquo Insufficient information to per-mit judgment

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationUnclear whether people phoning were blind to allocation

Low risk of bias in recall of falls Yes Falls ascertained by mail-in calendars returned monthly with telephonefollow up

Rubenstein 2000


Participants Setting community California USAN = 59Sample men recruited from Veterans Administration ambulatory care centre (volunteers)Age mean 74



Rubenstein 2000 (Continued)

Inclusion criteria aged 70 and over ambulatory with at least 1 fall risk factor lower limb weaknessimpaired gait impaired balance more than 1 fall in previous 6 monthsExclusion criteria exercised regularly severe cardiac or pulmonary disease terminal illness severejoint pain dementia medically unresponsive depression progressive neurological disease

Interventions 1 Exercise sessions (strength endurance and balance training) in groups of 16-20 3 x 90 minutesessions per week for 12 weeks2 Control usual activities


Notes

Risk of bias


Adequate sequence generation Yes Randomised in blocks of 16-20 at 3-6 month intervals using randomlygenerated sequence cards in sealed envelopes

Allocation concealment Unclear Cards in sealed envelopes

BlindingFalls

No Falls reported by participants who were aware of their group allocationPerson ascertaining falls was aware of group allocation

Low risk of bias in recall of falls No No active fall registration Fall ascertainment for intervention group atweekly classes Controls phoned every 2 weeks

Rubenstein 2007

Methods CCT Cluster randomised Participants ldquopreviouslyrdquo randomised to one of three primary care prac-tice groups using last two digits of Social Security number Two practice groups then randomisedto intervention or control Third group not included as used in prior pilot study (personal com-munication)Losses at one year 98 of 792 (12)

Participants Setting Sepulveda Ambulatory Care Center (Veterans Affairs Greater Los Angeles Health CareSystem) California (USA)N = 792Sample all patients receiving care at ambulatory care centre (only 3 women)Age mean 745 (SD 6)




Inclusion criteria aged 65 and over previously randomised to either of the two practice groupsinvolved in the trial having had at least one clinic visit in previous 18 months scoring 4 or moreon GPSSExclusion criteria living over 30 miles from care centre already enrolled in outpatient geriatricservices at care centre living in long-term care facility scoring less than 4 GPSS

Interventions 1 Structured risk and needs assessment and referral algorithm implemented by case manager(physician assistant) Targetting five geriatric conditions including falls Assessment followed byreferrals and recommendations for further assessment or treatment 3 monthly telephone contactwith case manager2 Control usual care


Notes

Risk of bias


Adequate sequence generation No Participants ldquopreviouslyrdquo randomised to one of three primary care practicegroups using last two digits of Social Security number Two practice groupsthen randomised to intervention or control Third group not included asused in prior pilot study (personal communication)

Allocation concealment No Two groups therefore alternation

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessment research staff blind blind to allocation

Low risk of bias in recall of falls No Retrospective recall Annual telephone follow up each year for 3 yearsText states participants asked ldquoabout incidence of falls in the previousyearrdquo but table 2 reports one or more falls in the preceding 3 months

Ryan 1996


Participants Setting community Baltimore Maryland USAN = 45Sample rural and urban dwelling women Volunteers from senior meal sitesAge mean 78 range 67-90Inclusion criteria aged 65 and over living alone in own home ambulatory with or withoutassistive devices with telephone for follow up



Ryan 1996 (Continued)

Interventions Interview and physical assessment by nurse prior to randomisation1 1 hour fall prevention education programme discussing personal (intrinsic) and environmental(extrinsic) risk modification in small groups of 7-8 women (nurse led)2 Same educational programme but individual sessions with nurse3 Controls received health promotion presentation (no fall prevention component) in smallgroups of 7-8


Notes Pilot research Primarily to test methodology of a fall prevention education programme andresulting changes in fall prevention behaviour

Risk of bias




BlindingFalls

No Falls reported by participants who were aware of their group allocationTelephone contact was not blinded (both groups asked about falls butintervention groups asked about recollection of intervention)

Low risk of bias in recall of falls No Retrospective recall by monthly phone call for 3 months

Salminen 2008


Participants Setting community Pori FinlandN = 591Sample recruited through local newspapers pharmacies Pori Health Cente Satakunta CentralHospital private clinics and written invitation from health professionals (84 women)Age 62 aged 65 - 74 38 aged ge 75Inclusion criteria aged ge65 years fallen in last year MMSE ge 17 able to walk 10 metersindependently living at home or sheltered housingExclusion criteria none described

Interventions 1 Intervention geriatric assessment individually tailored intervention targeting muscle strengthand balance (advised to carry out physical exercises x3 per week at home) exercise in groups(three levels according to physical performance) vision (referral) nutritional guidance or referral



Salminen 2008 (Continued)

medications depression treatment and prevention of osteoporosis home hazard modificationAll received calcium and vitamin D2 Control counselling and guidance after comprehensive assessments

Outcomes 1 Rate of falls2 Number of fallers

Notes

Risk of bias


Adequate sequence generation Unclear Quote ldquoRandomizedrdquo No description of sequence generation

Allocation concealment Yes Quote ldquousing consecutively numbered sealed envelopesrdquo

BlindingFalls


Low risk of bias in recall of falls Yes Quote ldquorecorded by fall diaries that subjects were asked to mail to theresearch assistants monthlyrdquo

Sato 1999


Participants Setting community dwelling JapanN = 86Sample elderly people with Parkinsonrsquos disease (mean Hoehn and Yahr Stage 3) (59 women)Age mean 706 range 65-88Inclusion criteria aged 65 or overExclusion criteria history of previous non-vertebral fracture non-ambulatory (Hoehn and YahrStage 5 disease) hyperparathyroidism renal osteodystrophy impaired renal cardiac or thyroidfunction therapy with corticosteroids estrogens calcitonin etidronate calcium or vitamin Dfor 3 months or longer during the previous 18 months or at any time in the previous 2 months

Interventions 1 1 alpha (OH) Vitamin D3 10 mcg daily for 18 months2 Control identical placebo




Sato 1999 (Continued)

Other outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Unclear Randomisation by computer generated random numbers


BlindingFalls

Unclear Falls reported by participants who were blind to their group allocation(placebo-controlled trial) ldquoFollowed up every two weeks at which timesclinical status was assessed and non-vertebral fractures were recordedrdquoUnclear whether data collectors were blind to allocation

BlindingFractures

Yes Fractures reported by participants who were blind to their group allocation(placebo-controlled trial) ldquoFollowed up every two weeks at which timesclinical status was assessed and non-vertebral fractures were recordedrdquoUnclear whether data collectors were blind to allocation

Low risk of bias in recall of falls Unclear Interval recall Quote ldquoFollowed up every two weeks at which timesclinical status was assessed and non-vertebral fractures were recordedrdquoNumber of falls per subject ldquorecordedrdquo during 18 months Presume everytwo weeks

Schrijnemaekers 1995


Participants Setting Sittard The NetherlandsN = 222Sample men and women living at home ( N = 146) or in residential homes (N = 76) (70women)Age At least 75 years 70 aged 77-84 30 ge85Inclusion criteria aged 75 and over living at home or in one of two residential homes havingproblems with one or more of the following IADL ADL toileting mobility or fallen in last 6months serious agitation or confusion informed consent from participant and their GPExclusion criteria living in nursing home received outpatient or inpatient care from geriatric unitin previous 2 years



Schrijnemaekers 1995 (Continued)

Interventions 1 Comprehensive assessment in outpatient geriatric unit (geriatrician psychologist socialworker) advice to participant and GP about treatment and support2 Control usual care


Notes Included in this review as the majority of participants were living at home (N = 146)

Risk of bias


Adequate sequence generation Unclear Stratified by living condition (home vs home for the elderly) then ldquoran-domly allocatedrdquo by researcher in blocks of ten


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationUnclear whether data collectors were blind to allocation

Low risk of bias in recall of falls No Retrospective recall Falls ascertained retrospectively at interview Presumeasked about falls in previous 6 months

Sherrington 2004


Participants Setting community Sydney AustraliaN = 120Sample identified through 6 hospitals in Sydney following hip fracture (80 women)Age mean 79 (SD 9) 57-95 yearsInclusion criteria community dwelling recent hip fractureExclusion criteria severe cognitive impairment medical conditions complications from fractureresulting in delayed healing

Interventions 1 Weight-bearing home exercise group2 Non weight-bearing home exercise group3 Control no intervention


Notes Data obtained from authors



Sherrington 2004 (Continued)

Risk of bias


Adequate sequence generation Yes Quote ldquothe randomisation schedule was produced with a random num-bers table in blocks of sixrdquo

Allocation concealment Yes Quote ldquoSealed in opaque envelopesrdquoComment probably done as research group has described ldquoconcealedallocationrdquo in previous study

BlindingFalls

No Falls reported by participants who were aware of their group allocationAssessors not blind to group allocation

Low risk of bias in recall of falls No Retrospective recall Falls data collected at home visits at 1 and 4 months

Shigematsu 2008


Participants Setting Kawage Mie JapanN = 68Sample people aged 65-74 living in Kawage (63 women)Age mean 69 (SD 3) yearsInclusion criteria 65-74 years old community dwellingExclusion criteria severe neurological or cardiovascular disease mobility-limiting orthopaedicconditions

Interventions 1 Exercise intervention square-stepping exercises (forward backward lateral and oblique stepson a marked mat 250 cm long) supervised group sessions 70 minutes (30 warm up and cooldown) x2 per week for 12 weeks Group ldquofurther dividedrdquo at end of 12 weeks and half (N = 16)continued with sessions ldquofrom December 2004 through February 2005rdquo ie a further 12 weeks2 Exercise intervention outdoor supervised walking session 40 minutes x1 per week for 12 weeksAs above half (N = 18) continued walking for a further 12 weeks


Notes

Risk of bias



Shigematsu 2008 (Continued)


Adequate sequence generation Yes Quote ldquoRandomly allocated by a public health nurse who used a com-puterized random number generation program in which the numbers 0and 1 corresponded to the two groups respectivelyrdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessors not blind to group allocation

Low risk of bias in recall of falls Yes Quote ldquoAll the persons received a pre-paid postcard at the beginning ofeach month which they returned at the beginning of the next monthrdquoInstructed to record falls on a daily basis Phoned if falls reported

Shumway-Cook 2007

Methods RCTLosses none for falls analysis

Participants Setting community USAN = 453Sample volunteers recruited by press releases and advertising seniors newsletters cable televisionetc (77 women)Age mean 756 (SD 63) range 65-96Inclusion criteria aged 65 and over community dwelling able to speak English have a primarycare physician they had seen in last 3 years able to ambulate independently (with or without caneor walker) willing to attend exercise classes for at least 6 months have access to transportationExclusion criteria more than minimal hearing or visual problems regular exercise in previous 3months unable to complete 10 ft rsquoTimed up and Gorsquo test in lt30 seconds five or more errors onPfeiffer Short Portable Mental Status Questionnaire

Interventions Both groups completed health history questionnaire at randomisation1 Group exercise class 1 hr 3x per week for up to 12 months 6 hours of fall prevention classes fallassessment summary (based on initial questionnaire) sent to participantsrsquo primary care physicianplus copy of fall prevention guideline (AGSBGS 2001)2 Control usual care plus two fall prevention brochures


Notes



Shumway-Cook 2007 (Continued)

Risk of bias


Adequate sequence generation Yes Computer random number generator used to generate sequence

Allocation concealment Yes Randomised using centralised randomisation scheme accessed by tele-phone

BlindingFalls


Low risk of bias in recall of falls Yes Prospective Falling ascertained by 12 monthly calendars with telephonefollow up

Skelton 2005


Participants Setting community N = 100Sample women recruited using posters newspapers and radio stationsAge mean 728 (SD 59)Inclusion criteria aged ge 65 living independently in own home ge3 falls in previous yearExclusion criteria acute rheumatoid arthritis uncontrolled heart failure or hypertension signifi-cant cognitive impairment significant neurological disease or impairment previously diagnosedosteoporosis

Interventions 1 FAME exercise class 1 hour x1 per week for 36 weeks plus home exercises 30 min x2 per week2 Control no exercise class Home-based seated exercises x2 per week


Notes

Risk of bias


Adequate sequence generation Unclear Quote ldquorandomly allocated (blind)rdquo




Skelton 2005 (Continued)

BlindingFalls


Low risk of bias in recall of falls Yes Daily diaries returned every two weeks

Smith 2007


Participants Setting Wessex EnglandN = 9440Sample men and women recruited from age sex registers of 111 participating general practicesites (54 women) Mainly community dwelling (98)Age mean 791 (IQR 769 to 826)Inclusion criteria men and women aged 75 and overExclusion criteria current cancer any history of treated osteoporosis bilateral total hip replace-ment renal failure renal stones hypercalcaemia sarcoidosis taking at least 400 IU of vitamin Dsupplements already

Interventions 1 300000 IU ergocalciferol (vitamin D2) by intramuscular injection every autumn for 3 years2 Placebo

Outcomes 1 Number of people falling2 Number sustaining a fractureFalls a secondary outcome of the study Other outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Yes Individual randomisation within blocks at each practice by allocation ofconsecutively numbered ampoules

Allocation concealment Yes Individual randomisation within blocks at each practice by allocation ofconsecutively numbered ampoules

BlindingFalls


BlindingFractures

Yes Fractures reported by participants who were blind to their group allocation(placebo-controlled trial)



Smith 2007 (Continued)

Low risk of bias in recall of falls No Retrospective ldquoInformation on falls was obtained at annual review (1224 and 36 months) by the practice nurse and on incident fractures bypostal questionnaire at 6 12 18 24 30 and 36 monthsrdquo

Speechley 2008


Participants Setting community Ontario CanadaN = 241Sample male Canadian veterans of WWII and Korean War living in south-west OntarioAge mean (SD) 81 (38) yearsInclusion criteria living independently in the community able to understand and respond toquestionnaire at least one modifiable risk factor for falling identified by initial screening ques-tionnaire

Interventions Initial postal risk factor screening questionnaire to all potential participants1 Specialised geriatric services group comprehensive geriatric assessment with individual recom-mendations for fall risk factor reduction2 Family physician group participants sent letter summarising risk factors reported in question-naire Similar letter sent to participantrsquos family physician Treatment left to discretion of familyphysician

Outcomes 1 Number of fallers

Notes

Risk of bias




BlindingFalls


Low risk of bias in recall of falls Yes Monthly falls calendars returned for one year Telephone follow up ifcalendar not returned or falls reported



Spice 2009

Methods RCT (cluster randomised 18 general practices)

Participants Setting community Winchester UKN = 516 (proportion of women not stated)Sample patients in 18 general practicesAge mean age 82 yearsInclusion criteria community-dwelling men and women aged over 64 years history of at leasttwo falls in previous yearExclusion criteria none described

Interventions 1 Secondary care intervention multidisciplinary day hospital assessment by physician OT andphysiotherapist2 Primary care intervention health visitorpractice nurse falls risk assessment referral3 Control usual care


Notes Published as an abstract only Data from authors

Risk of bias


Adequate sequence generation Yes Cluster randomised Quote ldquoPractices were stratified into urban (three)and rural (fifteen) and randomly allocated to the three arms in blocksof three using a random number generator on a Hewlett Packard 21Spocket calculatorrdquo


BlindingFalls

No Falls reported by participants who were aware of their group allocationldquoBlinding to the intervention group of those collecting and analysing datawas impracticalrdquo

Low risk of bias in recall of falls Yes Follow up monthly using postcards with a phone call if a card not re-turned

Steadman 2003


Participants Setting community London United KingdomN = 198



Steadman 2003 (Continued)

Sample attendees at a multidisciplinary falls clinic district general hospital ( women not re-ported)Age mean 827 (SD 56)Inclusion criteria ge 60 years Berg Balance Scale lt45 after ldquoadequate management of potentialrisk factorsrdquoExclusion criteria amputation unable to walk 10 metres recent stroke progressive neurologicaldisorder unstable medical condition severe cognitive impairment

Interventions 1 Enhanced balance training Conventional physiotherapy plus balance training 45 minutes x2per week for 6 weeks1 Control conventional physiotherapy alone


Notes

Risk of bias


Adequate sequence generation Yes Quote ldquocomputer generated random numbersrdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationData collector theoretically blind to allocation

Low risk of bias in recall of falls Unclear Interval recall Falls data collected for previous month at 6 weeks 12weeks and 24 weeks

Steinberg 2000

Methods RCT Cluster randomised Four groups with approximately equal numbers formed from 2 or 3National Seniors Branches Groups randomly allocated to 1 of 4 interventionsLosses 9 of 252 (4)

Participants Setting community Brisbane Queensland AustraliaN = 252Sample volunteers from branches of National Seniors Association clubsAge mean 69 range 51-87Inclusion criteria aged 50 and over National Seniors Club member with capacity to understandand comply with the projectExclusion criteria none stated



Steinberg 2000 (Continued)

Interventions Cumulative intervention1 Control oral presentation video on home safety pamphlet on fall risk factors and prevention2 Intervention 1 plus exercise classes designed to improve strength and balance 1 hour permonth for 17 months exercise handouts gentle exercise video to encourage exercise betweenclasses3 Intervention 2 plus home safety assessment and financial and practical assistance to makemodifications4 Intervention 3 plus clinical assessment and advice on medical risk factors for falls


Notes Younger healthier and more active sample than elderly population as a whole

Risk of bias


Adequate sequence generation Unclear Quote ldquoGroups were randomly allocated to receive the four interven-tionsrdquo

Allocation concealment No Cluster randomised Possibility of participants joining group after ran-domisation

BlindingFalls


Low risk of bias in recall of falls Yes Quote ldquoFalls were monitored prospectively using a daily calendar diary tominimise biasrdquo Diary returned monthly Telephone follow up of reportedfalls and no monthly returns

Stevens 2001

Methods RCT Some clusters Study population divided into four strata defined by age (lt80 years and gt 80years) and sex Within these strata index recruits allocated in 21 ratio to control or interventionCoinhabitants assigned to same group as index recruitLosses 264 of 1879 (14)

Participants Setting community Perth AustraliaN = 1737Sample aged 70 and over living independently and listed on State Electoral Roll and the WhitePages telephone directory Assigned numbers and recruited by random selection (53 women)Age mean 76Inclusion criteria aged 70 and over living independently able to follow study protocol (cognitivelyintact and able to speak and write in English) anticipated living at home for at least 10 out of



Stevens 2001 (Continued)

12 coming months could make changes to the environment inside the home had not modifiedhome by fitting of ramps and grab railsExclusion criteria if living with more than 2 other older people

Interventions 1 One home visit by nurse to confirm consent educate about how to recognise a fall andcomplete the daily calendar Sent information on the intervention and fall reduction strategiesto be offered Intervention home hazard assessment installation of free safety devices and aneducational strategy to empower seniors to remove and modify home hazards (see rsquoNotesrsquo)2 Control one home visit by nurse to confirm consent educate about how to recognise a falland complete the daily calendar


Notes Hazard list designed with OT input to include factors identified from literature and existing checklists Eleven hazards included All identified hazards discussed with subjects but only the threemost conspicuous or remediable selected to give specific advice on their removal or modificationSafety devices offered at no cost and installed by tradesman within 2 weeks of visit

Risk of bias


Adequate sequence generation Unclear Study population divided into four strata defined by age (lt 80 years andgt 80 years) and sex Within these strata index recruits allocated in 21ratio to control or intervention Coinhabitants assigned to same group asindex recruit


BlindingFalls


Low risk of bias in recall of falls Yes Falls recorded on daily calendar

Suzuki 2004


Participants Setting community Tokyo JapanN = 52Age mean 78 (SD 39) range 73-90



Suzuki 2004 (Continued)

Sample and inclusion criteria participants in the Tokyo Metropolitan Institute of GerontologyLongitudinal Interdisciplinary Study on Aging attending a comprehensive geriatric health exam-ination living at home (100 women)Exclusion criteria unable to measure muscle strength poor mobility due to hemiplegia poorlycontrolled blood pressure communication difficulties due to impaired hearing

Interventions 1 Exercise-centered fall-prevention programme + home-based exercise programme aimed at en-hancing muscle strength balance and walking ability Ten one-hour classes (every 2 weeks for 6months) plus individual home-based exercises for 30 minutes x3 per week2 Pamphlet and advice on prevention of falls


Notes

Risk of bias


Adequate sequence generation Unclear ldquoRandomizedrdquo but method of randomisation not described


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationDoes not state whether outcome assessors were blind to allocation

BlindingFractures

Unclear Fractures reported by participants who were aware of their group alloca-tion Does not state whether outcome assessors were blind to allocation

Low risk of bias in recall of falls No Retrospective recall Falls and fractures recorded retrospectively at inter-view at 8 months and 20 months (falls in previous year)

Swanenburg 2007


Participants Setting Zurich SwitzerlandN = 24Sample unclear Probably patients in Center for Osteoporosis of the Department of Rheumatology(100 women)



Swanenburg 2007 (Continued)

Age mean 712 (SD 68)Inclusion criteria aged ge 65 living independently with osteoporosis or osteopeniaExclusion criteria severe peripheral or central neurological disease known to influence gait balanceor muscle strength medical contraindications for exercise

Interventions 1 Intervention vitamin 400-800 IU cholecalciferol and calcium 500-1000 mg per day accordingto physician assessment at baseline plus 12 week training programme to improve balance and adaily nutritional supplement enriched with proteins 3 months2 Control vitamin 400-800 IU cholecalciferol and calcium 500-1000 mg per day according tophysician assessment at baseline plus leaflet on home exercises


Notes Pilot study

Risk of bias


Adequate sequence generation Unclear Quote ldquoRandom assignment with a stratified randomisation proce-durerdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationOutcome assessors were blind to allocation

Low risk of bias in recall of falls No Quote ldquoFalls were assessed by interview at each assessmentrdquo post inter-vention 6 9 and 12 months Interval recall of 3 month period

Tinetti 1994

Methods RCT Cluster randomised with randomisation of 16 treating physicians matched in 4 groups of4 into 2 control and 2 intervention in each group enrolled subjects assigned to same group astheir physicianLosses 10 of 301 (3)

Participants Setting community Southern Connecticut USAN = 301Sample independently ambulant community dwelling individuals (69 women)Age mean 779 (SD 53)



Tinetti 1994 (Continued)

Inclusion criteria aged over 70 independently ambulant at least one targeted risk factor forfalling (postural hypotension sedativehypnotic use use of gt 4 medications inability to transfergait impairment strength or range of motion loss domestic environmental hazards)Exclusion criteria enrolment in another study MMSE lt 20 current (within last month) partic-ipation in vigorous activity

Interventions 1 Interventions targeted to individual risk factors according to decision rules and priority lists3 month programme duration2 Control visits by social work students over same period


Notes Yale (New Haven) FICSIT trial Risk factors screened for included postural hypotension seda-tivehypnotic drugs eg benzodiazepine 4 or more medications impaired transfer skills environ-mental hazards for falls impaired gait legarm muscle strength range of movement

Risk of bias


Adequate sequence generation Yes Quote ldquoComputerised randomization programrdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationOutcome assessors blinded to assignment

BlindingFractures

Unclear Fractures reported by participants who were aware of their group alloca-tion Outcome assessors blinded to assignment

Low risk of bias in recall of falls Yes Prospective Falls ldquoRecorded on a calendar that subjects mailed to theresearch staff monthlyrdquo followed by personal or telephone contact if nocalendar returned of a fall reported

Trivedi 2003

Methods RCT Stratified by age and sexLosses 648 of 2686 (24)

Participants Setting community UKN = 2686Sample mailed letter and information sheet to people from the British doctors study and generalpractice register in Suffolk (24 women)



Trivedi 2003 (Continued)

Age mean 75 (SD 5) range 65-85Inclusion criteria aged 65-85 yearsExclusion criteria already taking vitamin D supplements conditions with contraindications forvitamin D supplementation eg renal stones sarcoidosis or malignancy

Interventions 1 Oral vitamin D3 supplementation (100000 IU cholecalciferol) 1 capsule every 4 months for5 years2 Control matching placebo 1 capsule every 4 months for 5 years

Outcomes 1 Number of people falling2 Number sustaining a fractureOther outcomes reported but not included in this review

Notes Although fracture and major illness data collected every four months after capsules sent out fallsdata not collected until end of study Falls not mentioned in statistical analysis section of methods

Risk of bias


Adequate sequence generation Yes Quote ldquorandomised after stratification by age and sexrdquoComment probably done since earlier reports from the same investigatorsclearly describe use of random sequences

Allocation concealment Yes ldquoIpswich pharmacy revealed the codingrdquo at the end of the study So assumerandomised centrally

BlindingFalls


BlindingFractures


Low risk of bias in recall of falls No Retrospecive recall over 12 month period

Van Haastregt 2000


Participants Setting community Hoensbroek The NetherlandsN = 316



Van Haastregt 2000 (Continued)

Sample community dwelling men and women registered with 6 general medical practices (66women)Age mean 772 (SD 51)Inclusion criteria aged 70 and over living in the community 2 or more falls in previous 6 monthsor score 3 or more on mobility scale of Sickness Impact ProfileExclusion criteria bed ridden fully wheelchair dependent terminally ill awaiting nursing homeplacement receiving regular care from community nurse

Interventions 1 Five home visits from community nurse over 1 year Screened for medical environmental andbehavioural risk factors for falls and mobility impairment advice referrals and ldquoother actionsrdquo2 Control usual care


Notes

Risk of bias


Adequate sequence generation Yes Randomisation by computer generated random numbers


BlindingFalls


Low risk of bias in recall of falls Yes Falls recorded in weekly diary

Van Rossum 1993

Methods RCT Some clusters as people living together allocated to same groupLosses 102 of 580 (18)

Participants Setting community Weert The NetherlandsN = 580Sample general population sampled not volunteers (58 women)Age range 75-84 yearsInclusion criteria aged 75 to 84 living at homeExclusion criteria subject or partner already receiving regular home nursing care

Interventions 1 Preventive home visits by public health nurse x 4 per year for 3 years Extra visitstelephonecontact as required Check list of health topics to discuss Advice given and referrals to otherservices2 Control no home visits



Van Rossum 1993 (Continued)


Notes

Risk of bias


Adequate sequence generation Yes Stratified by sex self-rated health composition of household and socialclass then randomised by computer generated random numbers Partici-pants in intervention group then randomised to nurses


BlindingFalls


Low risk of bias in recall of falls No Retrospecitve Follow up at 1frac12 years and 3 years by postal survey andinterview Falls in previous 6 months recorded

Vellas 1991

Methods RCT Randomised 7 days after a fallLosses 6 out of 95 (6)

Participants Setting community Toulouse FranceN = 95Sample community dwelling men and women presenting to their general medical practitionerwith a history of a fall (66 women)Age mean 78 yearsInclusion criteria no biological cause for the fall fallen less than 7 days previouslyExclusion criteria hospitalised for more than 7 days after the fall demented sustaining majortrauma eg hip fracture or other fracture unable to mobilise or be evaluated within 7 days of thefall

Interventions 1 Iskeacutedylreg (combination of raubasine and dihydroergocristine) 2 droppers morning and eveningfor 180 days2 Control placebo for 180 days


Notes



Vellas 1991 (Continued)

Risk of bias


Adequate sequence generation Unclear Quote ldquoRandomisedrdquo Method of randomisation not described


BlindingFalls

Yes Falls reported by participants who were blind to their group allocation(placebo-controlled trial) ldquoDouble blindrdquo so assessors also blind to groupallocation

Low risk of bias in recall of falls Unclear Retrospective recall at 30 60 120 180 days

Vetter 1992

Methods RCT Cluster randomised by householdLosses 224 of 674 (33)

Participants Setting community Wales UKN = 674Sample men and women aged over 70 years on the list of a general practice in a market town (women not described)Age over 70 yearsNo exclusion criteria listed

Interventions 1 Health visitor visits minimum yearly for 4 years with advice on nutrition environmentalmodification concomitant medical conditions and availability of physiotherapy classes if desired2 Control usual care


Notes

Risk of bias


Adequate sequence generation Yes Cluster randomised by household ldquousing random number tables withsubjectsrsquo study numbers and without direct contact with the subjectsrdquo



Vetter 1992 (Continued)

Allocation concealment Yes Randomised ldquousing random number tables with subjectsrsquo study numbersand without direct contact with the subjectsrdquo Introduction of bias un-likely

BlindingFalls

No Falls reported by participants who were aware of their group allocationControl group had no contact between baseline assessment and end ofstudy (4 years)

BlindingFractures

No Fractures reported by participants who were aware of their group alloca-tion Control group had no contact between baseline assessment and endof study (4 years)

Low risk of bias in recall of falls No Falling status and fractures ascertained by interview at end of study period

Voukelatos 2007


Participants Setting community Sydney AustraliaN = 702Sample men and women recruited through advertisements in local papers (84 women)Age mean 69 (SD 65) range 69-70 yearsInclusion criteria aged over 60 community dwellingExclusion criteria degenerative neurological disease severely debilitating stroke metastatic cancersevere arthritis unable to walk across a room independently unable to use English

Interventions 1 Tai chi classes for 1 hour per week for 16 weeks (8 to 15 participants per class) at 24 communityvenues Style of tai chi differed between classes majority (83) involved Sun style two classes(3) Yang style remainder (14) involved a mixture of styles2 Control placed on 24 week waiting list then offered tai chi programme


Notes

Risk of bias


Adequate sequence generation Yes Quote ldquoRandomization list was prepared for each venue using ran-domly permuted blocks of four or sixrdquo



Voukelatos 2007 (Continued)


BlindingFalls


Low risk of bias in recall of falls Yes Quote ldquoParticipants were given falls calendars and were instructed torecord on the calendar each day for 24 weeks whether they had had afallrdquo Pre-paid postage calendars returned at the end of each month withtelephone call if not returned within 2 weeks

Wagner 1994


Participants Setting community Seattle USAN = 1559Sample rsquohealthy elderlyrsquo men and women HMO enrollees (59 women)Age mean 72 yearsInclusion criteria aged 65 and over HMO members ambulatory and independentExclusion criteria too ill to participate as defined by primary care physician

Interventions 1 60-90 minute interview with nurse including review of risk factors audiometry and bloodpressure measurement development of tailored intervention motivation to increase physical andsocial activity2 Chronic disease prevention nurse visit3 Control usual care


Notes Risk factors identified inadequate exercise high risk alcohol use environmental hazards if in-creased fall risk high risk prescription drug use impaired vision impaired hearing

Risk of bias


Adequate sequence generation Unclear Quote ldquoRandomized into three groups in a ratio of 212rdquo


BlindingFalls




Wagner 1994 (Continued)

Low risk of bias in recall of falls No Falls retrospectively measured at 1 and 2 years by mailed questionnaireInterviewed by phone if questionnaire not returned Data supplementedby computerised hospital discharge files

Weerdesteyn 2006

Methods RCTLosses none for falls data

Participants Setting community Nijmegan The NetherlandsN = 58Sample recruited using newspaper advertisements (72 women)Age mean 74 (SD 6)Inclusion criteria ge 65 years community dwelling ge1 fall in previous year able to walk 15minutes without a walking aidExclusion criteria severe cardiac pulmonary or musculoskeletal disorders pathologies associatedwith increased falls risk eg PD osteoporosis using psychotropic drugs

Interventions Three arms described but one not randomised1 Low-intensity exercise programme 15 hours x2 per week for 5 weeks First weekly sessionincluded gait balance and coordination training including obstacle avoidance Second sessionwalking exercises with changes of speed and direction and practice of fall techniques derived frommartial arts2 Control no training


Notes

Risk of bias


Adequate sequence generation Unclear Quote ldquoBlock randomization (3 blocks of 20) with gender stratificationwith equal probability for either exercise or control group assignmentrdquo

Allocation concealment Unclear Quote ldquoThe group allocation sequence was concealed (to both researchersand participants) until assignment of interventionsrdquo ldquoWe had participantsdraw a sealed envelope with group allocation ticket from a box containingall remaining envelopes in the blockrdquo (personal communication)

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationPerson coding the registration cards not blind to group allocation



Weerdesteyn 2006 (Continued)

Low risk of bias in recall of falls Yes Quote ldquoFalls were monitored monthly using pre-addressed reply-paidfall registration cardsrdquo Asked asked whether a fall had occurred in thepast month Sent a reminder if no registration card received

Whitehead 2003

Methods RCTLosses none reported after randomisation

Participants Setting community or low care residential care (hostel accommodation) Adelaide AustraliaN = 140Sample patients presenting with a fall to the ED over 22 week period (71 women)Age mean 778 (SD 70)Inclusion criteria aged 65 and over fall-related attendance at ED community dwelling or in lowcare residential care (hostel accommodation)Exclusion criteria resident in nursing home presenting fall related to stroke seizure cardiac orrespiratory arrest major infection haemorrhage motor vehicle accident being knocked to theground by another person MMSE lt25 no resident carer not English speaking living out ofcatchment area terminal illness

Interventions 1 Home visit and questionnaire ldquoFall risk profilerdquo developed and participant given written careplan itemising elements of intervention Letter to GP informing him of participantrsquos fall invit-ing them to review participant highlighting identified risk factors suggesting possible strategies(evidence based) GP also given one page evidence summary 2 Home visit No intervention Standard medical care from GP

Outcomes 1 Number of people fallingPrimary outcome was uptake of prevention strategies rather than falls

Notes Potential strategies review of medication use especially psychotropic drugs home assessment

Risk of bias


Adequate sequence generation Yes Randomisation and allocation schedules created by a researcher externalto the trial

Allocation concealment Yes Randomised by a researcher external to the trial using numbered sealedopaque envelopes

BlindingFalls




Whitehead 2003 (Continued)

Low risk of bias in recall of falls Yes Falls ascertained by falls diary and phone calls monthly to encourage useof the diary

Wilder 2001


Participants Setting community Wisconsin USAN = 60Sample ldquofrail elderlyrdquo no other descriptionAge no descriptionInclusion criteria aged ge 75 years living at home using home services (ie Meals on WheelsTelecare or Lifeline)Exclusion criteria none described

Interventions 1 Home modifications plus home exercise programme monitored by a ldquotrained volunteer buddyrdquo2 Simple home modifications3 Control no intervention

Outcomes 1 ldquoNumber of fallsrdquo but no data

Notes Abstract only

Risk of bias


Adequate sequence generation Unclear Quote ldquorandomly assignedrdquo to three arms Method not described


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationUnclear whether data collector was blind to group allocation

Low risk of bias in recall of falls Unclear Falls monitored by weekly telephone calls Interval recall over a shortperiod



Wolf 1996


Participants Setting community Atlanta USAN = 200Sample men and women residing in an independent living facility recruited by local advertise-ments and direct contact (81 women)Age mean 762 (SD 47)Inclusion criteria aged over 70 ambulatory living in unsupervised environment agreeing toparticipate on a weekly basis for 15 weeks with 4 month follow upExclusion criteria debilitating conditions eg cognitive impairment metastatic cancer cripplingarthritis Parkinsonrsquos disease major stroke profound visual defects

Interventions Three arms1 Tai Chi Quan (balance enhancing exercise) Group sessions twice weekly for 15 weeks (Indi-vidual contact with instructor approximately 45 minutes per week)2 Computerised balance training Individual sessions once weekly for 15 weeks (Individualcontact with instructor approximately 45 minutes per week)3 Control group discussions of topics of interest to older people with gerontological nurse 1hour once weekly for 15 weeks

Outcomes Used modified definition of a fall rather than agreed definition for FICSIT trials described inBuchner 19931 Rate of falls2 Number of people falling

Notes Atlanta FICSIT trial [Province 1995] 1997 paper included under this Study ID reports on a sub-group of the trial reporting on outcomes other than falls

Risk of bias


Adequate sequence generation Yes Randomised using ldquocomputer-generated fixed randomization procedurerdquo


BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationBlinding of assessors not described

Low risk of bias in recall of falls Yes Falls ascertained by monthly calendar or by monthly phone call fromproject staff



Wolf 2003

Methods RCT Cluster randomisedLosses 93 of 311 (30)

Participants Setting community Atlanta USAN = 311 (N = 20 clusters)Sample congregate living facilities (independent living facilities) recruited in pairs by whetherHousing and Urban Development (N = 14) or private (N = 6) sites with at least 15 participantsrecruited per site (94 women)Age mean 809 (SD 62) range 70-97 yearsInclusion criteria aged 70 and over one or more falls in previous year transitioning to frailtyExclusion criteria frail or vigorous elderly major cardiopulmonary disease cognitive impairment(MMSE lt24) contraindications for exercise eg major orthopaedic conditions mobility restrictedto wheelchair terminal cancer evidence of other progressive or unstable neurological or medicalconditions

Interventions 1 Intense Tai Chi (TC) 6 out of 24 simplified TC forms 60 minute session progressing to 90minutes 2x per week (10-50 minutes of TC) for 48 weeks Progressing from using upright supportto 2 minutes of TC without support2 Wellness education programme 1 hour per week for 48 weeks Instruction on fall preventionexercise and balance diet and nutrition pharmacological management legal issues changes inbody function mental health issues Interactive material provided but no formal instruction inexercise


Notes ldquoTransitioning to frailtyrdquo if not vigorous or frail based on age gaitbalance walking activity forexercise other physical activity for exercise depression use of sedatives vision muscle strengthlower extremity disability (Speechley M et al J Am Geriatr Soc 19913946-52)

Risk of bias


Adequate sequence generation Unclear Facilities stratified by socioeconomic status and randomised in pairsQuote ldquoFirst site in the pair was randomized to an intervention Thesecond site received the other interventionrdquo

Allocation concealment Unclear Insufficient information to permit judgment although allocation of sec-ond site in the pair could be predicted after the first site was randomised

BlindingFalls

Unclear Falls reported by participants who were aware of their group allocationAssessors blind to group allocation

Low risk of bias in recall of falls Yes Prospective Falls recorded on forms and submitted to instructor weekly+ phone call



Woo 2007


Participants Setting community Hong Kong ChinaN =180Sample recruited by notices posted in four community centres in in Shatin township (50women)Age mean 69 (SD 26)range 65-74 yearsInclusion criteria able to walk gt8 meters without assistanceExclusion criteria neurological disease which impaired mobility shortness of breath or anginaon walking up one flight of stairs dementia already performing Tai Chi or resistance trainingexercise

Interventions 1 Tai Chi using Hang style with 24 forms x3 per week for 12 months2 Resistance training exercises x3 per week using a Theraband for 12 months3 Control no exercise prescribed

Outcomes 1 Number of people fallingFalls a secondary outcome of this study Other outcomes reported but not included in this review

Notes

Risk of bias


Adequate sequence generation Yes Quote ldquoComputer generated blocked randomisationrdquo


BlindingFalls


Low risk of bias in recall of falls Unclear Methods used to ascertain falls not described

Wyman 2005

Methods RCTLosses of 272 ()

Participants Setting community Minnesota USAN = 272Sample randomised sample of Medicare beneficiaries in Twin Cities Metropolitan Area (100women)Age mean 79 (SD 6) range 70 to 99 years



Wyman 2005 (Continued)

Inclusion criteria gt70 years community dwelling mentally intact ambulatory ge2 risk factorsfor falls medically stableExclusion criteria currently involved in regular exercise

Interventions 1 Multifactorial intervention comprehensive fall risk assessment by nurse practitioner exercise(walking with weighted balance and coordination exercises) fall prevention education provisionof two night lights individualised risk reduction counselling for 12 weeks followed by tapered16 week computerised telephone monitoring and support2 Control health education on topics other than fall prevention In-home intervention for 12weeks followed by tapered 16 week computerised telephone monitoring and support


Notes

Risk of bias


Adequate sequence generation Yes Quote ldquoParticipants were stratified according to age group and ran-domized using a permutated block design with varying block sizes of fourand six to assure that the number of participants was balanced in eachtreatment grouprdquo


BlindingFalls


Low risk of bias in recall of falls Yes Quote ldquoFalls were measured daily on a calendar that was mailed inmonthlyrdquo

AampE accident and emergency departmentADL activities of daily livingAMT abbreviated mental testBMD bone mineral densityBMI body mass indexCCT controlled clinical trial (quasi-randomised)CHF congestive heart failureCSH carotid sinus hypersensitivityCSM carotid sinus massageECG electrocardiogramERT estrogen replacement therapyd dayED emergency departmentFICSIT frailty and injuries cooperative studies of intervention techniquesGP general practitioner



GPSS Geriatric Postal Screening SurveyHMO health maintenance organisationHRT hormone replacement therapyIADL instrumental activities of daily living More complex than ADL eg handling personal finances preparing meals shoppinghousekeeping travelling using the telephoneiPTH intact parathyroid hormoneIQR interquartile rangem metersmcg microgramMMSE mini mental state examinationNSAID nonsteroidal anti-inflammatory drugsng nanogram (multiply by 2496 to convert to nanomolesL)nmol nanomoleOT occupational therapistPD Parkinsonrsquos diseasePTH parathyroid hormoneRCT randomised controlled trialSD standard deviationSF36 medical outcomes study 36-item short form questionnaire a standard measure of health related quality of lifeSF12 a validated abbreviated form of the above quality of life assessment toolx times25(OH)D 25-hydroxy-vitamin Dlt less thangt more than

Characteristics of excluded studies [ordered by study ID]

Alexander 2003 Controlled trial Not strictly randomised Intervention multifactorial fall risk assessment in day care centresFalls outcomes

Alp 2007 RCT Intervention self-management classes for osteoporotic women (post-menopausal or idiopathic os-teoporosis) Not just older women mean 66 (SD 12) mean minus 1SD lt60 Falls outcomes for outdoorfalls only

Armstrong 1996 RCT Intervention hormone replacement therapy in post menopausal women Not just older womenrange 45-70 mean 609 (SD 58) mean minus 1SD lt60 Falls outcomes

Barr 2005 Controlled trial 171 non responders added to intervention group after randomisation Interventionscreening for fracture risk and GPs advised to prescribe calcium and vitamin D Falls outcomes

Bogaerts 2007 RCT Intervention whole body vibration training for one year Falls recorded in laboratory setting duringdynamic computerized posturography testing

Buchner 1997b RCT Intervention endurance training (MoveIT study) No falls outcomes Same control group as includedFICSIT study (Buchner 1997a)



(Continued)

Byles 2004 RCT Intervention home-based health assessment No falls outcomes Mackenzie 2002 and 2006 reportan epidemiological sub-study of Byles 2004 using a stratified sample of 264 randomly selected participants

Chapuy 2002 RCT Intervention vitamin D plus calcium Falls outcomes Not community participants described asldquo583 ambulatory institutionalized womenrdquo in ldquo55 apartment homes for elderly peoplerdquo Administrationof vitamin D or placebo supervised by nurses at mealtimes ie intermediate level nursing care facilitiesIncluded in institutional falls review (Cameron 2005) after discussion with review authors

Cheng 2001 RCT Intervention symmetrical standing training and repetitive sit-to-stand training for stroke patientsNot just older people mean 627 (SD 79) mean minus 1SD lt 60 Falls outcomes

Crotty 2002 RCT Intervention accelerated discharge and home based rehabilitation after hip fracture Not interventionto prevent falls falls recorded as adverse events

De Deyn 2005 RCT Intervention antipsychotic (aripiprazole) versus placebo in patients with Alzheimerrsquos disease Notintervention to prevent falls only reported falls considered to be caused by the medication (adverse events)

Ebrahim 1997 RCT Intervention brisk walking in post menopausal women Not just older women mean 681 (SD 88)mean minus 1SD = lt60

Elley 2003 RCT (clustered) Intervention activity counselling and Green Prescription to increase physical activity inolder people Outcomes activity levels and quality of life Falls reported as adverse events

Faber 2006 RCT Intervention 1 functional walking Intervention 2 in balance (Tai Chi) Control usual activitiesFalls outcomes Excluded from this review as participants in 15 long-term care centres including self-careand nursing care facilities Included in institutional falls review (Cameron 2005) after correspondence withauthor

Freiberger 2007 Reported as an RCT but control group not randomised

Gill 2002 RCT Intervention home-based intervention including physical therapy to prevent functional decline Fallsreported as adverse events

Graafmans 1996 An epidemiological study of risk factors for falls in a self-selected subgroup of 368 subjects from an RCT ofdaily vitamin D versus placebo with 2578 participants Of 458 eligible subjects only 368 agreed to enrol inthis study (801) Percentage who fell in intervention and control groups are reported but it was felt thatthis paper should be excluded as the sample was a self-selected subgroup and the number in interventionand control groups were not provided There was no statistically significant difference in percentage offallers with or without vitamin D (OR 10 95 CI 06 to 15)

Hirsch 2003 RCT Intervention balance and resistance training versus balance Parkinsonrsquos disease Outcome balance(ability to balance under progressively more difficult conditions ie artificially induced falls)

Hu 1994 RCT Not fall prevention Falls artificially induced Balance parameters measured



(Continued)

Inokuchi 2007 Not RCT Was to have been an RCT but study design changed Potential participants and controls selectedfrom different sites Intervention nurse-led community exercise programme Falls outcomes

Iwamoto 2005 RCT Intervention whole body vibration (WBV) plus alendronate versus alendronate Aim to investigatewhether WBV enhanced effect of alendronate on BMD bone turnover and chronic back pain in peoplewith osteoporosis (age 55-88) Falls reported but only one person fell during year follow up in interventiongroup versus two in control group

Kempton 2000 Not RCT Evaluation of non-randomised community fall prevention programme targeting eight risk factorsGeographical control

Kerschan-Schindl 2000 Not RCT Sample selected from controlled trial of home exercise programme Falls outcomes

Larsen 2005 RCT Three intervention arms vitamin D plus calcium versus same plus home safety versus home safetyalone versus no intervention Outcome only rsquoseverersquo falls leading to acute hospital admission No significantdifference in number of rsquoseverersquo falls for any group

Lee 2007 RCT Intervention personal emergency response system (portable alarm and speaker microphone) Out-come anxiety and fear of falling Falls monitored as reason for using alarms Not designed to reduce falls

Lehtola 2000 RCT Intervention exercise Translated from Finnish Excluded because of apparent discrepancies in re-porting of data Clarification sought from authors but no response

Lin 2006 Not RCT Intervention Tai Chi Controlled trial with two intervention villages (selected because they hadthe largest older populations) versus four control villages Outcome injurious falls that required medicalcare

Linnebur 2007 Baseline data from ongoing RCT Intervention not described Falls not collected at follow up

Mansfield 2007 RCT Intervention perturbation-based balance training programme ldquoFallsrdquo monitored during perturbationby pressure on safety harness

Marigold 2005 RCT Intervention exercise for people with chronic stroke Falls outcomes Not just older people excludedas mean - 1SD lt60

Mead 2007 RCT Intervention endurance and resistance training versus relaxation for people who have had a strokeOutcomes functional measures Falls reported as adverse events

Means 1996 RCT nested within a pre-test post-test experimental design Both groups received the same exercise inter-vention randomisation was to test whether repeated exposure to the functional obstacle course used asa performance measure in the study resulted in an improvement in performance in that test Previouslyincluded in Cochrane review as falls data was presented by group this was a pilot study for a larger trialwhich has been included in this review (Means 2005)



(Continued)

Ondo 2006 Random order bilateral ventralis intermedius nuclei deep brain stimulation in patients with Parkinsonrsquosdisease or essential tremor Falls monitored during balance assessment with patients wearing a harness

Peterson 2004 RCT Intervention motivational video educational booklet supporting peer counselling and high inten-sity muscle strength training in hip fracture patients post discharge Outcome functional outcome (SF36)Trialists planned to include falls outcomes but insufficient falls data to carry out reliable analysis

Poulstrup 2000 Not RCT Community-based fall prevention intervention with non-randomised control communitiesOutcome fall related fractures

Protas 2005 RCT Eighteen participants with Parkinsonrsquos disease Analysed as pre-post intervention and not all partic-ipants included in analysis No data or results for inclusion in the review

Resnick 2007 RCT Intervention self-efficacy intervention alone exercise plus self-efficacy exercise alone (three arms)versus routine care in older women after hip fracture Author states falls were not an outcome (personalcommunication)

Robertson 2001b Not RCT Controlled trial in multiple centres Intervention home based exercise in over 80 year oldsSame programme as in Campbell 1997 Campbell 1999 and Robertson 2001a Outcome falls injuriesresulting from falls and cost effectiveness

Rosie 2007 RCT Intervention functional home exercise (repeated sit-to-stands versus low-intensity progressive resis-tance training) Outcomes multiple gait balance and falls efficacy assessments Falls reported as adverseevents

Rucker 2006 Not RCT Non-randomised ldquoon-off rdquo time series scheme Intervention educational intervention in com-munity-dwelling people aged ge50 with history of wrist fracture Outcome falls and fear of falling

Sakamoto 2006 RCT Intervention unipedal standing balance exercise Information from author institutional setting(special nursing homes for the aged and nursing care facilities) Included in institutional falls review (Cameron 2005) after correspondence with author

Sato 2002 RCT Intervention menatetrenone (vitamin K) for treating osteoporosis and preventing fractures in womenwith Parkinsonrsquos disease and vitamin D deficiency Control no intervention Not a fall-prevention interven-tion Report number of falls per subject (erratum published) but because of interaction with osteoporosisin risk of fracture

Sato 2005a RCT Intervention risedronate and ergocalciferol (vitamin D2) and calcium for preventing fractures inwomen with dementia and probable Alzheimerrsquos disease Control placebo risedronate and ergocalciferol(vitamin D2) and calcium Not a comparison of fall-prevention interventions as both groups receivedvitamin D Reports change in number of fallers pre-post intervention in both groups

Sato 2006 RCT Intervention alendronate plus vitamin D for prevention of fractures in people with Parkinsonrsquosdisease Control placebo plus vitamin D Not a comparison of fall-prevention interventions as both groupsreceived vitamin D Reports change in number of fallers pre-post intervention in both groups



(Continued)

Schwab 1999 Not RCT 1999 letter appeared to describe an RCT but not confirmed by subsequent publications orcorrespondence with authors

Shaw 2003 RCT with falls outcomes All had MMSE lt 24 Not community as 79 of participants lived in high andintermediate nursing care facilities Included in institutional falls review (Cameron 2005) after correspon-dence with author

Shimada 2003 RCT Not community institutional setting (geriatric health services facility in Japan) Included in institu-tional falls review (Cameron 2005) after correspondence with author

Singh 2005 RCT Intervention high versus low-intensity weight training versus GP care for depression in older peopleFalls reported as adverse events ie the hypothesis is that the intervention might increase falls not reducethem

Sohng 2003 RCT Intervention community-based ldquofall prevention exercise programmerdquo with no falls outcome Out-come muscle strength ankle flexibility balance IADL depression

Sumukadas 2007 RCT Intervention perindopril (ACE inhibitor) versus placebo Falls reported as adverse events

Tennstedt 1998 RCT Intervention to reduce fear of falling and increase activity levels Not fall prevention Falls reportedas possible adverse effect

Thompson 1996 Not RCT Pre-post intervention Environmental risk factor modification Falls outcomes

Tideiksaar 1992 Not RCT Community based survey and falls prevention programme Qualitative evaluation only Fallsoutcomes

Tinetti 1999 RCT Intervention home based multiple component rehabilitation after hip fracture Not intervention toprevent falls falls recorded but as adverse events

Von Koch 2001 RCT Intervention rehabilitation at home after a stroke Not intervention to prevent falls falls recordedas adverse events

Ward 2004 RCT Intervention to prevent skin sores and falls in people with progressive neurological conditions Notjust older people age range 22-89 years median 65 Excluded as not prevention of falls in older peopleand results not reported by age

Wolf-Klein 1988 Not RCT Pre-post intervention (multidisciplinary falls clinic) Falls outcomes

Wolfson 1996 RCT Intervention exercise Outcome balance strength and gait velocity No falls outcome FICSIT trial

Yardley 2007 RCT Intervention Internet provision of tailored advice on falls prevention activities for older people Nofalls outcomes



(Continued)

Yates 2001 RCT Multifactorial intervention to reduce fall risk Outcome decrease in selected fall risk factors No fallsoutcomes

Ytterstad 1996 Not RCT Quasi experimental with non-randomised controls Pre-post intervention design Outcomesinclude falling

AampE accident and emergencyBMD bone mineral densityGP general practitioner (family physician)RCT randomised controlled trialIADL instrumental activities of daily living

Characteristics of studies awaiting assessment [ordered by study ID]

Beyer 2007

Methods Randomised controlled trial

Participants Setting Copenhagen DenmarkN = 65Sample women with a history of a fall identified from hospital recordsAge 70-90 yearsInclusion criteria home-dwelling aged 70 to 90 years history of a fall requiring treatment in hospital emergencydepartment but not hospitalisation able to come to training facilityExclusion criteria lower limb fracture in last 6 months neurological diseases unable to understand Danish cognitivelyimpaired (MMSE lt24)

Interventions Supervised group exercise programme (flexibility lower limb resistance exercise balance training stretching) 60minutes 2x per week for 6 months

Outcomes Primary outcomes measures of muscle strength and function Falls a secondary outcome recorded for one year usingcalendar

Notes Not yet assessed

Di Monaco 2008

Methods Quasi-randomised trial (alternation)

Participants N = 95Sample women in hospital after a fall-related hip fractureInclusion criteria history of hip fracture community-dwelling aged ge60 years



Di Monaco 2008 (Continued)

Interventions Intervention multidisciplinary fall prevention programme during hospital stay plus single home visit by occupationaltherapist after dischargeControl as above but no home visit

Outcomes Falls recorded retrospectively at 6 months follow up

Notes Intervention commences in hospital but designed to prevent falls in the community Not yet assessed

Madureira 2007

Methods ldquoRandomized consecutively into two groupsrdquo

Participants 66 women with osteoporosis attending an outpatient clinic Unclear whether community-dwelling BrazilInclusion criteria osteoporosisExclusion criteria secondary osteoporosis visual deficiency hearing deficiency vestibular alteration unable to walkmore than 10 meters independently contraindications for exercise training

Interventions Intervention balance training programme for 1 hour a week for 40 weeksControl no intervention

Outcomes Falls a secondary outcome Primary outcomes are functional balance static balance and get up and go test

Notes No raw data usable summary statistics available Additional information required

Pfeifer 2004

Methods One-year randomised controlled trial

Participants 242 men and women aged over 70 years in Germany

Interventions 800 IU vitamin D3 and 1000 mg calcium or 1000 mg daily

Outcomes Falls and muscle power

Notes Published abstracts only Not yet assessed

Sato 2005b


Participants Two hundred ambulatory women with dementia and probable Alzheimerrsquos disease aged 70 years and over



Sato 2005b (Continued)

Interventions Intervention menatetrenone (vitamin K) and vitamin D2 and calciumControl no treatment

Outcomes Fractures and number of falls per participant

Notes

Weber 2008

Methods Cluster randomised by clinic site

Participants N = 620 peopleInclusion criteria aged over 70 community-dwelling at risk of falls based on age and medication use

Interventions Electronic medical record (EMR) system to identify at-risk patients and reduce medication use Standardised medi-cation review and recommendations to physician via EMR system

Outcomes Falls medication use and psychoactive medication useFalls self-reported at three month intervals for 15 months

Notes



Characteristics of ongoing studies [ordered by study ID]

Behrman

Trial name or title Prediction and prevention of falls in the elderly


Participants 500 individuals aged over 75 years at high risk of developing disabilities from each general practice inMaidenhead

Interventions 1 Intervention full geriatric assessment at day hospital and course of group exercises2 Control usual care

Outcomes Changes in Barthel score mental depression score change in residential status mortalityFalls not mentioned in list of outcomes but title and research question describe prevention of falls anddisability

Starting date April 1997 (completed data analysis ongoing)

Contact information Dr R BehrmanGeriatric DeptSt Markrsquos HospitalMaidenheadSL6 6DUBerksUKTelephone +44 1753 638532

Notes falls outcomes

Blalock

Trial name or title Preventing falls through enhanced pharmaceutical care

Methods Randomised controlled trial single blind (outcomes assessor)

Participants 200 men and women aged ge65Inclusion criteria taking ge 4 prescription medications taking ge 1 high risk medication ge 1 falls during 12month period before study entry able to speak and read EnglishExclusion criteria resident of long term care facility cognitive impairment housebound

Interventions 1 Pharmacist intervention participants receive written information about falls prevention and a personalconsultation from a community pharmacist concerning their medication regimen (identifying side effects etc)Pharmacist follow up as required with participantsrsquo physicians to coordinate any recommended medicationchanges2 Control written fall prevention information only



Blalock (Continued)

Outcomes Time to first fall and proportion of individuals who fall during the one-year follow-up period

Starting date August 2004 to September 2009

Contact information Dr S BlalockInjury Prevention Research CenterUniversity of North CarolinaChapel Hill North CarolinaUSA 27599-7505

Notes

Ciaschini

Trial name or title FORCE (Falls Fracture and Osteoporosis Risk Control Evaluation) study

Methods Randomised controlled trial Cross over at 6 months

Participants Community-dwelling Canada aged 55 years and over able to give consent at risk of falls or fracture Excludedif already receiving appropriate osteoporosis therapy

Interventions Osteoporosis risk assessment and evidence-based management Falls risk assessment intervention and occu-pational therapy or physiotherapy referral

Outcomes Primary outcomes are appropriate osteoporosis management and falls assessment by 6 months Secondaryoutcomes number of falls and fractures recorded in monthly diaries

Starting date March 2003 to January 2006

Contact information Dr M Ciaschini MD FRCPCGroup Health CentreSault St MarieOntarioCanada

Notes Protocol published 2008 but study completed in 2006

Cryer

Trial name or title A primary care based fall prevention programme evaluation of the Canterbury fall prevention programme




Cryer (Continued)

Participants One general practice Canterbury UK Fallers referred by GP staff and identified in AampEInclusion criteria falling in previous 2 weeks aged at least 65 years living independently in the communityregistered with target general practice able to communicate well enough to participateExclusion criteria unable to speak English too mentally confused medical reason for falling terminally illsudden onset of paralysis moved out of area

Interventions 1 Intervention home interview and assessment including medication review and referral to other agenciesgroup intervention 2 x per week for 6 months for seated exercise practice getting up from floor groupdiscussion re health and emotional needs2 Control usual careIntervention carried out by East Kent Health Promotion Service and nurses employed by the general practice

Outcomes Follow up at 6 12 and 18 monthsFalls

Starting date August 1996 (completed)

Contact information Dr Colin CryerCentre for Health Services StudiesGeorge Allen WingUniversity of KentCanterburyKentCT2 7NFUK

Notes Methods reported in Allen A Simpson JM Physiotherapy Theory and Practice (1999)15121-133

Donaldson

Trial name or title Action seniors A 12-month randomised controlled trial of a home-based strength and balance-retrainingprogramme in reducing falls


Participants People aged 70 or over seen at Falls Clinic due to presenting at AampE or to GP with fall or fall related injuryStratified by sex and Falls Clinic physician

Interventions 1 Twelve-month home-based strength and balance-retraining programme (Otago Exercise Programme)2 Control semi-structured interview about their presenting fall and their experience seeking care for the fallat AampE

Outcomes Fall rates injury rates time to first fallAlso changes in risk factors Falls recorded in monthly diaries



Donaldson (Continued)

Starting date October 2004

Contact information MG DonaldsonPhD CandidateHealth Care and EpidemiologyFaculty of Medicine University of British Columbia5804 Fairview AvenueVancouverBritish Columbia CANADAV6T 1Z3Telephone +1 604 875 4111 extension 62470Email meghangdinterchangeubccaAlternative contactProf Karim KhanFamily PracticeUniversity of British ColumbiaEmail khaninterchangeubcca

Notes Interim paper published (Liu-Ambrose et al 2008) reporting executive functioning outcomes

Edwards

Trial name or title Randomised controlled trial of falls clinic and follow up home intervention


Participants Volunteer community living seniors residing in apartments

Interventions 1 On site ldquofalls clinicrdquo assessment to identify those at high risk of falls followed by intensive in-homecomprehensive assessment and tailored intervention programmeControl low intensity educational session

Outcomes Incidence and risk of falls

Starting date (completed)

Contact information Prof Nancy EdwardsCareer ScientistSchool of NursingUniversity of OttawaCanadaEmail nedwardsuottawaca



Edwards (Continued)

Notes Ongoing trial described in Edwards N Cere M Leblond D A community-based intervention to prevent fallsamong seniors Family and Community Health 1993 15(4)57-65

Grove

Trial name or title Effects of Tai Chi training on general wellbeing and motor performance in patients with Parkinsonrsquos disease

Methods Randomised crossover trial

Participants 20 patients with Parkinsonrsquos disease recruited from a Parkinsonrsquos disease clinic

Interventions Tai Chi training

Outcomes Get up and go test ldquolog book of fallsrdquo

Starting date March 2000

Contact information Dr M GroveRoyal Cornwall Hospitals NHS TrustTreliskeTruroTR1 3LJUK

Notes

Haines

Trial name or title Assessment and prevention of falls functional decline and hospital re-admission in older adults post-hospi-talisation

Methods Randomised controlled trial Allocation via sequential opening of opaque envelopes containing computergenerated random number sequence

Participants Target sample size 156Inclusion criteria aged ge 65 using a gait aid to mobilise discharged from hospital to a community dwellingnot referred for post-discharge community rehabilitation servicesControl unstable severe cardiac disease cognitive impairment aggressive behaviour restricted weight-bearingstatus

Interventions 1 Intervention self-progressed home exercise program in DVD and booklet format to be completed 3 to 7times per week Active encouragement for 8 then 18 weeks without active encouragement2 Control usual daily activities



Haines (Continued)

Outcomes Number of falls (self recorded for 6 m then by monthly phone calls for 6 m

Starting date April 2007

Contact information Dr T HainesPhysiotherapy Department Geriatric Assessment and Rehabilitation Unit (GARU)Princess Alexandra HospitalIpswich RdWoolloongabbaQueensland 4102AustraliaEmail Terrence˙Haineshealthqldgovau

Notes

Hill a

Trial name or title RCT to evaluate the effectiveness of a targeted and personalised multifactorial program to reduce furtherfalls and injuries for community-dwelling older fallers presenting to and being discharged directly from anemergency department


Participants Aproximately 800 people aged 60 and over presenting to AampE (Melbourne Australia) because of a fall anddischarged directly homeInclusion criteria living in the community or a retirement village able to provide informed consent or hasconsent provided by a third party able to comply with simple instructions able to walk independently indoorswith or without a gait aid

Interventions 1 Intervention usual care put in place by AampE plus comprehensive falls risk assessment within one week ofbeing discharged home from AampE and again twelve month later2 Control usual care

Outcomes Falls and fall related injuries monitored for twelve months through a falls diary

Starting date December 2003 to December 2006

Contact information Irene Blackberry MB PhDNational Ageing Research InstituteMelbourneVictoria 3052AustraliaEmail iblackberrynariunimelbeduau



Hill a (Continued)

Notes

Hill b

Trial name or title Falls prevention for stroke patients following discharge home A randomised trial evaluating a multifactorialfalls prevention program (FLASSH)

Methods Randomised controlled trial Allocation sequence generated by computer Allocated using sealed envelopes

Participants 214 participantsInclusion criteria stroke patients (men and women aged ge 50) discharged home at risk of falls due to previousfall or balance impairmentExclusion criteria discharged to residential care facilities patients and carers without basic English

Interventions 1 Multifactorial individualised falls prevention program based on falls risk factors 12 month home exerciseprogram falls education (1 session) referral to address identified risk factors plus usual care ie therapyprescribed by the discharging facility2 Usual care therapy prescribed by discharging facility (variable but approximately 3 months)

Outcomes Falls time to first fall fall rate Falls data collected prospectively via monthly fall calendars for 12 months

Starting date June 2006

Contact information Prof K HillNational Ageing Research Institute34-54 Poplar RdParkvilleVictoria 3052AustraliaEmail khillnariunimelbeduau

Notes May not be included Depends on distribution of ages as recruiting people aged 50 or more

Jee

Trial name or title Incorporating vision and hearing tests into aged care assessment


Participants Target sample size 1400

Interventions 2 X 2 factorial designFour groups All receive standardized questionnaire plus vision tests hearing tests vision and hearing testsor no additional tests



Jee (Continued)

Outcomes One year follow upFalls quality of life physical and cognitive function use of health and community aged care services admissionto nursing home

Starting date 2005

Contact information Dr JJ WangSenior Research FellowCentre for Vision ResearchWestmead Millennium InstituteUniversity of Sydney C24Westmead HospitalSydneyNSWAustraliaEmail jiejin˙wangwmiusydeduau

Notes

Johnson

Trial name or title Community care and hospital based collaborative falls prevention project


Participants Target sample size 200Inclusion criteria male or female aged ge65 presenting to AampE or falls clinic community dwelling in PerthnorthExclusion criteria functional cognitive impairment unable to speak or read English

Interventions 1 Intervention community follow up by support worker (8 hours over 2-3 weeks) to review risk factors inthe home strategies to reduce risk factors assistance to implement Falls Action Plan provided by AampE orclinic (see ANZCTR website for further details)2 Control no community follow up after discharge

Outcomes Number of falls (falls calendar)


Contact information J JohnsonPerth Home Care Services30 Hasler RoadPO Box 1597Osborne Park



Johnson (Continued)

Western Australia 6017AustraliaEmail jayejphcsorgau

Notes

Kenny

Trial name or title SAFE PACE 2 Syncope and falls in the elderly - pacing and carotid sinus evaluation a randomised controlledtrial of cardiac pacing in older patients with falls and carotid sinus hypersensitivity


Participants 226 patients with carotid sinus hypersensitivity in over 30 centres across the UK Europe and North AmericaPatients screened in AampE geriatric medicine general medicine and orthopaedic facilitiesInclusion criteria gt50 years old 2 or more unexplained falls in previous 12 months cardioinhibitory response(gt3 seconds asystole) to carotid sinus massageExclusion criteria cognitive impairment (MMSE lt20) atrial fibrillation

Interventions 1 Intervention Medtronic Kappa 700 (Europe) or Kappa 400 (North America) pacemaker2 Control implantable loop recorder (Medtronic Reveal)

Outcomes Weekly fall diariesNumber of fallers in 24 months after interventionSecondary outcomesNumber of falls frequency of dizzy symptoms injury rates the use of primary secondary and tertiary carefacilities cognitive functionResource use and cost data collected

Starting date May 1999 (completed)

Contact information Prof RA KennyDept of Medical GerontologyTrinity College DublinDublin

Notes International multicentre trial



Klaber Moffett

Trial name or title PREFICS - Prevention of Falls and Injuries in a Community Sample effectiveness of a supervised exerciseprogram for falls prevention


Participants 1 Women aged over 60 years2 One fall or more in the year3 Independently mobile with or without a walking aid4 Able to follow simple instructions5 Resident in Hull and district

Interventions 1 Intervention supervised exercise class aimed at improving balance and strength2 Control home exercise sheets provided

Outcomes Number of fallsFall related injuriesFear of fallingQuality of lifePhysical data (balance etc)Follow up for 12 months using rsquofalls diariesrsquo The use of health care resources will be recorded for use in ahealth economic evaluation

Starting date April 2005 (completed)

Contact information Prof J Klaber MoffettProfessor of Rehabilitation and TherapiesDeputy DirectorInstitute of RehabilitationUniversity of Hull215 Anlaby RoadHullHU3 2PGUKTelephone +44 1482 675639Email jkmoffetthullacuk

Notes

Lesser

Trial name or title Vestibular rehabilitation in prevention of falls due to vestibular disorders in adults


Participants Adults with vestibular disorders



Lesser (Continued)

Interventions Vestibular rehabilitation (no further details available)

Outcomes Falls and quality of life


Contact information Mr THJ LesserOtolaryngologyUniversity Hospital AintreeLongmoor LaneLiverpoolL9 7ALUKTelephone +44 151 529 4035Fax +44 151 529 5263

Notes

Lips

Trial name or title Prevention of fall incidents in patients with a high risk of falling


Participants 200 peopleInclusion criteria aged 65 and over high risk of falling living independently or in residential home livingnear University Medical Center history of recent fallExclusion criteria unable sign informed consent or provide a fall history fall due to traffic or occupationalaccident living in nursing home acute pathology requiring long-term rehabilitation eg stroke

Interventions 1 Intervention multidisciplinary assessment in geriatric outpatient clinic and individually tailored treatmentregimen in collaboration with patientrsquos GP eg withdrawal of psychotropic drugs balance and strengthexercises home hazard reduction referral to specialists2 Control usual care

Outcomes One year follow up using fall calendarTime to first and second fallSecondary outcomes ADL quality of life physical performance adherence medication useEconomic evaluation

Starting date April 2005 to July 2008



Lips (Continued)

Contact information Prof P LipsDepartment of EndocrinologyVU University Medical CenterPO Box 7057AmsterdamThe NetherlandsEmail plipsvumcnl or gpeetersvumcnl

Notes

Lord

Trial name or title VISIBLE study (Visual Intervention Strategy Incorporating Bifocal and Long-Distance Eyeware)


Participants 580 peopleInclusion criteria using multifocal glasses outdoors 3 or more times per week community-dwelling aged65+ years with a recent fall OR aged 80+ years regardless of falls history Folstein Mini Mental score of 24+and adequate visual contrast sensitivity (Melbourne Edge Test score of 16+dB)

Interventions Assessor-blinded trialAll participants will receive an optometry assessment and updated multifocal glasses (if required) at baseline1 Intervention subjects will receive a pair of plain distance glasses and counselling for their use in predomi-nantly outdoor situations2 Control use their multifocal glasses in their usual manner

Outcomes Falls rates and compliance using monthly falls diariesSecondary outcomes Quality of life (SF-36) Instrumental Activities of Daily Living Adelaide ActivitiesIndex

Starting date June 2005 to March 2008

Contact information Prof SR LordPrince of Wales Medical Research InstituteUniversity of New South WalesRandwickSydneyNew South Wales 2031AustraliaEmailslordunsweduau

Notes



Maki

Trial name or title Evaluation of a balance-recovery specific falls prevention exercise program


Participants Inclusion criteria aged 65-80 community dwelling history of falls (at least 1 fall in the past 12 months) orpoor balance functional mobility (no dependence on mobility aids)Exclusion criteria neurological or musculoskeletal disorder cognitive disorder (eg dementia) osteoporosis

Interventions A training program involving perturbation-evoked reactions will be evaluated

Outcomes Primary outcome ability to recover balance by stepping and graspingSecondary outcome fall frequency clinical measures related to balance and fall risk (eg FallScreen Com-munity Balance and Mobility Scale balance confidence)

Starting date November 2005 to March 2008

Contact information Brian MakiPrincipal InvestigatorSunnybrook amp Womenrsquos College Health Sciences CentreUniversity of TorontoTorontoOntarioCanada

Notes Possibly laboratory induced falls while assessing balance rather than self-reported falls

Masud

Trial name or title Multifactorial day hospital intervention to reduce falls in high risk older people in primary care a multi-centre randomised controlled trial


Participants 400 people aged over 70 not resident in nursing or residential homes identified as being at high risk of fallingby a postal screening questionnaire registered with the participating general practices in Nottinghamshireand Derbyshire (UK)

Interventions 1 Intervention screening questionnaire information leaflet leaflet on falls prevention and invitation toattend the day hospital for assessment and any subsequent intervention2 Control screening questionnaire information leaflet leaflet on falls prevention and usual care from primarycare service until outcome data collected then offer of day hospital intervention

Outcomes Proportion falling during one year follow up

Starting date September 2004 to May 2006



Masud (Continued)

Contact information Prof T MasudDepartment of Rehabilitation and the Clinical Gerontology Research UnitNottingham City Hospital NHS TrustNottinghamNG5 1PBUKTelephone +44 (0)115 969 1169 x47193Email tmnchhcedemoncouk

Notes

Menz

Trial name or title Podiatry treatment to improve balance and prevent falls in older people

Methods Randomised controlled trial Simple randomisation by external telephone randomisation service

Participants Target sample size 300Inclusion criteria aged ge65 independently community dwelling ge1 falls in past year self-reported disablingfoot pain able to walk household distances without a walking aid able to read and speak basic EnglishExclusion criteria lower limb amputation (including partial foot amputation) Parkinsonrsquos disease activeplantar ulceration cognitive impairment

Interventions 1 Intervention assessment and if required footwear (assistance in purchasing more appropriate footwear) or-thoses (customised insoles to accommodate plantar lesions) home-based exercise instructions (ankle stretch-ing 1st metatarsophalangeal joint stretching toe strengthening 3x per week for 6 months) plus all partici-pants receive instructions on general foot exercises plus ldquousual carerdquo and booklet as for controls2 Control ldquousual carerdquo - general podiatric care ie nail trimming callus and corn reduction every 8 weeksfor 1 year booklet on falls

Outcomes Monthly falls calendar and phone calls Proportion of fallers and multiple fallers 12 month after baselineassessment rate of falls per person


Contact information Dr H MenzLa Trobe UniversityKinsbury DriveBundooraVictoria 3086AustraliaEmail hmenzlatrobeeduau

Notes



Miller

Trial name or title Individual nutrition therapy and exercise regime A controlled trialof injured vulnerable elderly (INTERACTIVE trial)


Participants 460 participantsInclusion criteria community-dwelling aged gt 70 in hospital after a proximal femoral fracture MMSE ge

1830 body mass index between 185 kgm2 and 35 kgm2

Exclusion criteria pathological fracture unable to give consent medically unstable 14 days after surgery

Interventions 1 Intervention six-month individualised exercise and nutrition program commencing within 14 days post-surgery Weekly home visits2 Attention control Weekly social visits

Outcomes Falls monitored at weekly visit for 6 months 12 month follow up in the community

Starting date June 2007 to September 2009

Contact information Michelle D MillerDepartment of Nutrition and DieteticsFlinders UniversityAdelaideSouth AustraliaAustraliaEmail michellemillerflinderseduau

Notes

Olde Rikkert

Trial name or title Randomized controlled trial to reduce falls incidence rate in frail elderly (CP)


Participants 160 patients referred to a geriatric outpatient clinic history of falling at least once in the last 6 months andtheir primary caregivers

Interventions A multifaceted fall prevention program for frail elders with physical and cognitive components and trainingprogram for caregivers

Outcomes Follow up for 6 months after interventionFalls incidence rateAlso numerous other secondary outcomes including fear of falling

Starting date January 2008 to July 2010



Olde Rikkert (Continued)

Contact information Dr Maria C FaesRadboud University Nijmegen Medical CentreNijmegen GelderlandNetherlands 6500 HBEmail mfaesgerumcnnl

Notes Principal investigator Prof dr M Olde Rikkert

Palvanen

Trial name or title The Chaos Clinic for prevention of falls and related injuries a randomised controlled trial

Methods Pragmatic randomised controlled trial

Participants Target sample size 3200Inclusion criteria Home-dwelling aged ge70 high-risk for falling and fall-induced injuries and fractures

Interventions 1 Intervention baseline assessment and general injury prevention brochure plus individual preventive mea-sures by Chaos Clinic staff based on baseline assessment physical activity prescription nutritional adviceindividually tailored or group exercises treatment of conditions medication review alcohol reduction smok-ing cessation hip protectors osteoporosis treatment home hazard assessment and modification2 Control baseline assessment and general injury prevention brochure alone

Outcomes Falls and fall-related injuries especially fracturesMeasured by phone calls at 3 and 9 months and on follow-up visits at 6 and 12 months from the beginning

Starting date January 2005 to December 2010

Contact information Dr M PalvanenThe Urho Kaleva Kekkonen (UKK) Institute for Health Promotion ResearchPO Box 30TampereFIN-33501Finland

Notes

Pighills

Trial name or title Environmental assessment and modification to prevent falls in older people




Pighills (Continued)

Participants 246 people recruited from 13 general practice lists in the catchment of Airedale NHS Trust (UK) Inclusioncriteria aged 70 and over with a history of at least one fall in the previous 12 months not currently receivingOT and not having had an OT environmental assessment for falls in the previous 12 months

Interventions Environmental assessment to reduce fall hazards provided by either occupational therapists or non profession-ally qualified domiciliary support workers Half of the participants receiving the environmental assessmentwill additionally receive follow through to support them in implementing recommendations

Outcomes Number of fallsTime to first fallFalls efficacy scale - International version (FES-I)SF-12 York versionEuroqol (EQ-5D)Modified Barthel Index

Starting date January 2006 to July 2007 (completed)

Contact information Alison PighillsRoom 228 Post Graduate AreaHYMS BuildingUniversity of YorkYorkYO10 5DDUKTelephone +44 1535 292706Email acp500yorkacuk

Notes

Press

Trial name or title Comprehensive interventions for falls prevention in the elderly


Participants 200 people living in Beer-Sheva and Ofakim (Israel)Inclusion criteria men and women aged 65 and over or more falls in past 12 month (self-reported) belongingto Clalit HMO living in Beer Sheva or Ofakim Israel mobile outdoors without wheelchairExclusion criteria seriously ill patients - as dyspnoea with light exercise unstable heart disease MMSE lt 18

Interventions 1 Intervention multidisciplinary assessment by geriatrician physiotherapist and OT (home hazard assess-ment) plus at least one of the following recommend medication adjustment or referral to optometrist orophthalmologist to family physician exercise sessions with physiotherapist OT advice to change unsafe homehazards2 Control usual care



Press (Continued)

Outcomes Participants to contact research assistant by phone soon after a fall Appear to be collecting fall data fromClalit and Medical Centre databasesPrimary outcome fall ratesSecondary outcomes safety cost of health care utilization and rate of hospitalisation

Starting date January 2008

Contact information Dr Yan PressBen-Gurion University of the NegevIsraelEmail yanpzahavnetil

Notes

Sanders

Trial name or title Vital D Primary care prevention of falls and fractures in the elderly by annual vitamin D supplementation


Participants 1500 ambulant women aged 70+ years on entry need to score at least 5 on algorithm (higher risk of hipfracture or low vitamin D status) Score 5 if osteoporotic fracture since the age of 50 years or rsquofrequent fallerrsquoExclusion criteria hypercalcaemia vit D supplement gt400 IUday HRT and SERM calcitriol renal disease(creatinine gt150 umolL) sarcoidosis TB or lymphoma

Interventions 1 Intervention annual oral dose of 500000 IU cholecalciferol every autumn for 5 years2 Control annual oral placebo dose

Outcomes Fall rate (monthly falls diary and phone calls) ldquotime to fallsrdquo fractures (all sites radiologically confirmed)total healthcare utilisation and mental health (depression)

Starting date 2003 to 2008

Contact information Dr Kerrie SandersClinical Research UnitDepartment Clinical and Biomedical Sciences Barwon HealthThe University of MelbourneGeelong HospitalPO Box 281Geelong 3220VictoriaAustraliaTelephone +61 3 52267834Email kerrieBarwonHealthorgau



Sanders (Continued)

Notes

Schumacher

Trial name or title Fall prevention by Alfacalcidol and training


Participants 484 men and women with chronic renal failureInclusion criteria aged 65 and over history of at least one movement-related non-syncopal fall either withinthe past year or earlier with increased fall risk identified by screening examination creatinine clearance of 30to 60 mlmin (ie moderately impaired kidney function)Exclusion criteria multiple exclusion criteria including being in an institution hypercalcaemia taking vitaminD dementia fracture or stroke in preceding 3 months etc (see ClinicalTrialsgov for details)

Interventions 1 Intervention 1microg Alfacalcidol and 500mg calcium daily mobility program (strength balance and gaittraining twice a week for one hour) patient education (single meeting with teaching lessons on risk factors forfalling and modes of fall prevention followed by an evaluation of the individual fall risk and correspondingrecommendations to reduce it)2 Control usual care

Outcomes Follow up for one year Number of fallers number of falls number of fractures fear of falling balanceperformance hypercalcaemia


Contact information Dr J SchumacherKlinik fuumlr Altersmedizin und Fruumlhrehabilitation Marienhospital Ruhr-Universitaumlt BochumHerne NRW Germany 44627Telephone +49 2323 499 0 ext 5918Email jochenschumacherrubde

Notes Open label trial sponsored by Teva Pharmaceutical Industries

Snooks

Trial name or title An evaluation of the Primary Care falls prevention services for older fallers presenting to the ambulance service


Participants 320 people aged over 65 who call for an ambulance after a fall and are not taken to hospital or are taken tohospital but not admitted People receiving a falls prevention services (in geriatric day hospitals or hospitalout-patient departments) will be excluded



Snooks (Continued)

Interventions 1 Intervention assessment by falls prevention service and interventions delivered as appropriate (six sessionsincluding physiotherapy and occupational therapy Balance training muscle strengthening reduction ofenvironmental hazards education about how to get off the floor and provision of equipment If medicalassessment required for medication check or visual problems refer to GP in first instance and then to thecommunity geriatrician if necessary2 Control no intervention by falls prevention service

Outcomes One year follow upFalls diaries returned monthly plus telephone prompts Postal assessment at 6 and 12 months (activity levelsfear of falling quality of life) service utilisationEconomic evaluation

Starting date 1 September 2005 to 31 December 2007

Contact information Dr P LoganB98 Division of Rehabilitation and AgeingMedical SchoolQMCNottinghamNG7 2UHUKTelephone +44 115 8230232Email piplogannottinghamacuk

Notes

Stuck

Trial name or title The PRO-AGE (PRevention in Older people-Assessment in GEneralistsrsquo practices) study


Participants GPs in London (UK) Hamburg (Germany) and Solothurn (Switzerland) trained in risk identification healthpromotion and prevention in older people Their consenting older patients (gt60 or 65 depending on site)randomised to intervention or controlAdditional GPs at each site did not receive the training and their eligible patients invited to participate as aconcurrent comparison groupExclusion criteria needing human assistance with basic ADL living in a nursingresidential home cognitiveimpairment terminal disease inability to speak the regional language

Interventions 1 Intervention Health Risk Appraisal for Older Persons (HRA-O) instrument feedback and site-specificintervention2 Control usual care



Stuck (Continued)

Outcomes Follow up at 1 year Sent questionnaire (HRA-O health care use and self-efficacy questions) Asked if fallenin previous year (yesno) multiple falls (yesno)

Starting date November 2000

Contact information Prof A StuckGeriatrische UniversitaumltsklinikSpital Netz Bern ZieglerMorillonstr 75-91CH-3001 BernSwitzerlandTelephone +41 31 970 73 36Email andreasstuckspitalnetzbernch

Notes International multi-centre study

Taylor

Trial name or title An evaluation of the Accident Compensation Corporation (ACC) Tai Chi programme in older adults doesit reduce falls

Methods RCT Central randomisation using specialist computer program (see httpwwwrandomizationcom) strat-ified by site and blocked to ensure balanced numbers over the three interventions

Participants Inclusion criteria men and women over 65 years (55 years if Maori or Pacific Islander) history of at least onefall in the previous 12 months or have a falls risk factor according to the Falls Risk Assessment Tool (FRAT)Exclusion criteria unable to walk independently (with or without walking aid) chronic medical condition thatwould limit participation in low-moderate exercise severe cognitive limitations (telephone Mini mental stateexamination score lt20) currently participating in an organised exercise programme of equivalent intensityas the study intervention

Interventions All training sessions are of 1 hour duration for a 20 week period1 Intervention Tai Chi training 1x week2 Intervention Tai Chi training 2X week3 Control flexibility training 1x week

Outcomes Falls at 20 weeks 6 months and 12 months

Starting date 30 August 2006

Contact information Dr Denise TaylorPhysical Rehabilitation Research CentreSchool of PhysiotherapyAuckland University of Technology (AUT)



Taylor (Continued)

Akoranga CampusNorthcoteAucklandTelephone +64 9 9219680Email denisetaylorautacnz

Notes

Tousignant

Trial name or title Falls prevention for frail older adults Cost-efficacy analysis of balance training based on Tai Chi

Methods Randomised controlled trial and economic evaluation

Participants 122 community-dwelling people aged ge 65 history of a fall in previous 6 m scoring lt4956 at the Bergtest cognitively intact (scoring gt65 at the 3MS test) able to exercise based on medical assessment

Interventions 1 Intervention Tai Chi two sessions of one hour per day for 15 weeks in groups of 4 to 6 subjects2 Control conventional physiotherapy balance training for two sessions of one hour per day for 15 weeks

Outcomes 1 year follow up1 Falls per person year2 Time to first fall3 Cost-effectiveness

Starting date 01102002 to 30062007 (Completed)

Contact information Dr Michel TousignantCentre de recherche sur le vieillissementIUGS - Pavillon DrsquoYouville1036 rue Belveacutedegravere SudSherbrookeJ1H 4C4Canada

Telephone +1 819-821-1170 (2351)Email MichelTousignantUSherbrookeca

Notes



Vind

Trial name or title Examination and treatment after a fall


Participants 400 people over 65 years treated in the emergency room or admitted to hospital after a fall

Interventions Assessment by doctor nurse and physical therapist followed by multifactorial intervention

Outcomes Primary falls and injurious fallsSecondary function health related quality of life balance confidence

Starting date September 2005 to March 2008

Contact information Dr AB VindDept of GeriatricsAmtssygehuset i GlostrupGlostrup 2600DenmarkTelephone +45 4323 4543Email anbovi01glostruphospkbhamtdk

Notes Anticipated completion date March 2008

Zeeuwe

Trial name or title The effect of Tai Chi Chuan in reducing falls among elderly people


Participants 270 community dwelling people age 70 and over identified from GPsrsquo files as having fallen in previous yearand suffering from two of the following risk factors disturbed balance mobility problems dizziness or theuse of benzodiazepines or diuretics

Interventions 1 Intervention Tai Chi Chuan (13 weeks twice a week)2 Control no treatment

Outcomes Primary falls recorded in diariesSecondary balance fear of falling blood pressure heart rate lung function parameters physical activityfunctional status quality of life mental health use of walking devices medication use of health care servicesadjustments to the house severity of fall incidents and subsequent injuries Cost-effectiveness analysis Followup at 3 6 and 12 months after randomisation

Starting date February 2004 through 2006



Zeeuwe (Continued)

Contact information Petra EM ZeeuweDepartment of General PracticeErasmus MCUniversity Medical CentreRotterdamPO Box 17383000 DR RotterdamThe NetherlandsEmail pzeeuweerasmusmcnl

Notes

Zijlstra

Trial name or title Evaluating an intervention to reduce fear of falling and associated activity restriction


Participants 360 people aged 70 and over community dwelling reporting some fear of falling and some associatedavoidance of activity

Interventions 1 Intervention cognitive behavioural group intervention designed to promote view that falls and fear of fallingare controllable set realistic goals for increasing activity modifying environment to reduce risk promoteexercise to increase strength and balance2 Control no intervention

Outcomes Primary fear of falling activity avoidance daily activitySecondary falls (falls calendar) general health satisfaction ADL anxiety depression social support loneli-ness perceived consequences of falling and risk of falling


Contact information GAR ZijlstraMaastricht UniversityFaculty of Health Medicine and Life SciencesDepartment of Health Care Studies6200 MD MaastrichtNetherlandsEmail RZijlstrazwunimaasnl

Notes

ABBREVIATIONS AND ACRONYMSAampE accident and emergency departmentADL activities of daily livingGP general practitionerIADL instrumental activities of daily living - eg use of telephone shopping housework managing finances



MMSE mini-mental state examination (cognitive assessment)OT occupational therapy



D A T A A N D A N A L Y S E S

Comparison 1 Exercise vs control

Outcome or subgroup titleNo of

studies

No of

participants Statistical method Effect size

1 Rate of falls 26 Rate ratio (Fixed 95 CI) Subtotals only11 Group exercise multiple

components vs control14 2364 Rate ratio (Fixed 95 CI) 078 [071 086]

12 Individual exercise athome multiple components vscontrol

4 666 Rate ratio (Fixed 95 CI) 066 [053 082]

13 Group exercise tai chi vscontrol


14 Group exercise gaitbalance or functional trainingvs control


15 Group exercisestrengthresistance training vscontrol


16 Individual exercise athome resistance training vscontrol


17 Individual exercisebalance training vs control


2 Number of fallers 31 Risk ratio (Random 95 CI) Subtotals only21 Group exercise multiple

categories of exercise vs control17 2492 Risk ratio (Random 95 CI) 083 [072 097]

22 Individual exercise athome multiple categories ofexercise vs control

3 566 Risk ratio (Random 95 CI) 077 [061 097]

23 Individual exercise athome multiple categories vsusual care (Parkinsonrsquos disease)


24 Individual exercisecommunity physiotherapy vscontrol (stroke)








28 Individual exercise athome resistance vs control




29 Individual exercisewalking vs control


3 Number of people sustaining afracture

5 719 Risk ratio (Fixed 95 CI) 036 [019 070]

Comparison 2 Group exercise multiple components vs control subgroup analysis by falls risk at baseline


studies

No of


1 Rate of falls 14 Rate Ratio (Random 95 CI) Subtotals only

11 Selected for higher risk offalling

8 1093 Rate Ratio (Random 95 CI) 075 [062 089]

12 Not selected for higherrisk of falling


2 Number of fallers 17 Risk Ratio (Random 95 CI) Subtotals only21 Selected for higher risk of

falling9 1139 Risk Ratio (Random 95 CI) 088 [078 099]


8 2171 Risk Ratio (Random 95 CI) 083 [062 111]

Comparison 3 Exercise vs exercise


studies

No of


1 Rate of falls 4 Rate ratio (Fixed 95 CI) Subtotals only11 Group exercise square

stepping vs walking1 68 Rate ratio (Fixed 95 CI) 070 [023 213]

12 Group exercise enhancedbalance therapy vs conventionalphysiotherapy post hip fracture


13 Group exercise balancetraining in workstations vsrsquoconventionalrsquo fall-preventionexercise class


14 Group exercise + homeexercise vs home exercise


2 Number of fallers 2 Risk ratio (Fixed 95 CI) Subtotals only21 Square stepping vs walking 1 68 Risk ratio (Fixed 95 CI) 064 [021 195]22 Group exercise + home

exercise vs home exercisemultiple components




Comparison 4 Vitamin D (with or without calcium) vs controlplacebocalcium


studies

No of


1 Rate of falls 5 3929 Rate Ratio (Random 95 CI) 095 [080 114]11 Vitamin D3 (by mouth)

vs control or placebo1 222 Rate Ratio (Random 95 CI) 112 [090 138]

12 Vitamin D3 (by mouth) +calcium vs control or placebo


13 Vitamin D3 (by mouth) +calcium vs calcium


14 Vitamin D2 (by injection)vs controlplacebo


2 Number of fallers 10 21110 Risk Ratio (Fixed 95 CI) 096 [092 101]21 Vitamin D3 (by mouth)

vs control or placebo2 2260 Risk Ratio (Fixed 95 CI) 098 [082 116]


2 3437 Risk Ratio (Fixed 95 CI) 093 [077 113]



24 Vitamin D2 (by mouth) +calcium vs calcium + placebo




26 Vitamin D (oral or IM)with or without calcium vscontrol studies with multiplearms combined




31 Vitamin D3 (by mouth)vs control or placebo










4 Number of people sustainingadverse effects

3 Risk Ratio (M-H Fixed 95 CI) Subtotals only

41 Hypercalcaemia 3 5744 Risk Ratio (M-H Fixed 95 CI) 170 [073 396]42 Renal disease (renal stones

and renal insufficiency)1 5292 Risk Ratio (M-H Fixed 95 CI) 057 [017 195]

43 Gastrointestinal effects 2 5594 Risk Ratio (M-H Fixed 95 CI) 091 [075 110]



Comparison 5 Vitamin D (with or without calcium) vs control subgroup analysis by falls risk at baseline


studies

No of


1 Rate of falls 5 Rate Ratio (Random 95 CI) Subtotals only11 Selected for higher risk of

falling2 3125 Rate Ratio (Random 95 CI) 087 [058 130]



2 Number of fallers 10 Risk Ratio (Fixed 95 CI) Subtotals only21 Selected for higher risk of

falling5 8838 Risk Ratio (Fixed 95 CI) 093 [083 103]



Comparison 6 Vitamin D (with or without calcium) vs control subgroup analysis by vitamin D level at baseline


studies

No of


1 Rate of falls 5 Rate Ratio (Random 95 CI) Subtotals only11 Selected for low vitamin

D level2 260 Rate Ratio (Random 95 CI) 057 [037 089]

12 Not selected for lowvitamin D level


2 Number of fallers 10 Risk Ratio (Fixed 95 CI) Subtotals only21 Selected for low vitamin

D level3 562 Risk Ratio (Fixed 95 CI) 065 [046 091]



Comparison 7 Any vitamin D analogue vs controlplacebo


studies

No of


1 Rate of falls 2 Rate Ratio (Fixed 95 CI) Subtotals only11 Alfacalcidol (vitamin D

analogue) vs placebo1 80 Rate Ratio (Fixed 95 CI) 108 [075 157]

12 Calcitriol (vitamin Danalogue) vs placebo

1 213 Rate Ratio (Fixed 95 CI) 064 [049 082]

2 Number of fallers 2 Risk Ratio (Fixed 95 CI) Subtotals only21 Alfacalcidol (vitamin D

analogue) vs placebo1 378 Risk Ratio (Fixed 95 CI) 069 [041 117]






2 Risk Ratio (Fixed 95 CI) Subtotals only

31 Alfacalcidol (vitamin Danalogue) vs placebo






41 Hypercalcaemia 2 624 Risk Ratio (M-H Fixed 95 CI) 233 [102 531]42 Renal disease (kidney

stone)1 246 Risk Ratio (M-H Fixed 95 CI) 033 [001 810]


Comparison 8 Medication (drug target) other than vitamin D vs control


studies

No of


1 Rate of falls 2 Rate ratio (Fixed 95 CI) Subtotals only11 Psychotropic medication

withdrawal vs control1 93 Rate ratio (Fixed 95 CI) 034 [016 073]

12 Hormone replacementtherapy vs placebo


2 Number of fallers 5 Risk ratio (Fixed 95 CI) Subtotals only21 Psychotropic medication

withdrawal vs control1 93 Risk ratio (Fixed 95 CI) 061 [032 117]

22 Hormone replacementtherapy vs controlplacebo


23 Medication review andmodification vs usual care


24 GP educationalprogramme and medicationreview and modification vscontrol




31 Psychotropic medicationwithdrawal vs control




Comparison 9 Surgery vs control


studies

No of


1 Rate of falls 3 Rate Ratio (Fixed 95 CI) Subtotals only11 Cardiac pacing vs control 1 171 Rate Ratio (Fixed 95 CI) 042 [023 075]12 Cataract surgery (1st eye)

vs control1 306 Rate Ratio (Fixed 95 CI) 066 [045 095]

13 Cataract surgery (2nd eye)vs control


2 Number of fallers 2 Risk Ratio (Fixed 95 CI) Subtotals only21 Cataract surgery (1st eye)

vs control1 306 Risk Ratio (Fixed 95 CI) 095 [068 133]





31 Cardiac pacing vs control 1 171 Risk Ratio (Fixed 95 CI) 078 [018 339]32 Cataract surgery (1st eye)




Comparison 10 Fluid or nutrition therapy vs control


studies

No of


1 Number of fallers 1 Risk ratio (Fixed 95 CI) Subtotals only11 Nutritional

supplementation vs control1 46 Risk ratio (Fixed 95 CI) 010 [001 131]

Comparison 11 Psychological interventions vs control


studies

No of


1 Number of fallers 1 Risk ratio (Fixed 95 CI) Subtotals only11 Cognitive behavioural

intervention vs control1 230 Risk ratio (Fixed 95 CI) 113 [079 160]



Comparison 12 Environmentassistive technology interventions vs control


studies

No of


1 Rate of falls 6 Rate ratio (Fixed 95 CI) Subtotals only11 Home safety intervention

vs control3 2367 Rate ratio (Fixed 95 CI) 090 [079 103]

12 Home safety interventionvs no home safety (severe visualimpairment)


13 Vision assessment and eyeexamination + intervention(with or without referral) vscontrol


14 Anti-slip shoe device foricy conditions vs control


2 Number of fallers 7 Risk Ratio (Fixed 95 CI) Subtotals only

21 Home safety interventionvs control






24 Visual acuity assessmentand referral vs control






Comparison 13 Environmentassistive technology interventions vs control subgroup analysis by risk of falling

at baseline


studies

No of


1 Rate of falls 4 Rate ratio (Fixed 95 CI) Subtotals only11 Selected for higher risk of

falling2 491 Rate ratio (Fixed 95 CI) 056 [042 076]










Comparison 14 Knowledgeeducation interventions vs control


studies

No of


1 Rate of falls 1 Rate ratio (Fixed 95 CI) Subtotals only

11 Education interventionsvs control


2 Number of fallers 2 Risk ratio (Fixed 95 CI) Subtotals only21 Education interventions

vs control2 516 Risk ratio (Fixed 95 CI) 073 [052 103]

Comparison 15 Multiple interventions


studies

No of


1 Rate of falls 6 Rate ratio (Fixed 95 CI) Subtotals only11 Exercise + vitamin D vs no

exerciseno vitamin D (severevisual impairment)


12 Exercise + ldquoindividualisedfall prevention advicerdquo vscontrol


13 Exercise + education + riskassessment vs control


14 Exercise + education +home safety vs control


15 Exercise + nutrition +calcium + vit D vs calcium +vit D


16 Exercise + education vseducation


17 Exercise + home safety +education vs education


18 Exercise + home safety +education + clinical assessmentvs education


2 Number of fallers 7 Risk Ratio (Fixed 95 CI) Subtotals only21 Exercise + home safety vs

control1 272 Risk Ratio (Fixed 95 CI) 076 [060 097]



22 Exercise + visionassessment vs control


23 Exercise + visionassessment + home safety vscontrol




25 Education + exercise +home safety vs control


26 Exercise + vitamin D vsno exerciseno vitamin D


27 Home safety + medicationreview vs control


28 Home safety + visionassessment vs control


29 Education + free access togeriatric clinic vs control








Comparison 16 Multifactorial intervention after assessment vs control


studies

No of


1 Rate of falls 15 8141 Rate ratio (Random 95 CI) 075 [065 086]2 Number of fallers 26 11173 Risk ratio (Random 95 CI) 095 [088 102]3 Number of people sustaining a

fracture7 2195 Risk Ratio (Fixed 95 CI) 070 [047 104]

Comparison 17 Multifactorial intervention after assessment vs control subgroup analysis by falls risk at baseline


studies

No of


1 Rate of falls 15 Rate ratio (Random 95 CI) Subtotals only11 Selected for higher risk of

falling13 4592 Rate ratio (Random 95 CI) 076 [064 091]


2 3549 Rate ratio (Random 95 CI) 057 [023 138]

2 Number of fallers 26 Risk ratio (Fixed 95 CI) Subtotals only







Comparison 18 Multifactorial intervention after assessment vs control subgroup analysis by intensity of inter-

vention


studies

No of


1 Rate of falls 14 Rate ratio (Random 95 CI) Subtotals only11 Assessment and active

intervention7 5314 Rate ratio (Random 95 CI) 070 [055 090]

12 Assessment and referral orprovision of information


2 Number of fallers 26 Risk ratio (Random 95 CI) Subtotals only21 Assessment and active

intervention10 6040 Risk ratio (Random 95 CI) 093 [084 103]



23 Unclassifiable 1 0 Risk ratio (Random 95 CI) Not estimable

F E E D B A C K

Definition of terms 26 June 2009

Summary

Please could you clarify the definitions of falls risk and rate of falls How do they differ from one another

Reply

We are unclear as to whether the question relates to ldquofalls riskrdquo or whether Dr Foley is actually meaning ldquorisk of fallingrdquoIn the review the term falls risk is used in relation to falls risk at enrolment In subgroup analyses we compared trials with participantsat higher versus lower falls risk at enrolment (ie comparing trials with participants selected for inclusion based on history of fallingor other specific risk factors for falling versus unselected) (see Data collection and analysis lsquoSubgroup analyses and investigation ofheterogeneityrsquo)The review reports two primary outcomes1 Rate of falls

This is the number of falls over a period of time for example number of falls per person year The statistic used to report this is therate ratio which compares the rate of events (falls) in the two groups during the trial or during a number of trials if the data are pooledBased on these statistics we report whether an intervention has a significant effect on the rate of falls2 Number of people falling during follow up

The statistic used to report this is the risk ratio which compares the number of participants in each group with one or more fall eventsduring the trial or during a number of trials if the data are pooled Based on these statistics we report whether an intervention has asignificant effect on the risk of fallingFor further details please refer to the Methods section in the review lsquoData relating to rate of fallsrsquo and lsquoData relating to number offallers or participants with fall-related fracturesrsquo



Contributors

Comment from Dr Charlotte Foley UKReply from Mrs Lesley Gillespie New Zealand

W H A T rsquo S N E W

Last assessed as up-to-date 7 October 2008

10 August 2009 Feedback has been incorporated Feedback added to clarify terms used

H I S T O R Y

Protocol first published Issue 2 2008

Review first published Issue 2 2009

13 May 2009 Amended Correction of several typographical errors

27 October 2008 Amended Converted to new review format

19 February 2008 Amended The published review ldquoInterventions for preventing falls in elderly peoplerdquo (Gillespie 2003) is notbeing updated Due to its size and complexity it is being split into two reviews ldquoInterventions forpreventing falls in older people living in the communityrdquo and ldquoInterventions for preventing falls inolder people in residential care facilities and hospitalsrdquo

C O N T R I B U T I O N S O F A U T H O R S

LD Gillespie the guarantor for this review conceived designed and coordinated the review developed the search strategy and carriedout the searches screened search results and obtained papers screened retrieved papers against inclusion criteria carried out qualityassessment and data extraction entered data into RevMan and wrote the review

MC Robertson contributed to the appraisal of quality extracted data from papers managed data and carried out statistical calculationswrote the economic evaluation section and Appendix 4 and commented on drafts of the review In addition she provided additionaldata about papers and a methodological perspective for measurement of outcomes and statistical analyses used in the papers and theeconomic evaluations

WJ Gillespie conceived and designed the review screened retrieved papers against inclusion criteria carried out quality assessment anddata extraction entered data into RevMan and wrote the review

SE Lamb conceived and led the design of the ProFaNE taxonomy that provided the framework for the structure of the review carriedout quality assessment and data extraction and commented on drafts of the review

S Gates provided statistical advice carried out quality assessment and data extraction and commented on drafts of the review

RG Cumming and BH Rowe carried out data extraction and quality assessment and commented on drafts of the review



D E C L A R A T I O N S O F I N T E R E S T

Three reviewers were investigators for eight included studies RG Cumming (Cumming 1999 Cumming 2007) WJ Gillespie (Carter1997) and MC Robertson (Campbell 1997 Campbell 1999c Campbell 2005 Elley 2008 Robertson 2001a) Investigators did notcarry out quality assessment on their own studies No other conflicts are declared

S O U R C E S O F S U P P O R T

Internal sources

bull University of Otago Dunedin New ZealandComputing administration and library services (MCR LDG)

External sources

bull Government of Canada Canada Research Chairs Program Ottawa CanadaSalary (BR)

bull Accident Compensation Corporation (ACC) New ZealandSalary (MCR)

D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W

Risk of bias assessment

The protocol was completed and submitted for publication prior to the general release of RevMan 5 and the supporting version of thersquoCochrane Handbook for Systematic Reviews of Interventionsrsquo (version 50) in February 2008 In the protocol we stated that we wouldassess methodological quality using the 11 item tool used in Gillespie 2003 Rather than use that tool we made a post hoc decision toconvert a number of these items for use in the new Cochrane Collaboration tool for assessing risk of bias (Higgins 2008a) and planto add additional items in future versions of the review

N O T E S

The published review ldquoInterventions for preventing falls in elderly peoplerdquo (Gillespie 2003) has been withdrawn from The CochraneLibrary Due to its size and complexity it has been split into two reviews this review and ldquoInterventions for preventing falls in olderpeople in residential care facilities and hospitalsrdquo which is nearing completion



I N D E X T E R M S

Medical Subject Headings (MeSH)

Accidental Falls [lowastprevention amp control] Accidents Home [lowastprevention amp control] Bone Density Conservation Agents [administrationamp dosage] Environment Design Exercise Patient Education as Topic Randomized Controlled Trials as Topic Tai Ji Vitamin D[administration amp dosage]

MeSH check words

Aged Humans



T A B L E O F C O N T E N T S

1HEADER 1ABSTRACT 2PLAIN LANGUAGE SUMMARY 3BACKGROUND 3OBJECTIVES 3METHODS 6RESULTS

Figure 1 13Figure 2 14Figure 3 20

21DISCUSSION 24AUTHORSrsquo CONCLUSIONS 25ACKNOWLEDGEMENTS 26REFERENCES 47CHARACTERISTICS OF STUDIES

185DATA AND ANALYSES 194FEEDBACK 195WHATrsquoS NEW 195HISTORY 195CONTRIBUTIONS OF AUTHORS 195DECLARATIONS OF INTEREST 196SOURCES OF SUPPORT 196DIFFERENCES BETWEEN PROTOCOL AND REVIEW 196NOTES 196INDEX TERMS

iInterventions for preventing falls in older people living in the community (Review)









A B S T R A C T

Background


Objectives


Search strategy


Selection criteria




Main results

























B A C K G R O U N D








O B J E C T I V E S


M E T H O D S


Types of studies










Primary outcomes


Secondary outcomes



Electronic searches



















related fractures






Data synthesis












Economics issues


R E S U L T S





Included studies


Design


Sample sizes


Setting

Location


Sampling frame





Participants



Interventions




Exercises






Study ID Gait bal-

ancefunctional

training

Strength re-

sistance

training


dance etc)

General phys-

ical activity

Endurance Other

Ashburn 2007

Ballard 2004

Barnett 2003

Brown 2002

Buchner1997a

Bunout 2005

Campbell1997

Campbell1999

Carter 2002

Cerny 1998

Cornillon2002

Day 2002

Fiatarone1997





Hauer 2001

Helbostad2004

Korpelainen2006

dance stamping

Latham 2003

Li 2005

Lin 2007

Liu-Ambrose2004



Lord 1995

Lord 2003 dance

Luukinen2007

self care

McMurdo1997

Means 2005

Morgan 2004

Nitz 2004

Pereira 1998


standupstep up

Resnick 2002

Robertson2001a




Rubenstein2000

Sherrington2004

Shigematsu2008


walkinggroup

Skelton 2005

Steadman2003

Suzuki 2004

Voukelatos2007

Weerdesteyn2006


Tai Chigroup

Wolf 2003


Tai Chigroup









Surgery




Psychological










Outcomes





Excluded studies



Ongoing studies












Allocation




Blinding








Exercises

















rion


747 (SD 383) NA NA 829 (SD 449) 664 (SD 317) No


267 (range 255to 280)

250 (range 237to 261)


Dukas 2004 726 (SD 279) 746 (SD 290)

706 (SD 267) NA NA No

Gallagher 2001 793 (SD 247) 780 (SD216)

805 (SD 274) NA NA No



29 (range 6 to85)

30 (range 12 to64)


Latham 2003 374 (95 CI349 to 449)

474 (95 CI399 to 524)

NA NA No


















Surgery



Cataract surgery




Psychological









signalling)




protection)




















D I S C U S S I O N





Exercises













problems

Poor vision




Post hip fracture


















Exercise





Vitamin D





Exercise












Poor vision

















R E F E R E N C E S











(4)407ndash14 [PUBMED 12851185]






(10)670ndash6 [PUBMED 9278463]



199341321ndash5 [PUBMED 8440857]



200534(1)80ndash3 [PUBMED 15591487]


































(2)175ndash81 [PUBMED 17302652]











































200433(1)45ndash51 [MEDLINE 14695863]































199881(5)352ndash6 [PUBMED 9623035]
































(4)489ndash98 [PUBMED 17397425]





































(6)40ndash9 [PUBMED 12071273]



















































































(11)1800ndash6 [PUBMED 7977921]

























200755(10)1548ndash56 [MEDLINE 17908058]


















200726(2)309ndash16 [MEDLINE 17074485]








































(2)157ndash63 [MEDLINE 15977465]







200081(3)319ndash23




























































































































































































AGSBGS 2001


Beswick 2008


Bischoff 2003


Boutron 2008




Broe 2007


Buchner 1993


Cameron 2005


Campbell 1990


Campbell 1999c


Campbell 2004


2004328(7441)702ndash8 [PUBMED 15031246]

Campbell 2005


Campbell 2006


Suppl 2ii60ndash4

Campbell 2007


Chapuy 2002



Close 2000


Cummings 1995


Excel


Findorff 2007


Flicker 2005



Gates 2008


Gillespie 2003


Goodwin 2008


200823(5)631ndash40 [MEDLINE 18181210]

Haas 2006


Hauer 2006


Higgins 2003


Higgins 2008a




Higgins 2008b


Jackson 2007


2007100(4)185ndash92 [MEDLINE 17308327]

Keene 1993


Kellogg 1987


Lamb 2005



Lamb 2007


Lefebvre 2008


Lord 2008


McAlister 2003


(19)2545ndash53 [MEDLINE 12759326]

RevMan 5


Richy 2008


Rizzo 1996


Robertson 2001c



Robertson 2001d


Robertson 2007



Sach 2007


Salkeld 2000


Sattin 1992


Sherrington 2008


Smeeth 2002


Stata


Tinetti 1988


Tinetti 1997




Vellas 1997


199726(3)189ndash93 [MEDLINE 9223714]

Zecevic 2006



Gillespie 2008







Ashburn 2007





Notes

Risk of bias




BlindingFalls


BlindingFractures





Assantachai 2002





Notes

Risk of bias




BlindingFalls



Ballard 2004









Notes

Risk of bias




BlindingFalls



Barnett 2003









Notes

Risk of bias




BlindingFalls












Notes

Risk of bias




BlindingFalls


BlindingFractures



Brown 2002








Notes

Risk of bias




BlindingFalls



Buchner 1997a










Risk of bias




BlindingFalls






Bunout 2005






Risk of bias




BlindingFalls





Campbell 1997






Risk of bias




BlindingFalls





Campbell 1999






Risk of bias







BlindingFalls



Campbell 2005






Risk of bias




BlindingFalls






Carpenter 1990





Notes

Risk of bias




BlindingFalls





Carter 1997






Risk of bias




BlindingFalls



Carter 2002









Notes

Risk of bias




BlindingFalls



Cerny 1998









Risk of bias




BlindingFalls



Clemson 2004









Risk of bias




BlindingFalls



Close 1999





Notes




Risk of bias




BlindingFalls



Coleman 1999





Notes

Risk of bias







BlindingFalls



Cornillon 2002





Notes

Risk of bias




BlindingFalls





Cumming 1999





Notes

Risk of bias




BlindingFalls



Cumming 2007









Notes

Risk of bias




BlindingFalls


BlindingFractures



Davison 2005










Risk of bias




BlindingFalls



Day 2002









Notes

Risk of bias




BlindingFalls





Dhesi 2004






Risk of bias




BlindingFalls



Dukas 2004









Notes

Risk of bias




BlindingFalls



Elley 2008









Notes

Risk of bias




BlindingFalls



Fabacher 1994








Notes

Risk of bias




BlindingFalls



Fiatarone 1997





Notes Abstract only

Risk of bias







BlindingFalls



Foss 2006





Notes

Risk of bias







BlindingFalls


BlindingFractures



Gallagher 1996





Notes

Risk of bias




BlindingFalls





Gallagher 2001





Notes

Risk of bias




BlindingFalls





Grant 2005





Notes

Risk of bias




BlindingFalls


BlindingFractures





Gray-Donald 1995





Notes

Risk of bias




BlindingFalls



Green 2002









Notes

Risk of bias




BlindingFalls



Greenspan 2005










Risk of bias




BlindingFalls



Harwood 2004










Risk of bias




BlindingFalls


BlindingFractures



Harwood 2005








Notes

Risk of bias




BlindingFalls


BlindingFractures



Hauer 2001








Notes

Risk of bias




BlindingFalls



Helbostad 2004








Notes

Risk of bias




BlindingFalls



Hendriks 2008








Notes

Risk of bias




BlindingFalls



Hill 2000





Notes




Risk of bias




BlindingFalls



Hogan 2001





Notes

Risk of bias







BlindingFalls


BlindingFractures



Hornbrook 1994





Notes

Risk of bias







BlindingFalls


BlindingFractures



Huang 2004





Notes

Risk of bias







BlindingFalls



Huang 2005









Risk of bias




BlindingFalls



Jitapunkul 1998





Notes

Risk of bias







BlindingFalls



Kenny 2001






Risk of bias







BlindingFalls


BlindingFractures



Kingston 2001





Notes

Risk of bias







BlindingFalls



Korpelainen 2006





Notes

Risk of bias




BlindingFalls





BlindingFractures



Lannin 2007





Notes Pilot study

Risk of bias




BlindingFalls





Latham 2003






Risk of bias




BlindingFalls





Li 2005





Notes

Risk of bias




BlindingFalls



Lightbody 2002










Risk of bias




BlindingFalls



Lin 2007








Notes

Risk of bias




BlindingFalls



Liu-Ambrose 2004








Notes

Risk of bias




BlindingFalls



Lord 1995








Notes

Risk of bias




BlindingFalls



Lord 2003










Risk of bias




BlindingFalls



Lord 2005








Notes

Risk of bias




BlindingFalls



Luukinen 2007





Notes

Risk of bias







BlindingFalls



Mahoney 2007





Notes

Risk of bias







BlindingFalls



McKiernan 2005





Notes

Risk of bias




BlindingFalls






McMurdo 1997





Notes

Risk of bias




BlindingFalls


BlindingFractures





Means 2005





Notes

Risk of bias




BlindingFalls



Meredith 2002









Notes

Risk of bias




BlindingFalls



Morgan 2004










Risk of bias




BlindingFalls



Newbury 2001









Risk of bias




BlindingFalls



Nikolaus 2003









Risk of bias




BlindingFalls



Nitz 2004








Notes

Risk of bias




BlindingFalls



Pardessus 2002





Notes

Risk of bias







BlindingFalls



Pereira 1998





Notes

Risk of bias




BlindingFalls





Pfeifer 2000





Notes

Risk of bias




BlindingFalls





Pit 2007





Notes

Risk of bias




BlindingFalls



Porthouse 2005








Notes

Risk of bias




BlindingFalls





Prince 2008





Notes

Risk of bias




BlindingFalls





Reinsch 1992






Risk of bias




BlindingFalls



Resnick 2002










Risk of bias




BlindingFalls



Robertson 2001a










Risk of bias




BlindingFalls


BlindingFractures



Robson 2003










Risk of bias




BlindingFalls



Rubenstein 2000









Notes

Risk of bias




BlindingFalls



Rubenstein 2007









Notes

Risk of bias




BlindingFalls



Ryan 1996









Risk of bias




BlindingFalls



Salminen 2008









Notes

Risk of bias




BlindingFalls



Sato 1999









Notes

Risk of bias




BlindingFalls


BlindingFractures












Risk of bias




BlindingFalls



Sherrington 2004









Risk of bias




BlindingFalls



Shigematsu 2008





Notes

Risk of bias







BlindingFalls



Shumway-Cook 2007





Notes




Risk of bias




BlindingFalls



Skelton 2005





Notes

Risk of bias







BlindingFalls



Smith 2007





Notes

Risk of bias




BlindingFalls


BlindingFractures






Speechley 2008





Notes

Risk of bias




BlindingFalls





Spice 2009






Risk of bias




BlindingFalls



Steadman 2003









Notes

Risk of bias




BlindingFalls



Steinberg 2000









Risk of bias




BlindingFalls



Stevens 2001










Risk of bias




BlindingFalls



Suzuki 2004









Notes

Risk of bias




BlindingFalls


BlindingFractures



Swanenburg 2007









Notes Pilot study

Risk of bias




BlindingFalls



Tinetti 1994










Risk of bias




BlindingFalls


BlindingFractures



Trivedi 2003










Risk of bias




BlindingFalls


BlindingFractures



Van Haastregt 2000









Notes

Risk of bias




BlindingFalls



Van Rossum 1993








Notes

Risk of bias




BlindingFalls



Vellas 1991





Notes




Risk of bias




BlindingFalls



Vetter 1992





Notes

Risk of bias







BlindingFalls


BlindingFractures



Voukelatos 2007





Notes

Risk of bias







BlindingFalls



Wagner 1994






Risk of bias




BlindingFalls






Weerdesteyn 2006





Notes

Risk of bias




BlindingFalls






Whitehead 2003






Risk of bias




BlindingFalls






Wilder 2001





Notes Abstract only

Risk of bias




BlindingFalls





Wolf 1996






Risk of bias




BlindingFalls





Wolf 2003






Risk of bias




BlindingFalls





Woo 2007





Notes

Risk of bias




BlindingFalls



Wyman 2005









Notes

Risk of bias




BlindingFalls
















(Continued)
















(Continued)
















(Continued)















(Continued)


















(Continued)





Beyer 2007






Di Monaco 2008









Madureira 2007






Pfeifer 2004






Sato 2005b








Notes

Weber 2008





Notes




Behrman









Blalock







Blalock (Continued)




Notes

Ciaschini









Cryer





Cryer (Continued)







Donaldson












Edwards










Edwards (Continued)


Grove








Notes

Haines







Haines (Continued)




Notes

Hill a










Hill a (Continued)

Notes

Hill b









Jee







Jee (Continued)


Starting date 2005


Notes

Johnson










Johnson (Continued)


Notes

Kenny











Klaber Moffett








Notes

Lesser






Lesser (Continued)





Notes

Lips









Lips (Continued)


Notes

Lord








Notes



Maki









Masud









Masud (Continued)


Notes

Menz








Notes



Miller










Notes

Olde Rikkert












Palvanen








Notes

Pighills











Notes

Press







Press (Continued)




Notes

Sanders










Sanders (Continued)

Notes

Schumacher









Snooks






Snooks (Continued)





Notes

Stuck







Stuck (Continued)





Taylor










Taylor (Continued)


Notes

Tousignant









Notes



Vind









Zeeuwe









Zeeuwe (Continued)


Notes

Zijlstra








Notes










studies

No of








































studies

No of













studies

No of

















studies

No of











































studies

No of












studies

No of












studies

No of

























studies

No of






















studies

No of


















studies

No of






studies

No of








studies

No of
























at baseline


studies

No of















studies

No of









studies

No of















































studies

No of






studies

No of














vention


studies

No of











F E E D B A C K


Summary


Reply






Contributors





H I S T O R Y


















Internal sources


External sources






N O T E S




I N D E X T E R M S



MeSH check words

Aged Humans










A B S T R A C T

Background


Objectives


Search strategy


Selection criteria




Main results

























B A C K G R O U N D








O B J E C T I V E S


M E T H O D S


Types of studies










Primary outcomes


Secondary outcomes



Electronic searches



















related fractures






Data synthesis












Economics issues


R E S U L T S





Included studies


Design


Sample sizes


Setting

Location


Sampling frame





Participants



Interventions




Exercises






Study ID Gait bal-

ancefunctional

training

Strength re-

sistance

training


dance etc)

General phys-

ical activity

Endurance Other

Ashburn 2007

Ballard 2004

Barnett 2003

Brown 2002

Buchner1997a

Bunout 2005

Campbell1997

Campbell1999

Carter 2002

Cerny 1998

Cornillon2002

Day 2002

Fiatarone1997





Hauer 2001

Helbostad2004

Korpelainen2006

dance stamping

Latham 2003

Li 2005

Lin 2007

Liu-Ambrose2004



Lord 1995

Lord 2003 dance

Luukinen2007

self care

McMurdo1997

Means 2005

Morgan 2004

Nitz 2004

Pereira 1998


standupstep up

Resnick 2002

Robertson2001a




Rubenstein2000

Sherrington2004

Shigematsu2008


walkinggroup

Skelton 2005

Steadman2003

Suzuki 2004

Voukelatos2007

Weerdesteyn2006


Tai Chigroup

Wolf 2003


Tai Chigroup









Surgery




Psychological










Outcomes





Excluded studies



Ongoing studies












Allocation




Blinding








Exercises

















rion


747 (SD 383) NA NA 829 (SD 449) 664 (SD 317) No


267 (range 255to 280)

250 (range 237to 261)


Dukas 2004 726 (SD 279) 746 (SD 290)

706 (SD 267) NA NA No

Gallagher 2001 793 (SD 247) 780 (SD216)

805 (SD 274) NA NA No



29 (range 6 to85)

30 (range 12 to64)


Latham 2003 374 (95 CI349 to 449)

474 (95 CI399 to 524)

NA NA No


















Surgery



Cataract surgery




Psychological









signalling)




protection)




















D I S C U S S I O N





Exercises













problems

Poor vision




Post hip fracture


















Exercise





Vitamin D





Exercise












Poor vision

















R E F E R E N C E S











(4)407ndash14 [PUBMED 12851185]






(10)670ndash6 [PUBMED 9278463]



199341321ndash5 [PUBMED 8440857]



200534(1)80ndash3 [PUBMED 15591487]


































(2)175ndash81 [PUBMED 17302652]











































200433(1)45ndash51 [MEDLINE 14695863]































199881(5)352ndash6 [PUBMED 9623035]
































(4)489ndash98 [PUBMED 17397425]





































(6)40ndash9 [PUBMED 12071273]



















































































(11)1800ndash6 [PUBMED 7977921]

























200755(10)1548ndash56 [MEDLINE 17908058]


















200726(2)309ndash16 [MEDLINE 17074485]








































(2)157ndash63 [MEDLINE 15977465]







200081(3)319ndash23




























































































































































































AGSBGS 2001


Beswick 2008


Bischoff 2003


Boutron 2008




Broe 2007


Buchner 1993


Cameron 2005


Campbell 1990


Campbell 1999c


Campbell 2004


2004328(7441)702ndash8 [PUBMED 15031246]

Campbell 2005


Campbell 2006


Suppl 2ii60ndash4

Campbell 2007


Chapuy 2002



Close 2000


Cummings 1995


Excel


Findorff 2007


Flicker 2005



Gates 2008


Gillespie 2003


Goodwin 2008


200823(5)631ndash40 [MEDLINE 18181210]

Haas 2006


Hauer 2006


Higgins 2003


Higgins 2008a




Higgins 2008b


Jackson 2007


2007100(4)185ndash92 [MEDLINE 17308327]

Keene 1993


Kellogg 1987


Lamb 2005



Lamb 2007


Lefebvre 2008


Lord 2008


McAlister 2003


(19)2545ndash53 [MEDLINE 12759326]

RevMan 5


Richy 2008


Rizzo 1996


Robertson 2001c



Robertson 2001d


Robertson 2007



Sach 2007


Salkeld 2000


Sattin 1992


Sherrington 2008


Smeeth 2002


Stata


Tinetti 1988


Tinetti 1997




Vellas 1997


199726(3)189ndash93 [MEDLINE 9223714]

Zecevic 2006



Gillespie 2008







Ashburn 2007





Notes

Risk of bias




BlindingFalls


BlindingFractures





Assantachai 2002





Notes

Risk of bias




BlindingFalls



Ballard 2004









Notes

Risk of bias




BlindingFalls



Barnett 2003









Notes

Risk of bias




BlindingFalls












Notes

Risk of bias




BlindingFalls


BlindingFractures



Brown 2002








Notes

Risk of bias




BlindingFalls



Buchner 1997a










Risk of bias




BlindingFalls






Bunout 2005






Risk of bias




BlindingFalls





Campbell 1997






Risk of bias




BlindingFalls





Campbell 1999






Risk of bias







BlindingFalls



Campbell 2005






Risk of bias




BlindingFalls






Carpenter 1990





Notes

Risk of bias




BlindingFalls





Carter 1997






Risk of bias




BlindingFalls



Carter 2002









Notes

Risk of bias




BlindingFalls



Cerny 1998









Risk of bias




BlindingFalls



Clemson 2004









Risk of bias




BlindingFalls



Close 1999





Notes




Risk of bias




BlindingFalls



Coleman 1999





Notes

Risk of bias







BlindingFalls



Cornillon 2002





Notes

Risk of bias




BlindingFalls





Cumming 1999





Notes

Risk of bias




BlindingFalls



Cumming 2007









Notes

Risk of bias




BlindingFalls


BlindingFractures



Davison 2005










Risk of bias




BlindingFalls



Day 2002









Notes

Risk of bias




BlindingFalls





Dhesi 2004






Risk of bias




BlindingFalls



Dukas 2004









Notes

Risk of bias




BlindingFalls



Elley 2008









Notes

Risk of bias




BlindingFalls



Fabacher 1994








Notes

Risk of bias




BlindingFalls



Fiatarone 1997





Notes Abstract only

Risk of bias







BlindingFalls



Foss 2006





Notes

Risk of bias







BlindingFalls


BlindingFractures



Gallagher 1996





Notes

Risk of bias




BlindingFalls





Gallagher 2001





Notes

Risk of bias




BlindingFalls





Grant 2005





Notes

Risk of bias




BlindingFalls


BlindingFractures





Gray-Donald 1995





Notes

Risk of bias




BlindingFalls



Green 2002









Notes

Risk of bias




BlindingFalls



Greenspan 2005










Risk of bias




BlindingFalls



Harwood 2004










Risk of bias




BlindingFalls


BlindingFractures



Harwood 2005








Notes

Risk of bias




BlindingFalls


BlindingFractures



Hauer 2001








Notes

Risk of bias




BlindingFalls



Helbostad 2004








Notes

Risk of bias




BlindingFalls



Hendriks 2008








Notes

Risk of bias




BlindingFalls



Hill 2000





Notes




Risk of bias




BlindingFalls



Hogan 2001





Notes

Risk of bias







BlindingFalls


BlindingFractures



Hornbrook 1994





Notes

Risk of bias







BlindingFalls


BlindingFractures



Huang 2004





Notes

Risk of bias







BlindingFalls



Huang 2005









Risk of bias




BlindingFalls



Jitapunkul 1998





Notes

Risk of bias







BlindingFalls



Kenny 2001






Risk of bias







BlindingFalls


BlindingFractures



Kingston 2001





Notes

Risk of bias







BlindingFalls



Korpelainen 2006





Notes

Risk of bias




BlindingFalls





BlindingFractures



Lannin 2007





Notes Pilot study

Risk of bias




BlindingFalls





Latham 2003






Risk of bias




BlindingFalls





Li 2005





Notes

Risk of bias




BlindingFalls



Lightbody 2002










Risk of bias




BlindingFalls



Lin 2007








Notes

Risk of bias




BlindingFalls



Liu-Ambrose 2004








Notes

Risk of bias




BlindingFalls



Lord 1995








Notes

Risk of bias




BlindingFalls



Lord 2003










Risk of bias




BlindingFalls



Lord 2005








Notes

Risk of bias




BlindingFalls



Luukinen 2007





Notes

Risk of bias







BlindingFalls



Mahoney 2007





Notes

Risk of bias







BlindingFalls



McKiernan 2005





Notes

Risk of bias




BlindingFalls






McMurdo 1997





Notes

Risk of bias




BlindingFalls


BlindingFractures





Means 2005





Notes

Risk of bias




BlindingFalls



Meredith 2002









Notes

Risk of bias




BlindingFalls



Morgan 2004










Risk of bias




BlindingFalls



Newbury 2001









Risk of bias




BlindingFalls



Nikolaus 2003









Risk of bias




BlindingFalls



Nitz 2004








Notes

Risk of bias




BlindingFalls



Pardessus 2002





Notes

Risk of bias







BlindingFalls



Pereira 1998





Notes

Risk of bias




BlindingFalls





Pfeifer 2000





Notes

Risk of bias




BlindingFalls





Pit 2007





Notes

Risk of bias




BlindingFalls



Porthouse 2005








Notes

Risk of bias




BlindingFalls





Prince 2008





Notes

Risk of bias




BlindingFalls





Reinsch 1992






Risk of bias




BlindingFalls



Resnick 2002










Risk of bias




BlindingFalls



Robertson 2001a










Risk of bias




BlindingFalls


BlindingFractures



Robson 2003










Risk of bias




BlindingFalls



Rubenstein 2000









Notes

Risk of bias




BlindingFalls



Rubenstein 2007









Notes

Risk of bias




BlindingFalls



Ryan 1996









Risk of bias




BlindingFalls



Salminen 2008









Notes

Risk of bias




BlindingFalls



Sato 1999









Notes

Risk of bias




BlindingFalls


BlindingFractures












Risk of bias




BlindingFalls



Sherrington 2004









Risk of bias




BlindingFalls



Shigematsu 2008





Notes

Risk of bias







BlindingFalls



Shumway-Cook 2007





Notes




Risk of bias




BlindingFalls



Skelton 2005





Notes

Risk of bias







BlindingFalls



Smith 2007





Notes

Risk of bias




BlindingFalls


BlindingFractures






Speechley 2008





Notes

Risk of bias




BlindingFalls





Spice 2009






Risk of bias




BlindingFalls



Steadman 2003









Notes

Risk of bias




BlindingFalls



Steinberg 2000









Risk of bias




BlindingFalls



Stevens 2001










Risk of bias




BlindingFalls



Suzuki 2004









Notes

Risk of bias




BlindingFalls


BlindingFractures



Swanenburg 2007









Notes Pilot study

Risk of bias




BlindingFalls



Tinetti 1994










Risk of bias




BlindingFalls


BlindingFractures



Trivedi 2003










Risk of bias




BlindingFalls


BlindingFractures



Van Haastregt 2000









Notes

Risk of bias




BlindingFalls



Van Rossum 1993








Notes

Risk of bias




BlindingFalls



Vellas 1991





Notes




Risk of bias




BlindingFalls



Vetter 1992





Notes

Risk of bias







BlindingFalls


BlindingFractures



Voukelatos 2007





Notes

Risk of bias







BlindingFalls



Wagner 1994






Risk of bias




BlindingFalls






Weerdesteyn 2006





Notes

Risk of bias




BlindingFalls






Whitehead 2003






Risk of bias




BlindingFalls






Wilder 2001





Notes Abstract only

Risk of bias




BlindingFalls





Wolf 1996






Risk of bias




BlindingFalls





Wolf 2003






Risk of bias




BlindingFalls





Woo 2007





Notes

Risk of bias




BlindingFalls



Wyman 2005









Notes

Risk of bias




BlindingFalls
















(Continued)
















(Continued)
















(Continued)















(Continued)


















(Continued)





Beyer 2007






Di Monaco 2008









Madureira 2007






Pfeifer 2004






Sato 2005b








Notes

Weber 2008





Notes




Behrman









Blalock







Blalock (Continued)




Notes

Ciaschini









Cryer





Cryer (Continued)







Donaldson












Edwards










Edwards (Continued)


Grove








Notes

Haines







Haines (Continued)




Notes

Hill a










Hill a (Continued)

Notes

Hill b









Jee







Jee (Continued)


Starting date 2005


Notes

Johnson










Johnson (Continued)


Notes

Kenny











Klaber Moffett








Notes

Lesser






Lesser (Continued)





Notes

Lips









Lips (Continued)


Notes

Lord








Notes



Maki









Masud









Masud (Continued)


Notes

Menz








Notes



Miller










Notes

Olde Rikkert












Palvanen








Notes

Pighills











Notes

Press







Press (Continued)




Notes

Sanders










Sanders (Continued)

Notes

Schumacher









Snooks






Snooks (Continued)





Notes

Stuck







Stuck (Continued)





Taylor










Taylor (Continued)


Notes

Tousignant









Notes



Vind









Zeeuwe









Zeeuwe (Continued)


Notes

Zijlstra








Notes










studies

No of








































studies

No of













studies

No of

















studies

No of











































studies

No of












studies

No of












studies

No of

























studies

No of






















studies

No of


















studies

No of






studies

No of








studies

No of
























at baseline


studies

No of















studies

No of









studies

No of















































studies

No of






studies

No of














vention


studies

No of











F E E D B A C K


Summary


Reply






Contributors





H I S T O R Y


















Internal sources


External sources






N O T E S




I N D E X T E R M S



MeSH check words

Aged Humans























B A C K G R O U N D








O B J E C T I V E S


M E T H O D S


Types of studies










Primary outcomes


Secondary outcomes



Electronic searches



















related fractures






Data synthesis












Economics issues


R E S U L T S





Included studies


Design


Sample sizes


Setting

Location


Sampling frame





Participants



Interventions




Exercises






Study ID Gait bal-

ancefunctional

training

Strength re-

sistance

training


dance etc)

General phys-

ical activity

Endurance Other

Ashburn 2007

Ballard 2004

Barnett 2003

Brown 2002

Buchner1997a

Bunout 2005

Campbell1997

Campbell1999

Carter 2002

Cerny 1998

Cornillon2002

Day 2002

Fiatarone1997





Hauer 2001

Helbostad2004

Korpelainen2006

dance stamping

Latham 2003

Li 2005

Lin 2007

Liu-Ambrose2004



Lord 1995

Lord 2003 dance

Luukinen2007

self care

McMurdo1997

Means 2005

Morgan 2004

Nitz 2004

Pereira 1998


standupstep up

Resnick 2002

Robertson2001a




Rubenstein2000

Sherrington2004

Shigematsu2008


walkinggroup

Skelton 2005

Steadman2003

Suzuki 2004

Voukelatos2007

Weerdesteyn2006


Tai Chigroup

Wolf 2003


Tai Chigroup









Surgery




Psychological










Outcomes





Excluded studies



Ongoing studies












Allocation




Blinding








Exercises

















rion


747 (SD 383) NA NA 829 (SD 449) 664 (SD 317) No


267 (range 255to 280)

250 (range 237to 261)


Dukas 2004 726 (SD 279) 746 (SD 290)

706 (SD 267) NA NA No

Gallagher 2001 793 (SD 247) 780 (SD216)

805 (SD 274) NA NA No



29 (range 6 to85)

30 (range 12 to64)


Latham 2003 374 (95 CI349 to 449)

474 (95 CI399 to 524)

NA NA No


















Surgery



Cataract surgery




Psychological









signalling)




protection)




















D I S C U S S I O N





Exercises













problems

Poor vision




Post hip fracture


















Exercise





Vitamin D





Exercise












Poor vision

















R E F E R E N C E S











(4)407ndash14 [PUBMED 12851185]






(10)670ndash6 [PUBMED 9278463]



199341321ndash5 [PUBMED 8440857]



200534(1)80ndash3 [PUBMED 15591487]


































(2)175ndash81 [PUBMED 17302652]











































200433(1)45ndash51 [MEDLINE 14695863]































199881(5)352ndash6 [PUBMED 9623035]
































(4)489ndash98 [PUBMED 17397425]





































(6)40ndash9 [PUBMED 12071273]



















































































(11)1800ndash6 [PUBMED 7977921]

























200755(10)1548ndash56 [MEDLINE 17908058]


















200726(2)309ndash16 [MEDLINE 17074485]








































(2)157ndash63 [MEDLINE 15977465]







200081(3)319ndash23




























































































































































































AGSBGS 2001


Beswick 2008


Bischoff 2003


Boutron 2008




Broe 2007


Buchner 1993


Cameron 2005


Campbell 1990


Campbell 1999c


Campbell 2004


2004328(7441)702ndash8 [PUBMED 15031246]

Campbell 2005


Campbell 2006


Suppl 2ii60ndash4

Campbell 2007


Chapuy 2002



Close 2000


Cummings 1995


Excel


Findorff 2007


Flicker 2005



Gates 2008


Gillespie 2003


Goodwin 2008


200823(5)631ndash40 [MEDLINE 18181210]

Haas 2006


Hauer 2006


Higgins 2003


Higgins 2008a




Higgins 2008b


Jackson 2007


2007100(4)185ndash92 [MEDLINE 17308327]

Keene 1993


Kellogg 1987


Lamb 2005



Lamb 2007


Lefebvre 2008


Lord 2008


McAlister 2003


(19)2545ndash53 [MEDLINE 12759326]

RevMan 5


Richy 2008


Rizzo 1996


Robertson 2001c



Robertson 2001d


Robertson 2007



Sach 2007


Salkeld 2000


Sattin 1992


Sherrington 2008


Smeeth 2002


Stata


Tinetti 1988


Tinetti 1997




Vellas 1997


199726(3)189ndash93 [MEDLINE 9223714]

Zecevic 2006



Gillespie 2008







Ashburn 2007





Notes

Risk of bias




BlindingFalls


BlindingFractures





Assantachai 2002





Notes

Risk of bias




BlindingFalls



Ballard 2004









Notes

Risk of bias




BlindingFalls



Barnett 2003









Notes

Risk of bias




BlindingFalls












Notes

Risk of bias




BlindingFalls


BlindingFractures



Brown 2002








Notes

Risk of bias




BlindingFalls



Buchner 1997a










Risk of bias




BlindingFalls






Bunout 2005






Risk of bias




BlindingFalls





Campbell 1997






Risk of bias




BlindingFalls





Campbell 1999






Risk of bias







BlindingFalls



Campbell 2005






Risk of bias




BlindingFalls






Carpenter 1990





Notes

Risk of bias




BlindingFalls





Carter 1997






Risk of bias




BlindingFalls



Carter 2002









Notes

Risk of bias




BlindingFalls



Cerny 1998









Risk of bias




BlindingFalls



Clemson 2004









Risk of bias




BlindingFalls



Close 1999





Notes




Risk of bias




BlindingFalls



Coleman 1999





Notes

Risk of bias







BlindingFalls



Cornillon 2002





Notes

Risk of bias




BlindingFalls





Cumming 1999





Notes

Risk of bias




BlindingFalls



Cumming 2007









Notes

Risk of bias




BlindingFalls


BlindingFractures



Davison 2005










Risk of bias




BlindingFalls



Day 2002









Notes

Risk of bias




BlindingFalls





Dhesi 2004






Risk of bias




BlindingFalls



Dukas 2004









Notes

Risk of bias




BlindingFalls



Elley 2008









Notes

Risk of bias




BlindingFalls



Fabacher 1994








Notes

Risk of bias




BlindingFalls



Fiatarone 1997





Notes Abstract only

Risk of bias







BlindingFalls



Foss 2006





Notes

Risk of bias







BlindingFalls


BlindingFractures



Gallagher 1996





Notes

Risk of bias




BlindingFalls





Gallagher 2001





Notes

Risk of bias




BlindingFalls





Grant 2005





Notes

Risk of bias




BlindingFalls


BlindingFractures





Gray-Donald 1995





Notes

Risk of bias




BlindingFalls



Green 2002









Notes

Risk of bias




BlindingFalls



Greenspan 2005










Risk of bias




BlindingFalls



Harwood 2004










Risk of bias




BlindingFalls


BlindingFractures



Harwood 2005








Notes

Risk of bias




BlindingFalls


BlindingFractures



Hauer 2001








Notes

Risk of bias




BlindingFalls



Helbostad 2004








Notes

Risk of bias




BlindingFalls



Hendriks 2008








Notes

Risk of bias




BlindingFalls



Hill 2000





Notes




Risk of bias




BlindingFalls



Hogan 2001





Notes

Risk of bias







BlindingFalls


BlindingFractures



Hornbrook 1994





Notes

Risk of bias







BlindingFalls


BlindingFractures



Huang 2004





Notes

Risk of bias







BlindingFalls



Huang 2005









Risk of bias




BlindingFalls



Jitapunkul 1998





Notes

Risk of bias







BlindingFalls



Kenny 2001






Risk of bias







BlindingFalls


BlindingFractures



Kingston 2001





Notes

Risk of bias







BlindingFalls



Korpelainen 2006





Notes

Risk of bias




BlindingFalls





BlindingFractures



Lannin 2007





Notes Pilot study

Risk of bias




BlindingFalls





Latham 2003






Risk of bias




BlindingFalls





Li 2005





Notes

Risk of bias




BlindingFalls



Lightbody 2002










Risk of bias




BlindingFalls



Lin 2007








Notes

Risk of bias




BlindingFalls



Liu-Ambrose 2004








Notes

Risk of bias




BlindingFalls



Lord 1995








Notes

Risk of bias




BlindingFalls



Lord 2003










Risk of bias




BlindingFalls



Lord 2005








Notes

Risk of bias




BlindingFalls



Luukinen 2007





Notes

Risk of bias







BlindingFalls



Mahoney 2007





Notes

Risk of bias







BlindingFalls



McKiernan 2005





Notes

Risk of bias




BlindingFalls






McMurdo 1997





Notes

Risk of bias




BlindingFalls


BlindingFractures





Means 2005





Notes

Risk of bias




BlindingFalls



Meredith 2002









Notes

Risk of bias




BlindingFalls



Morgan 2004










Risk of bias




BlindingFalls



Newbury 2001









Risk of bias




BlindingFalls



Nikolaus 2003









Risk of bias




BlindingFalls



Nitz 2004








Notes

Risk of bias




BlindingFalls



Pardessus 2002





Notes

Risk of bias







BlindingFalls



Pereira 1998





Notes

Risk of bias




BlindingFalls





Pfeifer 2000





Notes

Risk of bias




BlindingFalls





Pit 2007





Notes

Risk of bias




BlindingFalls



Porthouse 2005








Notes

Risk of bias




BlindingFalls





Prince 2008





Notes

Risk of bias




BlindingFalls





Reinsch 1992






Risk of bias




BlindingFalls



Resnick 2002










Risk of bias




BlindingFalls



Robertson 2001a










Risk of bias




BlindingFalls


BlindingFractures



Robson 2003










Risk of bias




BlindingFalls



Rubenstein 2000









Notes

Risk of bias




BlindingFalls



Rubenstein 2007









Notes

Risk of bias




BlindingFalls



Ryan 1996









Risk of bias




BlindingFalls



Salminen 2008









Notes

Risk of bias




BlindingFalls



Sato 1999









Notes

Risk of bias




BlindingFalls


BlindingFractures












Risk of bias




BlindingFalls



Sherrington 2004









Risk of bias




BlindingFalls



Shigematsu 2008





Notes

Risk of bias







BlindingFalls



Shumway-Cook 2007





Notes




Risk of bias




BlindingFalls



Skelton 2005





Notes

Risk of bias







BlindingFalls



Smith 2007





Notes

Risk of bias




BlindingFalls


BlindingFractures






Speechley 2008





Notes

Risk of bias




BlindingFalls





Spice 2009






Risk of bias




BlindingFalls



Steadman 2003









Notes

Risk of bias




BlindingFalls



Steinberg 2000









Risk of bias




BlindingFalls



Stevens 2001










Risk of bias




BlindingFalls



Suzuki 2004









Notes

Risk of bias




BlindingFalls


BlindingFractures



Swanenburg 2007









Notes Pilot study

Risk of bias




BlindingFalls



Tinetti 1994










Risk of bias




BlindingFalls


BlindingFractures



Trivedi 2003










Risk of bias




BlindingFalls


BlindingFractures



Van Haastregt 2000









Notes

Risk of bias




BlindingFalls



Van Rossum 1993








Notes

Risk of bias




BlindingFalls



Vellas 1991





Notes




Risk of bias




BlindingFalls



Vetter 1992





Notes

Risk of bias







BlindingFalls


BlindingFractures



Voukelatos 2007





Notes

Risk of bias







BlindingFalls



Wagner 1994






Risk of bias




BlindingFalls






Weerdesteyn 2006





Notes

Risk of bias




BlindingFalls






Whitehead 2003






Risk of bias




BlindingFalls






Wilder 2001





Notes Abstract only

Risk of bias




BlindingFalls





Wolf 1996






Risk of bias




BlindingFalls





Wolf 2003






Risk of bias




BlindingFalls





Woo 2007





Notes

Risk of bias




BlindingFalls



Wyman 2005









Notes

Risk of bias




BlindingFalls
















(Continued)
















(Continued)
















(Continued)















(Continued)


















(Continued)





Beyer 2007






Di Monaco 2008









Madureira 2007






Pfeifer 2004






Sato 2005b








Notes

Weber 2008





Notes




Behrman









Blalock







Blalock (Continued)




Notes

Ciaschini









Cryer





Cryer (Continued)







Donaldson












Edwards










Edwards (Continued)


Grove








Notes

Haines







Haines (Continued)




Notes

Hill a










Hill a (Continued)

Notes

Hill b









Jee







Jee (Continued)


Starting date 2005


Notes

Johnson










Johnson (Continued)


Notes

Kenny











Klaber Moffett








Notes

Lesser






Lesser (Continued)





Notes

Lips









Lips (Continued)


Notes

Lord








Notes



Maki









Masud









Masud (Continued)


Notes

Menz








Notes



Miller










Notes

Olde Rikkert












Palvanen








Notes

Pighills











Notes

Press







Press (Continued)




Notes

Sanders










Sanders (Continued)

Notes

Schumacher









Snooks






Snooks (Continued)





Notes

Stuck







Stuck (Continued)





Taylor










Taylor (Continued)


Notes

Tousignant









Notes



Vind









Zeeuwe









Zeeuwe (Continued)


Notes

Zijlstra








Notes










studies

No of








































studies

No of













studies

No of

















studies

No of











































studies

No of












studies

No of












studies

No of

























studies

No of






















studies

No of


















studies

No of






studies

No of








studies

No of
























at baseline


studies

No of















studies

No of









studies

No of















































studies

No of






studies

No of














vention


studies

No of











F E E D B A C K


Summary


Reply






Contributors





H I S T O R Y


















Internal sources


External sources






N O T E S




I N D E X T E R M S



MeSH check words

Aged Humans



Documents

Interventions for preventing falls in older people living in the community