http://cre.sagepub.com/Clinical Rehabilitation

http://cre.sagepub.com/content/18/6/660The online version of this article can be found at:

 DOI: 10.1191/0269215504cr746oa

2004 18: 660Clin RehabilChristine Dedding, Mieke Cardol, Isaline CJM Eyssen and Anita Beelen

outcome measurementValidity of the Canadian Occupational Performance Measure: a client-centred

  

Published by:

http://www.sagepublications.com

can be found at:Clinical RehabilitationAdditional services and information for    

  http://cre.sagepub.com/cgi/alertsEmail Alerts:

 

http://cre.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://cre.sagepub.com/content/18/6/660.refs.htmlCitations:  

What is This? 

- Jun 1, 2004Version of Record >>

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

Clinical Rehabilitation 2004; 18: 660-667

Validity of the Canadian Occupational PerformanceMeasure: a client-centred outcome measurementChristine Dedding Department of Rehabilitation, Academic Medical Center Amsterdam, Mieke Cardol Department ofRehabilitation, Academic Medical Center Amsterdam and Netherlands Institute of Health Services Research (Nivel), Utrecht,Isaline CJM Eyssen Department of Occupational Therapy, Joost Dekker Department of Rehabilitation Medicine, VUUniversity Medical Center, Amsterdam and Anita Beelen Department of Rehabilitation, Academic Medical Center Amsterdamand Department of Rehabilitation Medicine, VU University Medical Center, Amsterdam, The Netherlands

Received 8th March 2003; returned for revisions 15th July 2003; revised manuscript accepted 2nd October 2003.

Objective: To study the convergent and divergent validity of the CanadianOccupational Performance Measure (COPM).Design: Cross-sectional study.Setting: The occupational therapy departments of two university hospitals inAmsterdam.Subjects: One hundred and five consecutive outpatients.Outcome measures: The COPM is a measure of a client's self-perception ofoccupational performance in the areas of self-care, productivity and leisure. Outcomemeasures of the COPM are: the client's most important problems in occupationalperformance and a total score for performance and a total score for satisfaction forthese problems. Problems reported in the COPM were compared with the SicknessImpact Profile (SIP68), the Disability and Impact Profile (DIP) and an open-endedquestion.Results: Complete data were obtained for 99 clients. The identification ofoccupational performance problems with the COPM surpassed the items reportedin the SIP68, the DIP and the open-ended question, which confirms the surplusvalue of the COPM. Divergent validity was further demonstrated by the lowcorrelation coefficients between the total SIP68 scores and the COPM. Seventy-fourper cent of the occupational performance problems reported in the COPM had acorresponding item in the DIP and 49% had a corresponding item in the SIP68.Convergent validity was supported by the fact that 63% of the correspondingproblems in the DIP were reported to be a disruption of quality of life and 74% of thecorresponding problems in the SIP68 were identified as a disability.Conclusion: The results of this study provide supportive evidence for theconvergent and divergent validity of the COPM. The data support the assumptionthat the COPM provides information that cannot be obtained with currentstandardized instruments to measure health.

Address for correspondence: Anita Beelen, Department ofRehabilitation, Academic Medical Center, Amsterdam, PO Box22660, 1100 DD Amsterdam, The Netherlands.e-mail: j.a.beelen@Jamc.uva.nl

10. I 191/0269215504cr746oaC) Arnold 2004

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

Validity of the COPM 661

Introduction

Current health care policies focus on demand-induced and patient-oriented care. Demand-induced care addresses the problems experiencedby the patient as the starting point for treatment.In patient-oriented care hospital resources andpersonal care are organized around the patientrather than around various specialized depart-ments.1 However the concept of 'clinically definedneed' remains dominant in clinical practice, possi-bly because neither demand-induced nor patient-oriented care explicitly advocate active participa-tion of the client during the treatment process. Aclient-centred approach aims to do so.

In a client-centred collaborative model, theclient is involved in all phases of the treatmentprocess: the client's opinion is crucial with regardto defining priorities for treatment, deciding onadequate treatment strategies and achieving thedesired treatment outcome. The role of the thera-pist is to be attentive to the needs of the client andto provide all the information that is needed tomake balanced decisions.1"2 Research findingssuggest that client-centred practice leads to im-provement of client satisfaction and adherence tohealth service programmes." 3

Since the 1980s, considerable progress has beenmade in the measurement of perceived health, andmany different measurement instruments are nowavailable. Some health status measures have gainedwidespread use in rehabilitation research: forexample the Sickness Impact Profile (SIP), theNottingham Health Profile (NHP) and the MOSShort-Form 36 (SF-36). However, these instru-ments provide no information about the impact ofthe perceived health on daily life. In other words,no distinction is made between inability to performan activity on the one hand, and perceiving this asa problem on the other hand. Furthermore, it maywell be that activities that are really important tothe client are not listed in the questionnaire.

Unlike the above-mentioned questionnaires, anew generation of measurement instruments doesfocus on the client's point of view. Examples arethe Impact on Participation and Autonomy (IPA)questionnaire, focusing on social participation4and the Canadian Occupational PerformanceMeasure (COPM) that addresses activities. TheCOPM enables a detailed exploration of activities

that are important for a person. The COPM isbased on a semi-structured interview. The semi-structured design enables individuals to identifyany activity of importance that they find difficultto perform. The conceptual basis for the COPM isderived from the Canadian Model of OccupationalPerformance. In this model, occupational perfor-mance is defined as the ability to choose, organizeand satisfactorily perform meaningful occupationsthat are culturally defined and age appropriate forlooking after one's self, enjoying life and contri-buting to the social and economic fabric of acommunity.2The COPM has received international attention

because it is an important method of assessmentfor directing occupational therapy interventionsand measuring client-centred outcomes. Howeverno clear description of the psychometric propertiesof the COPM can be found in the literature. Itscontent validity is supported by the processthrough which it was developed,3 but informationabout the reliability of the COPM is difficult tofind: the authors of the COPM refer to threeunpublished studies that are claimed to showacceptable test-retest reliability.3 With regard toinformation about the validity of the COPM,authors refer to four studies (two of which areunpublished and the third focuses on children) thatare claimed to support the validity of the COPM.3Although there are different studies in which thevalidity and/or reliability of the COPM have beeninvestigated, some studies have only focused onspecific diagnoses,5'6 and in other studies smallstudy populations are investigated,7'0 so there isstill a lack of information about the psychometricproperties of the COPM. This article is part of aseries of two. It evaluates the extent to which theCOPM agrees with some other measures (conver-gent validity) and the extent to which the COPMprovides new data, different from pre-existingmeasures (divergent validity). A second articleabout the reliability of the COPM is forthcoming.

Method

During the study period 170 outpatients, whofulfilled the inclusion criteria, were referred tothe occupational therapy departments of two

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

662 C Dedding et al.

academic hospitals in Amsterdam, the AcademicMedical Center (AMC) and the VU UniversityMedical Center (VUMC). Inclusion criteria were:above 18 years of age, no difficulty in comprehend-ing the Dutch language, and perceiving limitationsin more than one activity of daily life. One hundredand five patients (50 from the AMC and 55 fromthe VUMC) were willing to participate in thestudy. The medical ethics committees of thehospitals involved approved the study. All subjectsgave written informed consent.

In occupational therapy, the COPM is used asan initial assessment for setting goals and planningtreatment. The interview focuses on activities thatthe client wants, needs or is expected to perform.The importance of each activity, as perceived bythe client, is first rated on a 10-point scale rangingfrom 1 (not important at all) to 10 (extremelyimportant). In the next step the client selects thefive most important activities, which are then ratedon a 10-point scale for performance, ranging from1 (not at all able) to 10 (able to perform extremelywell), and for satisfaction, ranging from 1 (not atall satisfied) to 10 (extremely satisfied). TheCOPM was translated into Dutch in 1999.The Disability and Impact Profile (DIP) covers a

wide range of activities that may be restricted by adisabling disease. It is a self-administered ques-tionnaire, and consists of 39 items with parallelquestions about disabilities and their importancefor or impact on the patient. Three items concernsymptoms, whereas the other 36 items cover fivedomains: mobility (10 items), self-care (six items),social activities (10 items), communication (fiveitems) and psychological status (five items). Eachitem is rated for its current disability aspect on a0-10 point scale (0= maximal disability; 10= nodisability) and for the importance or impact of thatparticular disability, also on a 0-10 point scale(0= not important at all; 10= most important ofall). A weighted score for each pair of questions iscalculated as follows: the deficit from the normalsituation is calculated by subtracting the actualdisability score from 10. This deficit is multipliedby the impact score for that item, resulting in a'weighted deficit' (representing both the objectiveand subjective aspects of activity restrictions). Thisvalue is divided by 100. Subtracting this normal-ized weighted deficit from 1 yields a weighted itemscore. 12

The Sickness Impact Profile (SIP68, a shortversion of the SIP136) assesses the impact of illnesson daily functioning and behaviour, and consists of68 items categorized into six subscales coveringthree broad dimensions: physical, psychologicaland social health. The physical dimension includesthe subscales 'somatic autonomy' (e.g., gettingdressed, walking), and 'mobility control' (beha-viour related to walking and arm function). Thepsychological dimension consists of the subscales'psychological autonomy and communication' and'emotional stability' (the effect of health status onemotional behaviour). The social dimension ad-dresses social and mobility aspects, and consists ofthe subscales 'social behaviour' and 'mobilityrange' (instrumental daily activities). All itemsare scored dichotomously; the number of con-firmed sickness impact items makes up thesub-scale scores and the total score. Internalconsistency and test-retest reliability of theSIP68 were satisfactory. The SIP68 is consideredto be useful to assess various rehabilitationgroups.'3"4

Finally, written responses to a self administeredquestionnaire with one open-ended question: 'Inyour own words, what do you experience as thethree main problems caused by your health condi-tion or disability?' were obtained.

Participants were interviewed at the hospital. Ineach hospital, the interviews were conducted by anoccupational therapist. The therapists were trainedand experienced in the use of the COPM. Allspontaneous remarks made by participants aboutthe COPM (during and after the interview) werenoted.

In the week after the COPM interview, partici-pants completed the SIP68 and the DIP, and wrotetheir response to the open-ended question at home.

Data analysisDivergent validity refers to the ability to differ-

entiate the concept under study from other con-structs. Convergent validity refers to thecorrespondence between different methods ofmeasuring the same theoretical construct.'5

Divergent validityGiven the semi-structured design and client-

centred approach, it was expected that theCOPM would identify problems that do not match

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

Validity of the COPM 663

any item in the SIP68 or the DIP. For eachproblem reported in the COPM two researchersassessed, independently, whether an item in theDIP or the SIP68 could be found that matched thereported COPM problem. In case of disagreementbetween the two researchers, a third persondecided on the matching. Problems that were notcovered by items in the SIP68 or the DIP wereidentified and described.

Additionally, Spearman's rank correlation coef-ficients were calculated between the performanceand satisfaction scores of the COPM on the onehand, and the total SIP68 score and physicaldimension score of the SIP68 on the other hand(two-sided tested). There were expected to be lowcorrelations between both COPM scores and thetotal SIP68 score because the total SIP score iscomposed of predefined items and focuses oncertain activities without taking importance intoconsideration. In contrast, the COPM addressesonly those activities that are important to theperson in question. Furthermore, it is known thatthere is no linear relationship between impairmentsand performance or problem-experience. ,16 Giventhe latter reason, it was also expected to find lowcorrelations between the performance and satisfac-tion scores of the COPM and the physical domainof the SIP68.

Convergent validityWith regard to the DIP, it was expected to find

agreement with items in the COPM because theDIP measures perceived health status as well asimportance. It was expected that the correspond-ing items in the DIP would have a score indicativeof a disruption of the quality of life. The authors ofthe DIP define a weighted score of less than 0.50 asa 'major disruption of quality of life'.'' In thepresent analyses it was not feasible to take 'majordisruptions' as the starting point for comparisonwith the COPM, since the COPM is not a norm-referenced measure. Therefore, a milder cut-offscore of 0.65 was chosen. Weighted scores < 0.65are regarded as disruptions of quality of life. Thepercentage agreement was calculated as the per-centage of the corresponding items that had aweighted score < 0.65. The various correspondingitems, which were not regarded as a disruption ofquality of life according to the weighted score inthe DIP, were described separately.

It was also expected to find agreement betweenproblems with the occupational performance re-ported in the COPM and corresponding items inthe SIP68. The percentage of agreement wascalculated as the percentage of the correspondingitems that were indicated as a disability by theclient. The various corresponding items that werenot indicated as a disability in the SIP68 weredescribed separately.

Finally, it was expected to find agreementbetween the problems reported in the COPM andthe problems reported in the open-ended question.To make responses compatible, all problems re-ported in the open-ended question and in theCOPM were categorized according to the Cana-dian Model of Occupational Performance by tworesearchers, independently, not knowing the re-sponses of the individual patients to the COPM,and then compared. The percentage of agreementwas calculated as the percentage of problemsreported in the open-ended question that werealso reported in the COPM.

Results

Study populationAlthough 105 consecutive patients met the

criteria for inclusion in the study, the data of 99patients (32 males) were used in the analyses. Sixpatients did not complete the questionnaires.The diagnoses of the clients were as follows:hand injury (n = 29), central neurological disorders(e.g., multiple sclerosis or stroke) (n = 23), neuro-muscular diseases (e.g., limb-girdle muscular dys-trophy, hereditary motor and sensory neuropathyand postpoliomyelitis syndrome) (n = 17), andother diagnoses (e.g., chronic pain, Ehlers-Danlos,diabetes mellitus, arthrosis or arthritis) (n = 30).

Reported problems in occupational performance(COPM)With the COPM, 443 occupational performance

problems were identified. These were subdividedinto 182 self-care activities (41%), 98 activities inthe area of productivity (22%), and 60 activities inthe area of leisure (14%). Although the COPMmeasures at the level of activity, and consequentlythe therapist should direct the client towardsactivity outcomes instead of functions, some

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

664 C Dedding et al.

respondents repeatedly reported a problem at theimpairment level, especially physical impairments(14%) and, to a lesser extent affective impairments(6%). The remaining 4% concerned cognitive,spiritual or environmental issues. Examples aregiven in Table 1.

Validity of the COPM

Divergent validityFor 81 problems reported in the COPM no

corresponding item could be found in either theDIP or the SIP68 (Table 2). These problemsconcerned sitting, caring for loved ones such asgrandchildren or a spouse, and personal appear-ance.

In agreement with the expectation, correlations(RJ) between the total SIP score and the COPMscore were low. The correlation between the SIP68

and the COPM performance scores was - 0.20(p = 0.05), and between the SIP68 and the COPMsatisfaction scores it was - 0.19 (p = 0.07). Thecorrelations between the performance and satisfac-tion scores of the COPM and the physical domainof the SIP68 were also low, respectively - 0.21(p = 0.04) and - 0.19 (p = 0.06). Therefore, onlythe correlation between the physical domain of theSIP68 and the COPM performance score was

significant as predicted.

Convergent validityOf the 443 occupational performance problems

reported in the COPM 328 (74%) had a corre-

sponding item in the DIP, and 216 problems (49%)had a corresponding item in the SIP68. Of the328 COPM problems that had a correspondingitem in the DIP, 205 (63%) problems were also

Table 1 Examples of problems reported in the COPM

n (%)

At the activity level

Self-care(Drive a car, ride a bike, tie one's shoelaces, turn the tap on/off, put on make-up, get up from the toilet, dress

oneself, lock the door)Productivity(Iron clothes, take care of children, work on the computer, open jars, maintain the garden, all kinds of paid and

unpaid jobs)Leisure(Visit friends and family, play with grandchildren, dance, travel, visit a theatre, all kinds of handicrafts and

sports)

At the impairment level

Physical(Strength, hyperpathy, grasping, reaching, pain, coordination)Affective(Asking for help, feeling vulnerable, feeling lonely, feeling dependent, accepting limitations)Cognitive(Concentration, memory)

At the spiritual level(Not wanting to depend on anybody, being able to do as you wish, being able to live/act spontaneously)

At the environmental level(House or surroundings not accessible for wheelchair, feeling cold in the house because energy costs are too

high)

182 (41)

98 (22)

60 (14)

59

26

6

(13)

(6)

(1)

8 (2)

4 (1)

All identified problems (n = 443) were categorized into four levels 'activity', 'impairment', 'spiritual' and 'environmental'. Withinthe 'activity' level, problems were categorized as 'self-care', 'productivity' and 'leisure'. Within the 'impairment' level, problemswere: categorized as 'physical', 'affective' and 'cognitive'.n = number of identified problems within the category; % = percentage of all identified problems.

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

Validity of the COPM 665

Table 2 Examples of individual-specific activities identifiedin the COPM only (n-=81)

SittingIn the carBehind the computerOn a couchSitting straight

InterdependenceTaking care of grandchildrenTaking care of spousesAsking for help

Personal appearanceWet shavingGood shaving and cleaning the shaverBlow-drying hairDoing one's hair (pigtail, a bun)Putting on make-upDepilating

RemainingUsing the telephoneOpening the garage doorMaking choicesTurning pages of the newspaperDrying dishesGoing to the toilet at night.Reading in bedInjecting insulin

Individual-specific activities identified in the COPM that werenot covered by items in the SIP68 or the DIP (n = 81) could becategorized into 'sitting' (n = 20), 'interdependence' (n = 14),'personal appearance' (n = 12) and 'remaining' (n = 35).n = number of identified problems within the category.

reported as a disruption of quality of life in theDIP.Of the 216 problems that had a corresponding

item in the SIP, 160 problems (74%) were alsoindicated as a disability in the SIP68.

Ninety-seven clients answered the open-endedquestion. Sixty-three per cent of the problemsreported in the open-ended question matched theproblems reported in the COPM.

FeasibilityClients expressed enthusiasm about the COPM;

they reported that it makes them feel that they aretaken seriously as individuals instead of beinggeneralized according to their diagnosis or age.Most clients reported that they found it helpful tospend time thinking in detail about their most

important occupational performance goals, parti-cularly with the knowledge that these goals wouldbe incorporated in the occupational therapy treat-ment. A considerable number of clients (n = 37)made remarks about the scoring system during theinterview. They sometimes found it difficult totranslate their problems into a score; they were

afraid of being too negative or too subjective, orfelt that the scoring was arbitrary, depending on

the day and mood. The researchers found theCOPM easy to administer, and very helpfulto identify issues that are relevant for the client.The average administration time was 30-45 min-utes.

Discussion

The COPM seems to be unique in that it enablesself-rating of any self-care, productivity or leisureactivity. The broad variation in problems identifiedwith the COPM confirms the belief that valueswith regard to occupational performance out-comes differ between clients, and are influencedby the physical, cultural and social environment ofthe client. '17 To a certain extent this broadvariation in problems is lost in the comparison ofindividual problems as reported in the COPM withthe structured format of the SIP68 and the DIP,which explains the low percentage (18%) ofproblems unique to the COPM. For instance, allleisure activities (e.g., playing a pipe organ, horseriding, fitness) were categorized into one DIP item'leisure activities' and into two items on the SIP68

Clinical messages

* The Dutch Canadian Occupational Perfor-mance Measure (COPM) is a valid measureof a client's self-perception of occupationalperformance in the areas of self-care, pro-ductivity and leisure.

* The COPM provides information that can-not be obtained with current standardizedinstruments.

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

666 C Dedding et al.

concerning the amount of time spent on leisureactivities.The convergent validity of the COPM was

satisfying. A significant correlation was foundbetween the physical domain of the SIP68 andthe performance score of the COPM, and therewas substantial agreement between the problemsreported in the COPM and the items in the DIPand the SIP68. The items in the DIP are lessdetailed than the items in the SIP68, whichprobably explains the fact 74% of the COPMproblems matched items in the DIP, and only49% matched items in the SIP68. The higheragreement between the items indicated as a dis-ability in the SIP68 (74%), compared to itemsreported as a disruption of quality of life accordingto the DIP (63%), can be explained by the fact thatthe DIP is a norm-referenced measure, whereas theSIP68 is not. Furthermore, it must be realized thatfor the items of the DIP an arbitrary cut-off valuefor disrupted quality of life was used (a weightedscore < 0.65).

Divergent validity was supported by the lowcorrelation between the satisfaction score of theCOPM and the SIP68, and the low correlationbetween the performance score of the COPM andthe total SIP score. The identification of occupa-tional performance problems using the COPMprovided more detailed information, and theproblems exceeded those identified by clients inthe open-ended question. It was expected to finda higher agreement between the problems reportedin the COPM and the issues raised in the open-ended question. The discrepancy can be explainedby the interaction between the client and theinterviewer, which enables clients to reflect on theirsituation and reformulate the problems theyexperience. In addition, the wording of theopen-ended question was not directed towardsreporting problems in activities and hence pro-blems such as pain, a loss of strength or restrictedrange of motion were frequently reported as one ofthe three main problems instead of the activityitself that was restricted. A few respondentsrepeatedly reported problems at the impairmentlevel in the COPM as well, although the research-ers tried to focus the interview on activity out-comes. This was respected, since in a client-centredapproach it is the client who defines the goals for

treatment, even when this is not in accordance withthe design of the instrument.Although in the Netherlands one is used to

scoring on a scale between 1 and 10, individualinterpretation is inevitable. For one person a'6' indicates dissatisfaction, while others considerthis to be a moderate score. In practice, however,the interpretation problem is of less importance,since the COPM is used within the individual andchanges in performance and satisfaction are eval-uated by comparing the initial assessment scoreswith the reassessment scores. In research one mustbe aware that the scores are not as objective as theyseem.

Given this study, the COPM was not consideredto be useful for clients with one clear problem inthe field of activity, since they have alreadyindicated their treatment goal. However, in clinicalpractice the scoring for performance and satisfac-tion for that particular problem can still be useful,since this reflects the perception of the client andoffers the possibility to detect change from theclient's point of view.

In general, the COPM is experienced as a veryhelpful tool with which to bring the theory of theclient-centred approach into practice. Thiscan only be achieved when the therapist is trainedin the principles of client-centred practice and hasbeen trained to administer the semi-structuredinterview. Moreover, introduction of the COPMin a multidisciplinary team demands reconsidera-tion of current clinical practice, since this can be inconflict with the principles of client-centred prac-tice.

In conclusion, the data suggest that the COPMprovides information that cannot be obtained withstandardized instruments to measure health.Its strength lies in the self-rating of any activityof importance, leading to an open dialogue be-tween client and therapist about expectations andtreatment outcomes. The results of this studyprovide supportive evidence for the convergentand divergent validity of the COPM.

AcknowledgementThe study was supported by grant, number

1310.0005 from ZonMw.

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from

Validity of the COPM 667

References

1 Sumsion T (ed.) Client-centred practice inoccupational therapy. A guide to implementation.Edinburgh: Churchill Livingstone, 1999.

2 Canadian Association of Occupational Ther-apists. Enabling occupation: an occupationaltherapy perspective. Ottawa: CAOT Publications,1997.

3 Law M, Baptiste S, Carswell A, McColl MA,Polatajko H, Pollock N. Canadian OccupationalPerformance Measure, third edition. Ottawa: CAOTPublications ACE, 1998.

4 Cardol M, De Haan RJ, De Jong BA, Van den BosGAM, De Groot IJM. Psychometric properties ofthe 'Impact on Participation and Autonomy (IPA)questionnaire. Arch Phys Med Rehabil 2001; 82:210-16.

5 Carpenter L, Baker GA, Tyldesley B. The use of theCanadian Occupational Performance Measure asan outcome of a Pain Management Program. Can JOccup Ther 2001; 68: 16-22.

6 Cup EHC, Scholte op Reimer WJM, ThijssenMCE, Van Kuyk-Minis MAH. Reliability, validityand utility of the Canadian OccupationalPerformance Measure in stroke patients. ClinRehabil 2003; 17: 402-409.

7 Bodiam C. The use of the Canadian OccupationalPerformance Measure for the Assessment ofOutcome on a Neurorehabilitation Unit. Br JOccup Ther 1999; 62: 123-26.

8 Ripat J, Etchevery E, Cooper J, Tate R. A comp-arison of the Canadian Occupational Performance

Measure and the Health Assessment Questionnaire.Can J Occup Ther 2001; 68: 247-53.

9 Chan CCH, Lee TMC. Validity of the CanadianOccupational Performance Measure. Occup TherInt 1997; 4: 229-247.

10 McColl M, Paterson M, Davies D, Doubt L, LawM. Validity and community utility of the CanadianOccupational Performance Measure. Can J OccupTher 1999; 67: 22-30.

11 Pfennings LEMA. Health-related quality of life inpatients with multiple sclerosis. Dissertation VrijeUniversiteit, Amsterdam, 1999.

12 Lankhorst GJ, Jelles F, Smits RC et al. Quality oflife in multiple sclerosis: the disability and impactprofile (DIP). J Neurol 1996; 243: 469-74.

13 Post MWM, De Bruin AF, De Witte L, SchrijversA. The SIP68: a measure of health- relatedfunctional status in rehabilitation medicine. ArchPhys Med Rehabil 1996; 77: 440-45.

14 De Bruin AF, Diederiks JPM, De Witte LP, StevensFCJ, Philipsen H. The development of a shortgeneric version of the Sickness Impact Profile. JClin Epidemiol 1994; 47: 407-18.

15 Polit DF, Hungler BP. Nursing research. Principlesand methods. Philadelphia: Lippincott Williams andWilkins, 1999.

16 Badley EM. An introduction to the concepts andclassifications of the international classification ofimpairments, disabilities, and handicaps. DisabilRehabil 1993; 15: 161-78.

17 Townsend E, Brintnell S, Staissey N. Developingguidelines for client-centred occupational therapypractice. Can J Occup Ther 1990; 57: 69-76.

at UCSF LIBRARY & CKM on November 19, 2014cre.sagepub.comDownloaded from