Improving resident safety in care homes · This project would not have been possible without the enthusiastic support of a large number of people and organisations along with the

Improvingresidentsafetyincarehomes

LearningfromthePROSPERprogrammeinEssex

November2016

2

Foreword

More than18,000carehomes inEnglandprovide support forover 386,000 people. Many of these people have complexhealthcareneeds,withmultiplelong-termconditions,disabilityand frailty. Care home teamswork hard to support residentsbut generally there is a high turnover of staff and limitedinvestment in developing skills. Systematic approaches toimprovingquality,whicharebecomingmorecommonplace in

theNHSandother sectors, arevirtuallyunknown in carehomes.ThePROSPERprogrammewasagroundbreaking initiativetotestwhetherquality improvementmethodscouldbe implemented inthecarehomecontext.

PROSPER was a collaboration between care homes, Essex County Council, the health sector,UCLPartners and Anglia Ruskin Health Partnership. Funded by The Health Foundation, theprogrammefocusedonusingqualityimprovementmethodstoreducepreventableharmfromthreeofthemostcommonsafetyissuesincarehomes:falls,urinarytractinfectionsandpressureulcers.

AteamofqualityimprovementfacilitatorsbasedatEssexCountyCouncilprovidedcarehomeswithtrainingaboutquality improvementmethods, tools to trackchangesover time,andsignpostingtoresources and other training. Supporting the programme was an embedded and formativeevaluationexaminingtheextenttowhichtheinterventionsworkedandtheirmechanismsofaction.Thisformofparticipatoryevaluationisbothnovelandchallengingtodeliver.Asisthenorminthequality improvement field, the intervention flexed and adapted during implementation and ourevaluationhadtoacknowledgeandreflectthisprocess.

WearepleasedtoreportthatthePROSPERinitiativehelpedtoincreaseknowledgeandawarenessofresidentsafetyamongstcarehomestaff,encouragedhomestotestnewapproaches,andinsomehomes,resultedintangiblereductionsinharms.Althoughimprovementswerenotuniversalacrossallparticipatinghomes,atleasthalfofthehomesreportedtheybenefitedfromtakingpart.

We found that the context in which care homes operate is fundamentally important to howimprovementinitiativesareimplementedandevaluated.Theinteractionbetweentheintervention,themethod of implementation and the context is complex and poorly understood. Relationshipsbetweenlocalauthoritiesandcarehomesarecomplex;withpolitical,commercialandsafeguardingsensitivitiesthatreachwellbeyondPROSPERandinfluenceboththeimplementationandevaluation.Carehomestaffhavemanycompetingprioritiessoimprovementinitiativesmaynotalwaysbegiventimeandfocus.

PROSPER’ssuccessesareareflectionoftheenormousgoodwillofallinvolved–carehomes,Councilstaff, healthcare partners and the evaluation teammadeupofUCLPartners,Anglia RuskinHealthPartnership, UCL and The Evidence Centre. Without everyone’s passion and commitment to

3

improving care for people living in care homes, this programme would not have been possible.PROSPER has shown what can be achieved. The challenge now, for all of us, is to spread theselessonsandsuccessessothatawiderrangeofhomesandtheirresidentscanbenefit.

MartinMarshallProfessorofHealthcareImprovement,UCL

PrincipalInvestigatorforPROSPER

4

Acknowledgements

This projectwould not have beenpossiblewithout the enthusiastic support of a large number ofpeople and organisations along with the core PROSPER evaluation team. We are particularlyindebtedtothefollowing:The participating care homes in Essex, including the staff, home managers, residents and theirfamiliesandthehomeowners.The PROSPER Advisory Group comprising of Professor Bryony Dean Franklin, Professor ClaireGoodman,ProfessorSteveIliffe,Dr.YoginiJani,Dr.JamesMountford,ProfessorMaxinePowerandProfessorMikeRoberts.The officers and staff of Essex County Council for their vision and courage in initiating andchampioningtheprojectthroughthickandthin.Research staff providing support for the core evaluation team, including ZahraaMohammed Ali,ChrisSinghandClemWhite.TheHealthFoundationforfundingtheworkandothergrantholdersintheClosingtheGapfundingschemewhowerewillingtosharetheirprojectlearningatjointevents.Theevaluationteamcomprised:

• MartinMarshall,UCLandUCLPartners(principalinvestigator)• DebideSilva,TheEvidenceCentre(qualitativeevaluationlead)• LiWei,UCL(quantitativeevaluationlead)• JamesAnderson,AngliaRuskinHealthPartners• JennyShand,UCLPartners• LesleyCruickshank,EssexCountyCouncil• KieranAttreed-James,EssexCountyCouncil

5

Executivesummary

ImplementingimprovementmethodsincarehomesQuality improvement approaches such asPlan-Do-Study-Act (PDSA) cycles, run charts and cultureassessmenttoolsarebecomingincreasinglypopularintheNHS,butarelessoftenusedinthecarehome sector. From July 2014, Essex County Council and local care homes began testing whetherusingquality improvementtechniquescould improveresidentsafety.Atotalof118homes infourseparatecohortssigned-upto takepart in thePROSPERprogramme(PromotingSaferProvisionofcareforElderlyResidents)and90homesremainedpartoftheprogrammeasofMay2016(76%).

Theprogramme,whichwasfundedbyTheHealthFoundation,providedcarehomeswithtraininginquality improvement methods, a resource toolkit, tools to help monitor change in residentoutcomes, opportunities to share learning with other homes, and support visits from Councilimprovement facilitators.Homesreceivedsix-monthsof targetedsupport thenremainedpartofacommunity of practice, receiving newsletters, invitations to meetings and ad-hoc support. The‘interventions’ evolved over time as the facilitators came to understand more about qualityimprovementmethods,andastheapproachesandcommunicationstrategieswereadaptedforthecarehomecontext.Bytheendofthestudythepackageofinterventionswassubstantiallydifferenttothatoriginallyenvisaged.

EvaluationapproachAparticipatoryevaluationwasundertakenusingamulti-methodpragmaticbefore-and-afterdesignto document the impacts of the programme and lessons learnt. Data were collected using 203telephone interviews, 12 in-person discussion groupswith staff from 90 homes, an online surveywith51homes,49discussionswiththeimplementationteamandotherstakeholders,observationof12programmeeventsandactivities,monthly incident figures from65carehomesandanalysisof523programmedocuments.

Inlinewiththeparticipatoryandformativenatureoftheevaluation,findingswerecontinuouslyfedback to the participants through monthly meetings between the implementation and evaluationteamsandthroughbuildingacloserelationshipbetweenthefield-basedevaluatorsandcarehomestaff.

ImpactsPROSPER aimed to help care homes improve the safety culture, undertake activities to improvesafety, reduceratesof falls,pressureulcersandurinary tract infections,andreducecosts forcarehomesandtheNHS.

ImprovementsincareprocessesInqualitativeinterviewsandanonlinesurvey,two-thirdsofcarehomesreportedchangingsomeoftheircareprocessesasaresultofPROSPER.

6

Thethreemostcommonlyreportedchangeswere:

• proactivelyusingdata tohelp track changesover timeand identify resident safety issues;

• implementing ideas for change, suchas increasing theavailabilityof fluids,using colouredcoastersortumblerstoidentifypeoplewhoneededmorefrequentdrinks,andpersonalisingandoptimisingwalkingframes;and

• making safety issuesmore visible andaprioritybydiscussing them in teammeetings anddisplayingmaterialsonboardsinpublicareasandstaffrooms.

These changes were evidenced during observations by the evaluation team, feedback from carehomes in interviews, an online survey and in progress reports and observations by theimplementationteam.

Homesthatsaidtheyhadnotimplementedanychangesasaresultoftheprogrammewereeitherthose thathadonlyrecently joinedPROSPERoroneswhichdidnot feel that the informationandsupport available via PROSPER added to the knowledge and skills they already had. Homes thatreceived fewer visits fromPROSPER teammembers and those from the initial cohortswere leastlikelytoreportchangestocareprocesses.

ChangesinsafetycultureTwo-thirdsofhomesreportedchangesinsafetyculture.Thisincluded:

• managersandstaffbeginningtoredefinesafetyintermsofpreventionandminimisingharmrather than the regulatory-driven imperatives of reducing safeguarding incidents andinspections.Thisappearstobeasignificantshiftforthecarehomesector;

• greater focus on proactive prevention and monitoring of safety incidents;

• engagingawiderrangeofcarehometeammembersininterventionstoimprovequalityandsafety,includingmorejuniormembersofstaff.Healthprofessionals,andtoalesserextent,families were involved in some homes. Residents were included in a small number ofinitiatives in selected homes, such as a project to personalise walking frames; and

• care staff feelingmore empowered to suggest ideas and support change,whereas in thepastthismayhavepredominantlybeenthedomainofcarehomemanagers.

Thesechangeswereevidencedbyobservationsofthehomes,feedbackfromhomesininterviews,astatistically significantdifference in anonline survey completedwhenhomesbeganPROSPERandagaininMarch2016,andfeedbackfromtheimplementationteam.

Homes in cohort one took more time to experience culture changes, perhaps because theinterventionswerebeingdevelopedwiththiscohort.Withadaptionoftheinterventionsandgreaterexperience of the improvement team, benefits were more quickly reported by cohorts two andthree.

7

ResidentoutcomesTheimpactofPROSPERonresidentoutcomeswasevaluatedusingdatafromcohortsonetothree.64 care homes from these cohorts provided outcome data. Four homes (6%) focusedmainly onreducing pressure ulcers, 17 homes (27%)mainly focused on reducing falls and 13 homes (20%)predominantlyfocusedonreducingurinarytract infections.Fourhomes(6%)focusedonbothfallsand pressure ulcers. 26 homes (41%) did not specify a main focus. Cohort four’s data was notincludedintheanalysisbecausehomesinthiscohortbeganbetweenNovember2015andFebruary2016sotherewerenotenoughfollow-updataavailableatthetimeoftheanalysis.

Twodifferentanalyseswerecarriedout,oneaggregatingdataprovidedbyhomeswhichgaveanydataandonefromthemorelimitedsetofhomesthatprovidedbothpre-andpost-interventiondataforaperiodofsix-to12-monthsbeforeandafterthestartoftheintervention.

Theresultssuggestthat:

• whenusingallavailabledatafromcarehomes(n=64)therewerestatisticallysignificantreductionsintheratesoffallsandpressureulcersbeforeandafterPROSPER,andstatisticallysignificantincreasesinurinarytractinfections,A&Eattendancesandanyhospitaladmissions;

• usingonlydataforthehomeswhichfocusedonreducingspecificsafetyevents,theresultsweresimilartothatofalldataanalysis;and

• usingdataonly from those carehomesprovidingbothpreandpost-interventiondata,nostatisticallysignificantreductioninsafetyeventsorhospitalutilisationwerefound.

Therewerenodifferences in impactsonresidentoutcomesbasedongeographicarea,homesize,whetherhomeshadoneassignedGP, levelofengagementwithPROSPERandcohortnumber.Thedataseemtosuggestsomekindof‘cohorteffect’,forexamplecohortthreehomesseemtomakeadisproportionate contribution to reducing falls whilst cohort one homes seem to have a stronginfluence on the observed changes in pressure ulcers. The reasons for these effects are not clearfromeitherthequantitativeorqualitativedata.

The analysiswas challenging for a number of reasons: (a) it is difficult to attribute change to thePROSPER intervention,not justbecauseof thestudydesignbutbecausethe interventionchangedduringthestudy, (b)an increase in reportingof incidentsasaconsequenceofparticipation in theprogrammemightbeexpectedanddesirablebutthiswouldhideanybeneficialimpact,(c)homesdonot have a long track-record of collecting data, particularly for pressure ulcers and urinary tractinfections and for many of the homes their methods of collecting data are often manual,complicatedandunreliable,and(d) it isdifficulttoreducehospitalutilisationasthisdependsonawiderangeof factors. Inaddition, in interpretingtheresults, theevaluationteamwassensitivetotheriskthatbeforeandafterdesignstendtoover-estimateinterventioneffects.

8

CostsIn the original proposal the evaluation team said that itwould carry out a cost analysis to assesswhethersavingswererealisedasaresultofthePROSPERintervention.WecollateddatarelatingtotheoveralldirectandindirectcostsofrunningthePROSPERprogrammeandthestandardisedunitcosts for each safety outcome.We carried out a preliminary cost analysis based on the residentoutcomeanalysiswhichusedalltheavailabledatafromthehomes.

Theanalysis involvedthesamerisksandlimitationsastheanalysisofresidentoutcomes,togetherwith additional complications around accurate costing of falls, pressure ulcers, and otherharms. Care homes treat these harms in different ways, with different associated costs to thesystem. For this reason, theanalysis shouldbe treatedwithextremecaution,and furtherwork isrecommendedtodevelopamorerobustpicture. Thekeyfindingswere:

• The overall cost of the PROSPER project was £282,000. This includes a significantcomponent of set-up and development costs for the intervention, that would not berequiredforfuturesimilarprojects;

• The total cost savings across PROSPER associatedwith the reduction in falls and pressureulcerswasintherange£86,000-£143,000;and

• These cost savings were offset by the costs of the increased incidence of Urinary TractInfections, A&E attendances, and admissions, which were in the range of £121,000 -£465,000. By far the biggest component of these costs were the increased admissions(£100,000-£442,000)andthismaybeexplainedbyageneralrise inadmissionsfromcarehomesoverthetimeperiodofthePROSPERproject.

SustainabilityCarehomesthatparticipatedinthefirsttwo-yearsreportedthattheywouldcontinuetousesomeof the tools and techniques they learnt. In particular, this included theSafety Cross and graphingtheir incident rates plus changes to care processes, such as undertaking special activities fornutritionandhydrationweek,givingresidents jellytoboosttheir fluid intake,usingmotion-sensorpads inroomsandusingmirrorstocheckskin.SomecohortoneandcohorttwocarehomeshavecontinuedmakingthesechangesintheabsenceofintensivesupportfromthePROSPERteam.

As a result of the successes of PROSPER and promotion and campaigning by the implementationteam,EssexCountyCouncilhasagreed to fundanextensionof theproject foranadditionalyear.Thiswill extend the intervention toa largernumberof carehomesacross thecounty, sustain thecommunityofinterest,thechampions’studydays,andprovidealongertimeframefortrackingtheimpact.

Inaddition,peopleplanningsimilarinitiativesinotherpartsoftheUKhavebeenintouchwiththePROSPERteamandlearningisbeingsharedwidely.TheprocessoflearningisbeingformalisedusingtheNational InstituteforHealthResearchENRICH(EnablingResearchinCareHomes)networkandthroughnationalprogrammessuchasCareQualityCommission(CQC)reviewsinEngland.

9

ImplementationchallengesWhilst two-thirdsof carehomes felt theyhadgainedsomething fromtakingpart inPROSPERandabout half of homes thought that some of the changesmadewould not have happenedwithoutPROSPER,otherhomesreportedthattheyhadnotchangedagreatdealandthattheyfeltliketheywere being asked to give more than they received. This included providing data for evaluationpurposesaswellassharingideasforimprovementtobepassedontootherhomes.Aboutonethirdofhomesthattookpartininterviewsandfocusgroupsexpressedthisconcern,mainlyfromcohortone and two. Almost all homes suggested that there were areas for improvement, in particularregarding the level and consistency of support provided. These comments were less evidentamongstcohortfourthathadaccesstodedicatedmembersoftheimplementationteamemployedsolelytoworkonPROSPER.

Therewasaperception that somecarehomes receivedmore sustained support thanothers. ThiscouldbeduetothevaryingstylesoftheimprovementfacilitatorsassignedtodifferentcarehomesandcapacityissuesfortheCouncil’squalityimprovementteam.Homesthatreportedmoreon-goingcontactwithPROSPERweremostlikelytobesustaininganychangestheymade.Theuseofspecificquality improvement methodologies, such as PDSA cycles, were not seen as something that hadbeenprioritisedbyPROSPERorvaluedby thecarehomes,apart fromtheuseofdatamonitoringtools.

There were tensions between the evaluation team’s desire to present unbiased assessments ofprogress and the implementation team’s desire to celebrate achievements, maintainmomentumandkeepthehomesengaged.

KeysuccessfactorsBearing in mind the qualified successes of PROSPER and learning from the implementationchallenges, the top ten tips that other areas may wish to bear in mind when implementing aprogrammeofthisnatureinclude:

1. Inthecarehomecontext,providingsubstantivenewideasandresourceswasasimportantastrainingandsupportinqualityimprovementmethods;

2. Simple introductory training about quality improvement was useful when it includedexamplesandlanguagerelevanttocarehomesandfocusedonpracticalapplicationratherthantheory;

3. Simpletoolshelpedhomestousedataeffectively.ThisincludedtheSafetyCrossandgraphsshowingmonthlyincidentrates;

4. Providingopportunities forhomes toshare ideas and learn fromeachotherworkedwell,includinghaving regularget-togethers formanagersandcarerswitha ‘taught’ componentbutalsoampleopportunitytosharelearning;

5. Homesreportedthatbeingabletocomparethemselveswithotherhomeswasmotivating,such as through anonymised ‘average’ incident rates andmonthly newsletters. However,anyperceivedjudgementsaboutdifferencesinperformancewerenotwelcomed;

10

6. Havingwaystoengageawiderrangeofcarehomestaff,ratherthansolelymanagers,wascrucial to success. PROSPER ‘champions’ included carers and domestic staff. A ‘train thetrainer’ approach was used so that these champions took responsibility for rolling outlearningtoothers;

7. It is important to allocate enough capacity and capability in the implementation team toprovideregularproactivesupporttohomes.Developingeducationalprogrammesandtoolsand giving support to 90 homes required considerable resource. The facilitatorswere notexperts in quality improvementmethods and this impactedon the speed implementationoccurredandtheextentofexpertiseavailabletohomes,particularlyinearliercohorts;

8. Thereisbenefitfromhavingawidersupportteamtoinputideas,includingcarehomestaff,membersfromelsewhereintheCouncil,healthcareprofessionalsandimprovementexpertsfromtheevaluationteam.JointworkingwithNHScolleagueshasbeenimportantinofferingawiderangeofsubstantivetraining. Jointownershipbythe localauthorityandNHScouldbeworthwhileinthefuture;

9. Onlineforumsandwebsitesmaynotworkwelltoengagecarehomes.Face-to-facecontactor regular telephone follow-upsarepreferredby carehomesmore thanwrittenoronlinematerials;and

10. Itisimportanttoberealisticaboutthetimeittakestoembedchangeandthecapacityandmotivation of homes to use specific tools. This includes acknowledging the burdens thatdata collection can impose. It also includes understanding how change is reported tostakeholders,andensuringthatanyclaimsmadearesubstantiatedandsustainable.

11

Contents

Foreword 2

Acknowledgements 4

Executivesummary 5

1.ThePROSPERprogramme 131.1 Context 131.2 Evaluationapproach 151.3 PROSPERcomponents 16

2.Impacts 292.1 Safetyculture 292.2 Safetyprocesses 322.3 Residentoutcomes 352.4 Costs 542.5 Beneficiaries 55

3.Helpfulandhinderingfactors 583.1 FeedbackaboutPROSPERcomponents 583.2 Implementatonsuccessfactors 613.3 Implementationchallenges 65

4.Implications 694.1 Achievingobjectives 694.2 Keylessons 704.3 SustainabilityforEssexCountyCouncil 724.4 Dissemination 734.5 Conclusion 74

Appendices A1 Evaluationmethods 76A2 Ethicsapproval 84A3 ExamplesoftoolsusedinPROSPER 85A4 Impactsonsafetyculture 92A5 Impactsonsafetyprocesses 99A6 ResultsfromtheCarehomeswithbothPreandpostdatafor6–12months 102

12

A7 PROSPERandUnitCosts 112A8 ImpactsonResidentOutcomes 114A9 FeedbackaboutPROSPERcomponents 116A10 Communicationsplan 131

13

1. ThePROSPERprogramme

Residentialandnursinghomesprovideahomeandcareforsomeofthemostvulnerablepeopleinour communities. Overall, care home staff work hard to provide a clean, safe and caringenvironmentforresidents,butthereisgrowingawarenessofthegapbetweenwhatpeopleaspireto achieve and what actually happens in practice. This report describes the impacts and lessonslearntduringthePROSPERprogrammeinEssex–apartnershipbetweencarehomes,EssexCountyCouncil, thehealth sectorandUCLPartners,working together to improvesafety for residents.TheprogrammewasfundedbyTheHealthFoundation.

This sectionof the reportoutlineswhy suchaprogramme is important,what it comprisedofandhowitwastested.SectiontwoexplorestheimpactsoftheprogrammeandSectionthreefocusesonsome of the factors that have helped and hindered success. Finally, Section four highlights theimplicationsforotherswantingtoimplementasimilarmodelofimprovementandforthePROSPERprogramme itself,whichwill continuewhenTheHealthFoundation fundingcomes toanend.Thereport is basedon informationderived fromaparticipatory evaluationwhich started in July 2014andendedinMarch2016.

1.1 ContextOnehundred-years ago, people in Englandwere expected to live to about 45 to 50-years of age.Now,lifeexpectancyisabout80-years1andthenumberofpeopleaged60oroverisanticipatedtorise by over 50% in the next 25-years.2 Older people often have long-term and age-relatedconditionswhichaffecthowtheirbodiesandmindsfunction.Thismeanshealthandsocialservicesneed tobe ready tocopewithan increase indemand forhigh-quality safecare thathelpspeoplemaintaintheirdignityandqualityoflife.

Around4%ofpeopleaged75to84-yearsliveinresidentialornursingcarehomes.Thisrisestoonein five people aged 85-years and older, and half of 90-year olds.3 Peoplemay live in care homesbecause they need helpwith physical ormental health problems. Care home residents aremorelikely thanothers tohave impairment, frailty and an increasedneed for specialist equipment andsupport.4

There are around 500,000 places in care homes in the UK, 90% of which are managed by theindependentsector.5InEngland,about40%ofcarehavefewerthan10beds,withonly1%ofcarehomesoffering75ormorebeds.6

Care homes are sometimes the subject of negative publicity. They are characterised by frequentpolicyandregulatorychanges,highstaffturnover,limitedopportunitiesforstaffeducationanduse

1 WiseJ.Numberof“oldestold”hasdoubledinthepast25years’.BMJ2010;340:1266.2 OfficeofNationalStatistics,NationalPopulationProjections.London:OfficeofNationalStatistics,2014.3 MyHomeLife/NationalCareHomesResearchandDevelopmentForum.QualityofLifeinCareHomes:A

reviewoftheliterature.London:HelptheAged,2007.4 OfficeofFairTrading.SurveyofOlderPeopleinCareHomes.London:OfficeofFairTrading,2005.5 CareofElderlyPeopleUKMarketSurvey2010.London:LaingandBuisson,2010.6 SkillsforCare.TheStateoftheAdultSocialCareWorkforceinEngland2010,London:SkillsforCare,2010.

14

of information for improvement. These factors may influence the quality of life of residents,includingtheextenttowhichtheyexperiencecarethatissafeandensuresdignity.

Research from the UK and around the world highlights that people living in care homes mayexperience preventable safety concerns, including medication errors, falls, pressure ulcers andurinary tract infections.7,8 In recent years, a number of programmes have attempted to improveresident safety, for example through staff education, computerised records, decision supportsystemsandpartnershipswithprimarycare.9,9a,10,10a,11,12,13,14Theyhavehadvaryinglevelsofsuccessandhaveoftenencounteredchallengesinimplementation.Ratherthanreplicatingproblem-focusedimprovementapproaches,EssexCountyCouncilwantedtotest a more innovative approach; up skilling care home teams in quality improvementmethodologiessoteamscouldapplytheseapproachestoawiderangeofdifferentsafetychallengesandtherebyimprovetheoverallqualityandsafetyofresidentcare.PROSPER is not the first systematic attempt to improve quality in care homes in Essex. In recentyears,carehomeshaveworkedwiththelocalauthorityandNHStodevelopnewskillsandimprovethe quality of care in a number of improvement programmes. In particular,My Home Life is anationalorganisationaldevelopmentandleadershipinitiative.Essexwasoneofthefirstareasthatimplementedtheapproachin2009.Followingonfrominitiativessuchasthis,carehomesandEssexCountyCouncil identifiedaneedtofocusmoreexplicitlyonimprovingthesafetyofresidents.ThisledtothedevelopmentofthePROSPERprogramme.

7 StubbsB,DenkingerMD,BrefkaS,DallmeierD.Whatworkstopreventfallsinolderadultsdwellinginlong

termcarefacilitiesandhospitals?Anumbrellareviewofmeta-analysesofrandomisedcontrolledtrials.Maturitas2015;81(3):335-342.

8ModyL,KreinSL,SaintS,MinLC,MontoyaA,LansingB,McNamaraSE,SymonsK,FischJ,KooE,RyeRA,GaleckiA,KabetoMU,FitzgeraldJT,OlmstedRN,KauffmanCA,BradleySF.Atargetedinfectionpreventioninterventioninnursinghomeresidentswithindwellingdevices:arandomizedclinicaltrial.JAMAInternMed2015;175(5):714-723.

9Buljac-SamardzicM,vanWijngaardenJD,Dekker-vanDoornCM.Safetycultureinlong-termcare:across-sectionalanalysisoftheSafetyAttitudesQuestionnaireinnursingandresidentialhomesintheNetherlands.BMJQualSaf(PublishedonlineaheadofprintJuly2015).

9a GoodmanC,DeningT,GordonAL,etal.Effectivehealthcareforolderpeoplelivinganddyingincarehomes:arealisticreview.BMCHealthServicesResearch2016;16:269DOI:10.1186/s12913-016-1493-4

10 ModyL,MeddingsJ,EdsonBS,McNamaraSE,TrautnerBW,StoneND,KreinSL,SaintS.EnhancingResidentSafetybyPreventingHealthcare-AssociatedInfection:ANationalInitiativetoReduceCatheter-AssociatedUrinaryTractInfectionsinNursingHomes.ClinInfectDis2015;61(1):86-94

10aSzczepuraA,NelsonS.WildD.In-reachspecialistnursingteamsforresidentialcarehomes:uptakeofservices,impactoncareprovisionandcost-effectiveness.BMCHealthServicesResearch2008;8:269

11 LimRH,AndersonJE,BucklePW.WorkDomainAnalysisforunderstandingmedicationsafetyincarehomesinEngland:anexploratorystudy.Ergonomics2016;59(1):15-26.

12MarasingheKM.Computerisedclinicaldecisionsupportsystemstoimprovemedicationsafetyinlong-termcarehomes:asystematicreview.BMJOpen2015;5(5):e006539.

13 ZúñigaF,AusserhoferD,HamersJP,EngbergS,SimonM,SchwendimannR.TherelationshipofstaffingandworkenvironmentwithimplicitrationingofnursingcareinSwissnursinghomes--Across-sectionalstudy.IntJNursStud2015;52(9):1463-1474.

14JiangT,YuP.Theimpactofelectronichealthrecordsonclientsafetyinagedcarehomes.StudHealthTechnolInform2014;201:116-123.

15

1.2 EvaluationapproachPROSPERwas evaluated over a 21-month period, fromwhen implementation began in July 2014through toMarch 2016. The evaluationwas undertaken by a team comprisingUCL, UCLPartners,Anglia Ruskin Health Partnership, Essex County Council and The Evidence Centre. The aim of theevaluationwastoexaminetheimpactofthePROSPERprogrammeandthefactorswhichinfluencedwhetherornottheprogrammewassuccessful.Theevaluationquestionswere:

1.WhatwasthePROSPERintervention?• Whatwasinvolvedintheinterventionandhowwasitimplemented?

2.WhatimpactdidPROSPERhave?• DidPROSPERimpactonsafetyculturewithincarehomes?• DidPROSPERimpactonsafetyprocesseswithincarehomes?• DidPROSPERimpactonresidentoutcomes?• DidPROSPERimpactoncostsforcarehomes,socialcareandtheNHS?

3.WhatinfluencedtheimpactofPROSPER?

• Whathelpedorhinderedtheintervention?• Whatwouldhelpembedandsustainanybenefitsderivedfromtheprogramme?

The evaluation also considered whether learning from improvement initiatives in the care homesectorhadimplicationsforsafetyimprovementintheNHS.Toaddresstheevaluationaims,amulti-method‘before-and-after’designwasused,withqualitativeandquantitativecomponentsandastrongorientationtoparticipatorymethods.

Quantitativemethods explored the impact of the intervention on outcomes for residents in carehomesthatprovidedmonthlydata.Theevaluationteamanalyseddataprovidedbycarehomesforthe pre- and post-intervention periods relating to the rates of falls, pressure ulcers, urinary tractinfections,emergencydepartmentadmissions,anyhospitaladmissionsandhospitaladmissionsduetoafalls.

Qualitativemethodswereusedtodescribetheintervention(s),understandstaffexperiencesoftheintervention(s) and to examine impacts on organisational processes and safety culture. Methodsincludedobservingprogrammeactivities, interviewing carehomemanagers and staff, and visitingcarehomestoobserveandlistentoteams.AnonlinesurveywasalsoundertakenatthestartoftheprogrammeandagainattheendinMarch2016.Insummarythequalitativedatacollectionmethodscomprisedof:

• reviewof523documents• 127hoursworthofhomevisitsandobservations• 203individualtelephoneinterviews• 12discussiongroupswithcarehometeams• anonlinesurveywith51carehomeswhentheybeganPROSPERandagainattheendofthe

data collectionperiod inMarch 2016 (betweeneight- and 20-months later, dependingonthecohort)

16

• 26meetings,discussiongroupsandinterviewswithPROSPERteammembers• 23interviewswithotherstakeholders,includingGPs,NHSteamsandEssexCountyCouncil• 10interviewswithcarehomesnottakingpartinPROSPER

TheevaluationmethodsaredescribedindetailinAppendix1.EthicsapprovalwasgrantedbyEssexCountyCouncilandresearchgovernanceapprovalwasgrantedbyUCL(seeAppendix2).1.3 PROSPERcomponentsThissectiondescribes the initialaimofPROSPER,how itwas implementedandhowtheapproachchangedovertime.Dataforthissectionwassourcedfrom:

• reviewof523documents• 127hoursworthofhomevisitsandobservations• 203individualtelephoneinterviews• 12discussiongroupswithcarehometeams• 26meetings,discussiongroupsandinterviewswithPROSPERteammembers

AimsPROSPERaimedtoimprovesafetyforcarehomeresidentsinEssexbysupportingstafftousequalityimprovement approaches. Box 1 describes the aims as outlined in the original proposal to TheHealthFoundation.

Box1:AimofPROSPERasoutlinedinoriginalproposalforfunding

Aim: To improve safety and reduceharm for carehome residents acrossNorth East andWestEssex by implementing a multi-faceted improvement intervention with educational,measurementandculturechangecomponents.Undertheoverallaim,specificobjectivesareto:

• Co-design solutions with residents and their relatives and introduce new QI trainingopportunitiestobuildstaffcapabilitiestoaddresssafetyconcerns;

• ReducethepercentageofresidentsthatpresentatA&E;• Increase the proportion of residents who are ‘harm-free’ (as defined by the NHS Safety

Thermometer);• Reducetheprevalenceoffalls,pressureulcersandcatheterinfectionsacrosscarehomes;• Increase staff understanding of, capacity and capability to create a safe environment for

residents;• Understandtheconsequenceonhealthandsocialcarecostsasaresultoftheabove;and• Establishanevidencebasefortheintervention

Source:EssexCountyCouncilfundingapplicationtotheHealthFoundation

TheoryofchangePROSPER was based on the theory of change that supporting care homes to use qualityimprovementmethods and tounderstand their safety culturewould lead to changes in how carehomestaffthinkandact,andtotheenvironmentinwhichtheywork.Itwashopedthatthiswouldimpact thequality of care provided; improving resident safety and reducing costs for care homesandtheNHS.Figure1illustratesthistheoryofchange.

17

Figure1:PROSPERtheoryofchange

ComponentsThethreemaincomponentsoftheplannedPROSPERprogrammewere:

1.Educationaboutqualityimprovementmethods,suchasPDSAcyclesandruncharts.Thiswastobedeliveredusingan initial trainingsession in improvementmethods,monthly follow-upvisitsbyimprovement facilitators for six-months, developing and distributing a toolkit of resources, andsetting-up community of interest practice events where care homes shared their learning everythree-tosix-months;

2.UsingtheManchesterPatientSafetyFramework (MaPSaF)tooltohelpcarehomesunderstandtheirsafetyculturetoidentifyareasfordevelopment;and

3. Using theNHS Safety Thermometer to help care homes collect and analyse data about safetyincidentssotheycouldidentifyareasforimprovementandtrackchangesovertime.

Quality improvement approaches such as PDSA cycles, run charts, data feedback and cultureassessment tools have been used in the NHS and other sectors for many years, but are notcommonlyusedincarehomes.ExamplesofthesetoolscanbefoundinAppendix3.

PROSPER• educationaboutimprovementmethods• implementingsafetyculturetool• trackingincidentdataovertime

Improvedbehaviour(safetyprocesses)

Improvedsafety(reducedfalls,pressureulcersandurinarytract

infections)

Reducedcosts forhomes,

socialcareandthehealthsystem

Improvedthinkingaboutsafety(safetyculture)

18

Care homeswere invited to focus on one ormore of threemost commonharms experienced byresidents: falls, pressure ulcers or urinary tract infections. Areas were selected according to theprioritiesofthecarehomesbutthesechangedoverthedurationoftheinitiative.

ChangestotheinterventionanditsimplementationTheoriginalcomponentsofPROSPERwereimplementedtovaryingextents.Asthestudyprogressedand learning emerged, Essex County Council improvement team introduced a number of newinterventionsthathadnotbeenenvisagedatthestart,andinsomecases,reducedthefocusontheoriginal interventions. In particular, at the care homes’ requests they introduced training aboutsubstantivecontentareas,suchasmedicinesmanagement(particularlywhenlinkedtoriskoffalls)anddiabetes,extendedtrainingsupportforlongerthantheoriginalsix-monthsandnewtools,e.g.the Safety Cross. At the same time they put less emphasis on theNHS Safety Thermometer andMaPSaFbecauseneitherwereperceivedasusefuloreasytouse.

The PROSPER implementation team decided early on to move away from solely testing theimplementationofquality improvementmethodsandexamining safetyculture towardsabroaderandmoreflexiblemodelofsupport.Thisflexibilityiscompatiblewiththeliteratureonimprovementinitiatives.15,16,17,18,19,20

Thechangeswereinpartaresultofresponsivenesstotheneedsandrequestsofthecarehomesbutalsoreflectedpragmaticconstraints.Sometoolsappearednottohavebeenimplementedbecausethe implementation team had insufficient resources, inadequate training, felt uncomfortablefacilitatingthem,e.g.MaPSaF,orbecausethecarehomesdidnotseethemasapriority.

As outlined in Appendix 4 not every home had the same opportunity to test all interventions.Sometimesthereweretensionsbetweenwhatthecarehomeswantedandwhattheimprovementteamhadthecapacityorcapabilitiestodeliver.

Table1.1comparestheoriginalandfinalcomponentsofPROSPER.

Table1.2liststhenumberofadditionaltrainingsessionsoffered.Thesetopicswereselectedpartlybecause they were a focus of PROSPER, but in other instances they were selected due to theavailabilityofhealthpartnerstofacilitatetraining(aspartoftheirexistingworkpriorities).

15 ArlingPA,AbrahamsonK,MiechEJ,InuiTS,ArlingG.CommunicationandeffectivenessinaUSnursing

homequality-improvementcollaborative.NursHealthSci2014;16(3):291-297.16 RantzMJ,Zwygart-StauffacherM,FlesnerM,HicksL,MehrD,RussellT,MinnerD.Theinfluenceofteams

tosustainqualityimprovementinnursinghomesthat"needimprovement".JAmMedDirAssoc2013;14(1):48-52.

17 HockleyJ.Learning,supportandcommunicationforstaffincarehomes:outcomesofreflectivedebriefinggroupsintwocarehomestoenhanceend-of-lifecare.IntJOlderPeopleNursing2014;9(2):118-130.

18 PuxtyJ,BranderRA,MurphyS,ByrnesV.Promotingqualityimprovementinlong-termcare:amulti-sitecollaborationtoimproveoutcomeswithpneumonia,falls,bacteriuriaandbehaviouralissuesindementia.HealthcQ2012;15(2):70-74.

19 OuslanderJG,LambG,TappenR,HerndonL,DiazS,RoosBA,GrabowskiDC,BonnerA.Interventionstoreducehospitalizationsfromnursinghomes:evaluationoftheINTERACTIIcollaborativequalityimprovementproject.JAmGeriatrSoc2011;59(4):745-753.

20 RantzMJ,Zwygart-StauffacherM,HicksL,MehrD,FlesnerM,PetroskiGF,MadsenRW,Scott-CawiezellJ.Randomizedmultilevelinterventiontoimproveoutcomesofresidentsinnursinghomesinneedofimprovement.JAmMedDirAssoc2012;13(1):60-68.

19

Further detail about the stakeholders perception of each intervention and details ofwhy specificinterventionsweremodifiedorstoppedcanbefoundinAppendix9.

Table1.1:Extenttowhichoriginallyplannedinterventionswereimplemented

Key:Green-completedasplanned;Amber-partlycompletedasplanned;Red=notcompleteddoneasplanned

Component Original/added

Implementation Uptake

NHSSafetyThermometer Original Implementedwithcohortoneandofferedtoallofcohorttwo.SafetyCrosswasimplementedfromcohorttwoonwards.Inaddition,aMonthlyMappingtoolwasaddedfromcohortthree.

TwothirdsofcohortonetriedtheThermometer.Aboutonethirdinputdata.80%ofhomesreportedusingtheSafetyCross.60%supplieddatafortheMonthlyMappingtool.Thisincreasedwhenhomeswereencouragedbytheevaluationandimplementationteams.

Staff,residentsandrelativesactivelysharingperformanceontheNHSSafetyThermometerandco-creatingsolutionsforon-goingimprovement

Original PROSPERdidnotfocusheavilyonwaystoinvolveresidentsandrelativesinco-creatingsolutions.HomesshareddatawithstaffandhadPROSPERnoticeboardswhereresidents,familiesandstaffcouldseeinformation.

Fewerthan10%ofcohortonesharedNHSSafetyThermometerdata.80%ofhomesusedtheSafetyCrossanddisplayedthisforstaff,residentsandfamiliestosee.60%displayedgraphsfromtheMonthlyMappingtool.

Carehomestafftrainedinimprovementmethodologies

Original Qualityimprovementtrainingwasprovidedandrevisedtobemorerelevanttocarehomes.

90homestookpartintraining.Incohortonethiswaslargelyhomemanagers.Fromcohorttwoonwardssomeseniorcarersalsoattended.

Participantsabletodeliverthetrainingtopeers(train-the-trainer)

Original Formaltrain-the-trainermodelwasnottestedbutPROSPERchampionswereimplementedtorolloutlearningtoothers.

Championshavebeenfoundtoworkwelltospreadlearninginformally.

Interventiontoolkitcontainingacompendiumofevidenced-basedinterventionsforeachofthedomainsoftheNHSSafetyThermometer

Original Toolkitwithworksheetsandinformationsheetsdeveloped.Informationfocusesonfalls,pressureulcersandurinarytractinfections.

Allhomesreceivedacopyandthiswasavailableonline.

SafetycultureassessedusingtheMaPSaFtoolatthreetimepoints(before,duringandafterPROSPER),usingthetooltounderstandandaddressbarrierstochange

Original MaPSaFrevisedandtestedindifferentwayswithvariouscohorts.PROSPERhasnotfocusedonusingthisasakeytooltoshapethework.

Twohomesfromcohortonehadfacilitateddiscussions.10homescompletedasurvey-typeform.Nohomesfromcohorttwoorthreeusedthetool.Arevisedversionwastestedwithasmallsampleofhomesfromcohortfour

Communitiesofpractice Original Aproject-widecommunityofpracticeeventwasruninNovember2014andagaininMay2015.

Allhomeswereinvited.Abouttwothirdsattendedthelargerevents.Abouthalfofhomes

21

Component Original/added

Implementation Uptake

Thesecoveredallhomes.InOctober2015,smallercommunityofpracticeeventswereheldonamorelocalbasis

attendedthelocalevents.

Improvementtoolsandcasestudiesuploadedtoresourcetoolforpeerlearning

Original Knowledgehubsetup.Documentswereloadedeverysooften,mainlycopiesofthingssentbyemail.

10%(9)ofcarehomessignedup.Noneposted.

On-goingscheduleofinteractionthroughmeetingsandtelephoneconferences

Original Facilitatorsvisitedhomeswithvariedregularity.Duringtheintensivephase,somehomeswerevisitedmonthlyandotherseverythreetofourmonths.Grouptelephoneconferenceswerenotused.Adhoctelephonecallsweremadetoindividualhomesifneeded.

Adhoc.Somehomesreceivedregularsupportandothersdidnot.Somehomesreportedthattheyhadnocontactwiththeirallocatedimprovementadviserforsixmonths.

SafetyCrossfordisplayinginformationaboutmonthlyincidents

Addition Usedfromcohorttwohomesonwardsthenalsorolledouttocohortone.

Allhomesreceivedacopyaspartofthetoolkit.About80%reportedusingit.

Provisionofgraphswithmonthlydatatotrackchangesovertimeandcompareaverages

Addition Allhomeswereinvitedtoprovidedataaboutthemonthlyincidenceofharms.Fromcohortthreeonwards,homesweregivenaccesstoanonlinetool.

About60%ofhomesprovidedsomedata.Onequarterusedthetoolregularlywithoutprompting.

Provisionofresourcessuchasposters,certificates,mirrorstoviewpressureulcers,andothertangibleresources

Addition Resourcesdevelopedadhoc. Homesofferedtoolsduringcommunityofpracticeandvisits.VariableuptakedependingonPDSAcycles/focus.Resourceswereappreciated.

Provisionofadditionaltraining,suchastrainingaboutinfectioncontrolandpressureulcers

Addition 26trainingsessionsrun. Aroundhalfofhomeshavetakenpartinthisextratraining.

Co-ordinationwithpartnerorganisationssuchasCCGsandNHSproviderorganisations

Addition VariesdependingonCCGarea. VariesdependingonCCGarea.

Monthlynewsletter Addition SenttoparticipatinghomesmonthlyfromJanuary2015.

60%ofhomemanagersreportedreadingit.

22

Table1.2:AdditionaltrainingfacilitatedundertheauspicesofPROSPER

Topic Numberoftrainingsessionsorganised WastrainingrunbyPROSPERteamorNHSpartners?

Diabetes 1sessionfor20homes DeliveredbyhealthpartnerFalls 8sessionsfor6homes Deliveredonceperweekforeightweeksby

healthpartnerInfectioncontrol 1sessionfor13homes Deliveredbyhealthpartner Pressureulcers 1sessionfor4homes DeliveredbyhealthpartnerUTIsandhydration 1sessionfor11homes DeliveredbyhealthpartnerChampion’studydays 1dayfor13homes

1dayfor10homes1dayfor13homes1dayfor13homes1dayfor14homes1dayfor13homes

DeliveredbyPROSPERwithhealthpartners

Therewasalsoa change in the implementation timeline. Theoriginalproposal indicated that116care homes would participate in training in quality improvementmethods in 2014, with supportprovided to undertake PDSAs and to use other improvement approaches throughout 2015. Thuscarehomeswerescheduledtobepartoftheprogrammeforafulltwo-years,fromJuly2014toJune2016,allowingtimeforskillstobuildandtoolstoembed.

Forpracticalreasons,thePROSPERimplementationteamdecidedonaphasedapproach,withfourseparate cohortsof carehomes spaced three to six-monthsapart. Each cohort receivedabout sixmonthsof support (which includedvisits to thehomeeveryone- to four-months), tailoringoff toquarterly or six-monthly group gatherings. This stepped approach allowed the PROSPERimplementationteamtotestoutandthenadapttheapproacheswhilstproperlyresourcingthem,soreducingtherisk forEssexCountyCouncilofmakingmistakesatscale.However italsoresulted insomecarehomesreceivinglesssupportthanoriginallyenvisioned,andin lessdatabeingavailableforoutcomesanalysisforonethirdofparticipatingcarehomes(cohortfour).

ParticipationMembers of the PROSPER implementation team invited care homes to participate on a voluntarybasis,basedonbeingidentifiedthroughCouncilnewslettersorontheteam’sknowledgeofthecarehomes from other local initiatives, CQC inspection reports and other internal intelligence. Thus,initially,existinglinkswereakeysourceofsampling.Inevitablythen,manyofthecarehomesintheearly cohortswere known to the Council and had been part of other improvement initiatives. Astimemovedon, recommendations fromotherstakeholdersbeganto targetcarehomesthatwerethoughttohavemosttogainfromimprovementsupport,e.g.carehomesthatmayhavehadalargenumberoffalls.Carehomeswerenotspecificallytargetedbasedonsizeorlevelsofperformance.

PROSPERwasoriginally scheduled to focuson twopartsofEssex,butwasexpanded to cover theentire county following requests from managers in other localities. This wider geographic scopemeant that the programme spanned the boundaries of the five clinical commissioning groups(CCGs).

ThestartdateandnumberofcarehomesineachofthefourcohortsisshowninTable1.3.

Table1.3:NumberofcarehomesparticipatinginPROSPER

Cohort Startdate Numberofhomes1 July2014 182 January2015 183 July2015 214 November2015-February2016 33

Atotalof118carehomesinitiallysigneduptotakepartandofthese90(76%)continuedwiththe

24

initiative.Ofthe28carehomesthatdecidednottocontinue,twowithdrewimmediatelybeforeandeightafter the initialquality improvement training. 18carehomesstayedpartof theprogrammebutwerenotactive,withdrawingwhenamanagerchangedorwhencarehomesfeltthattheirtimeneededtobedevotedtootherpriorities.InterviewswiththesecarehomessuggestedthattheyfeltPROSPERwouldbetootime-consumingordifficult.Fourofthesecarehomesmovedontotakepartinlatercohorts,thoughtheyremainedrelativelyinactive.

Thelowertotalnumberthanplannedrelatesmainlytothecapacityoftheimprovementfacilitationteamratherthanalackofinterestfromthecarehomes.EssexCountyCouncilplanstocontinuetheprogrammeforatleastoneyearafterfundingfromTheHealthFoundationends,withfurthercarehomesexpressinganinteresttotakepart.

Table 1.4 lists the characteristics of care homes that took part. Essex has a higher than averageprevalence of older people. There aremore than 600 care homes in Essex, 277 ofwhich providebothnursingandresidentialcareforolderpeople.Alloftheseolderpeople’scarehomesarerunbytheindependentsectorratherthanthelocalauthority.Comparedtothenationalpicture,Essexhasfewer small care homes. Table 1.5 compares the characteristics of participating and non-participating care homes in Essex. The breakdown of residential versus nursing care provided byparticipatingcarehomeswasbroadlysimilartoothers inEssex.21However,somecohorts includedmorecarehomesfromlargecorporations.22

Themedian number of residents per home ranged from 31 (Interquartile range (IQR), 23-36) forcohortthreeto44(IQR,38-50) forcohortone.Theproportionof femaleresidentsamongcohortsone to four were 73%, 66%, 72% and 68%, respectively. There were no statistical differencesbetween the four cohorts for type of care (residential or nursing), ownership, proportion of oldresidents, geographical location, proportion of occupancy capacity, home-owned in the UK,proportion of residents with dementia, proportion of residents wholly-funded by the Council,proportionof residentspart-fundedby theCouncil,proportionof residents in theirhomeforoversix-monthsandwhetherresidentshadtheirownGP(Table1.4).

Throughout thisdocument, carehomesare referred toas small,mediumor large. Fordescriptivepurposes small care homes refer to thosewith 30 or fewer residents,medium-sized care homeshaveupto70residentsandlargecarehomesgreaterthan70residents.Thisisadistinctionmadeinpublishedliterature,aswellasrecommendedbythePROSPERevaluationadvisorygroup.23

21 www.cqc.org.uk/search/services/care-homes?location=essex&sort=default&la=&distance=15&mode=html22 www.cqc.org.uk/content/how-get-and-re-use-cqc-information-and-data#directory23 www.pssru.ac.uk/archive/pdf/dp2815.pdf

Table1.4:Characteristicsofparticipatingcarehomes Cohort1 Cohort2 Cohort3 Cohort4 Overal

lPROSPERstartdate July2014 January

2015July2015 Nov2015–

Feb2016

Numberofparticipatingcarehomesthatprovideddata*

17 16 13 18 64

Mediannumberofresidents(interquartilerange)

41(36-77)

38(31-45)

31(23-36)

44(38-50)

39(30-51)

Averageproportionoffemaleresidents 73% 66% 72% 68% 71%LocationMidEssexNorthEastEssexSouthEssexSouthWestEssexWestEssex

19%31%25%6%19%

6%44%13%6%31%

9%55%18%9%9%

39%11%17%11%22%

20%33%18%8%21%

OwnershipOneprivateowner,e.g.familyOwnedbysmallgroup(2or3homes)Ownedbylargecorporation

41%24%35%

6%13%81%

50%17%33%

28%22%50%

30%19%51%

HowmanyhomesownedintheUKbysameowner12-56-1011-2021+

12%44%00

44%

07%13%33%47%

0

64%0

18%18%

17%22%6%11%44%

8%32%5%15%40%

TypeofcareprovidedResidentialcareNursingcareBothresidentialandnursingcare

69%6%25%

80%0

20%

80%10%10%

56%22%22%

70%10%30%

Averageproportionofresidentsover80 82% 82% 78% 76% 80%Averageproportionofresidentswithdementia

51% 57% 63% 57% 57%

Average%residentsfullyfundedbylocalauthority

54% 46% 50% 47% 49%

Average%ofresidentsinhomeforoversix-months

80% 72% 64% 82% 76%

EachresidenthastheirownGPOneGPpracticecoverswholehomeAsingleGPpracticeformostresidentsbutsomeresidentshaveadifferentGP

38%25%37%

27%20%53%

46%18%36%

35%41%24%

36%27%37%

*Note:Figuresbasedondataprovidedbyhomes(26homeschosenottoprovidebackgrounddata)

26

Table1.5:Comparisonbetweenparticipatingandnon-participatingcarehomesinEssexandnationally PROSPERcare

homesn(%)

CarehomesinEssexn(%)

CarehomesinEnglandn(%)

Geographicallocation

MidEssex 12(20%) 126(19%) --

NorthEastEssex 20(33%) 324(50%) --SouthEssex 11(18%) 133(20%) --

SouthWestEssex 5(8%) 31(5%) --

WestEssex 13(21%) 40(6%) --

Ownership

Oneprivateowner,e.g.family 19(30%) 172(41%) 6190(36.2%)Ownedbysmallgroups(2/3homes) 12(19%) 84(20%) 4526(26.5%)

Ownedbylargecorporation 32(51%) 165(39%) 6360(37.3%)

Typeofcare

Residentialcare 41(70%) 351(73%) 12372(72.5%)

Nursingcare 6(10%) 77(16%) 4411(25.8%)Bothresidentialandnursingcare 12(20%) 53(11%) 293(1.7%)Note:ThesedataweresourcedfromtheCQConlinedatabase.Ownershipdataareavailablefor421carehomesinEssex.Dataaboutthetypeofcareprovidedareavailablefor481carehomesinEssexand64ofthePROSPERcarehomes.Itincludesdatafromhomesthatpredominatelycateredforspecialistresidentgroups,suchasthosewithlearningdisabilitiesandthesewerenottargetedbyPROSPER.

FacilitationThePROSPERimplementationteamwasmadeupofaprojectleadanduptosixfacilitators,mostofwhomwere part-time. Table 1.6 lists the number of PROSPER teammembersworkingwith eachcohort.

Table1.6:WholetimeequivalentstaffinPROSPERimplementationteam

Cohort Numberofhomesoriginallytargeted

Numberofhomestakingpart

NumberinPROSPERimplementationteam(WTE)

1 28 18 1managingandfacilitating2 25 18 1managing,2.5facilitating3 30 21 1managing,2.5facilitating4 35 33 1managing,2.5facilitating

AllstaffwerepartoftheEssexCountyCouncilqualityimprovementteam.Initially,theprojectleadwasassignedtoworkanaverageofthree-daysperweekonPROSPER,thoughinpracticesheworkednearly full-time on the project. From November 2014 to June 2015 there was one additionalfull-timeBand4workingonPROSPERwithotherBand4staffspendingabout5%oftheirtimeontheproject on an ad-hoc basis. In the final quarter of 2015, two Band 3 staff joined the team asimprovementassistantsandspentalloftheirtimeonPROSPERactivities.Together,theprojectlead,facilitators and improvement assistants are referred to as the ‘implementation team’ throughoutthisreport.

UCLPartners staff provided quality improvement training for the first cohort but the Esseximplementationteamtookoverresponsibilityfortheothercohortsandprovidedindividualsupportvisitstothecarehomes.

Originallytheaimwastovisitcarehomesmonthly foruptosix-months,but inrealitycarehomesreceived between one and five visits, spanning up to a two-year period. On average, each homereceivedtwovisitsduringthefirstsix-months,butthisincreasedincohortfourwhendedicatedstaffwereemployedontoPROSPER.

Theteamworkedwithotherservices,includingsomefromtheNHS,tooffertrainingandresources,identify interventions to support the care homes, run a community of practice event every six-months, attendmeetings and eventswithin Essex and externally to promote PROSPER and sharelearning.

The implementation team were supported by UCLPartners providing advice about qualityimprovementapproaches,byaprojectgroupmadeupofstakeholdersfromtheCouncil,NHS,carehomesandbyasteeringgroupmadeupprimarilyofseniorCouncilstakeholders.

AnothercorecomponentofPROSPERwastheevaluationteamcomprisingofindividualsfromUCL,

28

UCLPartners,AngliaRuskinHealthPartnership,TheEvidenceCentreandEssexCountyCouncil,whoworked inaparticipatorymannertofeed in informationtoassist in theprogrammedevelopment.Together,theimplementationteam,evaluationteamandcarehomesmadeupthePROSPERteam.

29

2. Impacts

This section summarises key learning about the impacts of PROSPER after 21-months ofimplementation.Asummaryofkeyfindingsarereportedinthismainbodyofthereport.Additionaldetailisprovidedintheappendices.

2.1 SafetycultureDataforthissectionweresourcedfrom:

• reviewof523documents• 127hoursworthofvisitsandobservations• 203individualinterviewswithcarehomestaff• 12discussiongroupswithcarehometeams• onlinesurveywith51carehomeswhentheybeganPROSPERandeight-to20-monthslater

Safety culture refers to the attitudes, beliefs, perceptions and values that care home staff shareaboutthe‘wayofdoingthings’relatedtosafety.Interviewswithstaffandmanagersfrom67outofthe 90 care homes taking part suggested that PROSPER helped change how teammembers viewsafety.Thiswasnotconfinedsolelytoseniormanagement.Two-thirdsofcarehomesself-reportedthatachangehadoccurredinhowmanagersandstaffthinkaboutsafety intheorganisation.Thismanifestedpredominantlyin:

• redefiningsafetyasbeingaboutminimisingriskforresidentsratherthanavoidingsanctionsforstaff;

• trackingchangesinincidentratesovertimeandusingthemtoconsiderwaystoimprove;

• involvingawiderrangeofstaffindiscussionsabouthowtominimiserisk,suchasinformingdomesticstaffthatthedrinkstheyprovidedhelpedtokeepresidentshydratedandtherebytoavoidsafetyincidents;

• greatervisibilityof improvementefforts,suchasdisplayingincidentdataandothersafety

informationonboardsinreceptionareas,residentloungesorstaffrooms;

• recognisingthevalueofsharingideaswithothercarehomes,ratherthanviewingthemascompetitors;

• adesiretocontinueusingspecifictoolssuchastheSafetyCrossandmonthlygraphsonce

PROSPERendsasthetoolswerethoughttohelpimprovethelivesofresidents;and

• beginning to involve families and residents in improvement discussions at some carehomes,particularlyarounddisplayboards.

SeeAppendix4 for furtherdetailsabout impactsonsafetyculture.Box2describes the journeyofonehome.

30

Box2:Onehome’sjourneytowardsimprovedsafetyculture

“WestartedPROSPER[in2015]andIdidn’tknowwhattoexpectatall.Itwasabitdaunting

atfirstwithallthenewwordsandconcepts.Wedidn’tknowifwewantedtodoitbecauseit

sounded complicated.Oncewewent to the second studyday though, things started to fall

intoplace.

PROSPERhasdefinitelyhelpedus to focusmoreand lookatother things.Weare trying to

reduce falls andwe’ve setourselvesa target toachieve.Weareworkingmorewith family

membersaboutgettinggoodfootwearandwe’regettingstafftocleanpeople’sglassesand

makesurepeopledon’tfeeldizzywhenthey’regoingtomove.Youmightthinkthatstopping

falls is all aboutmobility problems and weakness, but there is a lotmore to it than that.

PROSPER has helped us think of all the things that could contribute. It has broadened our

minds.

PROSPERhashelpedtheseniorteamfocusonsafetymoreandweinvolveawiderrangeof

staffinthings;likecleanersaskpeopletohaveadrinknow.Weputtogetheraboardforstaff

andrelativesaboutthingstoask.Lastyearotherhomeswentthroughthefirstcohortsowe

stoletheirideasaboutputtingupboards.It’sgoodwecanalllearnfromeachother.Butit’s

not just the thingswearedoingthatmakeadifference it is sortofhowweare thinking, if

thatmakes sense. Staffarea lotmoreawareof thingsand lookout for thingsnow. It has

almostbecomewhatwejustdoeveryday.

Familiesseetheboardasitisinthereceptionareaandhasinformationaboutpressureareas

and so on, sowe are educating familymembersmore now.Wewould not have done this

before.Someofthefamilieshavebeenabitscaredwhentheyseeallthenittygrittybutthey

aresupportiveofwhatwearetryingtodo.Itopensthedoorsforconversations.Likeweare

allononeteam.

ThemonthlySafetyCross isworkingwellas it isvisibleandwecanseehowmanyfallsand

pressureulcerswehave.ItishavingitvisiblethatmakesitmoreofapriorityIthink.Idon’t

like that it is a cross as that is off-putting for some people and there has been a bit of

backlash from some people about that. If it was a calendar or a different shape I think it

wouldbebetter.

WearesubmittingdatafortheMonthlyMappingwhichissohelpful.TheSafetyCrossisok

butitdoesn’tshowusovertime.Themappinggraphsgorightbackforayearormoresowe

canreallyseewhatishappeningandthehighsandlows.Havingthecomparisonswithother

homesisuseful.Itmakesyouseewhereyouaredoingwellandnotsowellanditmakesyoua

bitcompetitiveinagoodway.BeingabletolookatdataisgoodandIhavestartedtomake

graphsforotherthingstoo.Wecanseeitisusefulsowearetryingitforotherthings.

Weareenjoyingbeingpartofitandrecommendittootherpeople.Wewillcontinuetouse

thetoolsprovidedoncePROSPERis finishedbecause ithelpspeoplestayfocusedandkeeps

safetyapriority.Itdoeschangethewayyouthinkaboutthings.Wehavesomuchgoingon

hereanditiseasytoforgetaboutthings.PROSPERishelpingusgetbacktobasicsandthink

aboutwhatisimportant.Andwearedoingitasateam.”[Managerofamedium-sizedhome]

31

Figure2showsresultsfromabeforeandaftersurveycompletedbycarehomemanagerswhichwasdesigned to track changes betweenwhen care homes began PROSPER and a follow-up inMarch2016,betweeneight-and20-monthslater.Datafromthe51carehomesthatcompleteditsuggeststhattherewerechangesinself-reportedaspectsofsafetyculture.Inparticular,towardstheendofPROSPERcarehomesweremorelikelytosaythattheywereusingdatatoguideimprovement,thatsafety was a priority and they had more knowledge and understanding of safer care. Theseimprovementsheldupregardlessofthesizeofthehome,thepartofEssexortheownershiptype.Figure2:Proportionofcarehomesthatagreedwithvariousstatementsaboutsafetyculture

Note:Dataisbasedonmatchedcomparisonsof51carehomesspanningeight-to20-months.

51

51

51

82

82

84

94

94

94

94

100

10

21

33

51

33

67

59

51

59

10

33

0 10 20 30 40 50 60 70 80 90 100

WeregularlyusesmalltestsofchangetomakeimprovementsegPDSAcycles

Weregularlyworkwithrelativesandresidentstoplanhowtoimproveresidentsafety

Stafffeelvaluedforimprovingresidentsafety

Ourotherstaffhavegoodknowledgeandskillsforimprovingresidentsafety

Ourmanagersandstaffhavebetterknowledgeandskillstoimproveresidentsafetythanthey…

Allstaffatourcarehomethinkresidentsafetyisapriority

Staffatourcarehomeunderstandwhatasafeenvironmentis

Allthemanagersandstaffatthehomeactivelytrytoimprovesafetyforresidents

Ourmanagershavegoodknowledgeandskillsforimprovingresidentsafety

Overthepastsixmonthsourhomehasstartedthinkingmoreaboutresidentsafety

Weregularlyusedatatoplanhowtoimprovecare

%agreeatstartofPROSPER %agreetowardsendofPROSPER

32

2.2 SafetyprocessesDataforthissectionweresourcedfrom:

• reviewof523documents• 127hoursworthofhomevisitsandobservations• 203individualtelephoneinterviews• 12discussiongroupswithcarehometeams• 26meetings,discussiongroupsandinterviewswithPROSPERteammembers

About two-thirds of carehomes involved in PROSPERmade some changes to their processes in awaythatimprovedsafety.Themostcommonlyreportedchangeswere:

• improvingtheuseofdataaboutincidentstohelptrackchangesovertimeandthinkaboutthereasonsfortrends.Graphsshowingcarehomestheirmonthlyrateofincidents,includingupto20-month’sofdata,werewellreceived.Carehomesreportedusingtheirdataatteammeetingsorindiscussionswithrelativegroupstohelpidentifyareasforimprovement.Onehomeuseda‘SafetyWalkingStick’(similartotheSafetyCross)atthefrontofpeople’snotestoeasilyseethenumberanddateoffallsratherthanthesebeing‘hidden’incarerecords.Othershaveadaptedasimpletableprovidedbytheevaluationteamtocompileinformationaboutotherharms(outsidethescopeofPROSPER);

• implementing changes in small cycles and testing their impacts. In quality improvementtermsthisisknownasaPlan-Do-Study-Act(PDSA)cycle.Howevermostcarehomesdidnotusethisterminologyandanumberdidnotkeepformalrecordsoftheirchangesandtests.Theydid,however,use thebroadapproachof testingnew ideas.Examples includedusingdifferent coloured drinks coasters or cups as a reminder that people should be offeredbeverages frequently or providing families with a leaflet about where to buy non-slipslippers.Box3listsfurtherexamples.SomeoftheseideaswereexamplesprovidedaspartofPROSPERdiscussionsandtoolkits.Othersweredevelopedbythecarehomesthemselves;

• takingpartintrainingaboutinfectioncontrol,diabetes,medicinesmanagement,hydration,

fallspreventionandassociatedtopics;

• making safety issuesmore visible through theuseof ‘champions’ (namedanddesignatedindividuals),displayboardsanduseoftheSafetyCross(safetyincidentscalendar);and

• using resources provided by the PROSPER implementation team, such as smallmirrors tohelpcheckforpressureulcersonheels.

Appendix5describesthesechangesinmoredepth.Box4providesanexamplefromonehome.CarehomesthatdidnotmakechangestotheircareprocesseswereeithernewtotheprogrammeordidnotfeelthatPROSPERofferedanynewinformationorskillsforthem.

33

Box3:ExamplesofchangesreportedlymadebycarehomesasaresultofPROSPER

Sharingwithothercarehomes

• Sharingideaswithothercarehomesthroughchampions’studydays,communitiesofpracticeeventsandnewsletters

• Invitingothercarehomestoattendeventsrunatthehome,suchasnutritionandhydrationactivities,dementiadayandtrainingactivities

Gettingtheteaminvolved

• Encouragingawiderangeofstafftobeinvolved;throughaskingforideasatstaffmeetings,usingtheSafetyCrossandhavingPROSPERchampionstotrainotherstaff

• Havingasuggestionsboxtoactivelygetpeople’sideasforimprovement

Usingdataforimprovement

• UsingtheSafetyCrosstoraiseawarenessaboutthenumberoffalls,pressureulcersand/orurinarytractinfections

• UsingtheMonthlyMappinggraphstolookatlonger-termchangesandcomparingwiththeaverageofothercarehomes

• ShowingCQCrecordsofchangesmadeandtheimpactsofthisovertime

Promotingbettercare

• Helpingtochangepeople’sperceptionsofcarehomesbyshowingthatstaffareworkinghardtohelpresidents

• Displayingmaterialsonboards,e.g.theSafetyCross,graphs,newsletters,tipssheets,whichhelpedstaffandfamiliesseewhatishappeningandaskquestions

• Cascadingwhatislearnttoothercarehomesinthecompany

Learningnewthings

• Learningnewinformationfromtraining,suchasChampions’dayswherecarerslearnhowtoteachotherstaffmembersaboutwaystoimprovecare

Makinglittlechanges

• Offeringmoredrinkstoresidents,havingmoredrinksstationsinthelounge,usingcolouredcoastersorcupstoshowwhoneedsmorefrequentfluids,usingjellies,fruitplattersandmelonstoincreasehydration

• Usingmirrorstocheckforpressureulcers• Decoratingwalkingframestoindividualisethem,changingferrules(tostabilisetheendof

walkingsticks)andusingthefall’schecklist.

34

Box4:Exampleofchangesinsafetyprocessesatonehome“WearenowusingPROSPERmoreasatoolforusratherthanatoolforitsownsake.Itishelpingus

topreventthingsfromhappening.Wewereslowtogetgoingbutnow,thinkingback,wehavedone

alotofthings.Thisprojectisusefulforremindingpeopleofthings;itisallaboutbasicsbutisuseful

asweareseeingresults.Wearedoingsmallthingslikethebodymapforpressuresores,withevery

memberofstaff involved.Carersarecompletingabodymapeverydayforabout80%ofresidents.

PROSPERgaveusthebodymaptouseandaseriesofquickquestionsforcarerstothinkabout,anda

space forcarers to sign theirname. It isauser-friendly form.Gettingengagement fromall staff is

important.Itdoesn’ttakepeoplelongoutoftheirdaytodothesethingsandwenowhaveaneasy

wayofrecordingthings.

PROSPERhasbeenreallygoodfor tworeasons.Firstlywehaveseen improvementswhich isbetter

forresidentsandtheirsafety.Butsecondlyitismakingcarersthinkoutsideoftheboxandconsider

allthereasonsforthings.Likefallsisnotjustaboutmobility,theremaybeotherreasonspeoplefall.

Andwhatisthereasonforpressureulcers?Ithasmademanagersandstaffconsiderallofthethings

involvedinreducingrisk.Westartedtoanalysethefallstoseewhetheritistodowithcapacityand

weakness. We look at how often people fall, howmany people fall and when.We look at what

precautions are needed. Accident reports nowgo on the desk of themanager rather than in care

plans.Fallsaremorevisiblenowratherthanhiddenaway.

Thereare lotsofchangesasa resultofPROSPER.Everyone ismoreaware, like thedomestic team

automaticallyofferpeopleadrinkwhentheygointocleanrooms.Itmakespeoplefeelmoreapart

oftheteam.Wewerealwaysproactivewithpressureulcers,butnowthestaffaremoresharpabout

thisandhavebeentrainedabouthowtospotissues.Thedrivershelpedusfocusonwhattoupskill

staffabout.Wenowoffer10minutetrainingabouteachthing.Thecarergetsacopyandtheysignit

tosaytheyhavedoneit.Thismeansthemanagerislessanxiousaboutstaffskills.Wegotthisidea

from thePROSPER trainingwhere they talkedabout safetydrivers.Wedo the training inpeople’s

firstweekaspartofinductionnow,sowehave10minutetrainingoneachofthesafetydrivers.

WestartedusingtheSafetyCrossforpressureulcers,butnowwehaverolleditouttofallsandUTIs.

We’veputitacrossthewholehome.Itgivespeoplemotivationtokeepgoingandmakesthemproud

toseeeverythingisallgreen.TheMonthlyMappinggraphshelpusthinkabouthowtoimproveand

thatwasn’ttherebefore.Wehadthemetricsbeforebutnowwecanseewhatitmeansforus.

Thishashelpeduswithaperson-centredapproach.It isnotonlystatistics.Behindthestatisticsare

residents.Thisisaverybusyhome.Manytimeswearedoingthingsatspeed.Nowweknowwecan

doall thegraphsquicklyand then lookat them in-depthandanalysebecauseeverything is inone

place.Wearediscussing thingsasa teamand thinkinghow todo thingsbetter together.There is

goodteamworkingandweaskforadvicefromseniors.Thetoolshavehelpedustostopbehaving

likerobots,tostandbackandthinkaboutthings.Beforewelookedatthingsonanindividualbasis

but nowwe look at things ona bigger scale and focus on improvements throughout the home. It

pushesustothinkaboutnewideasforhelpingpeopleandwearediscussingthingswithfamiliesand

residents.”[Managerofalargehomeownedbyacorporation]

35

2.3 ResidentoutcomesThe outcome data and care home characteristics were collected by The Evidence Centre using amonthly data collection form.Datawas collected between a six- to 12-month period prior to thestart of the intervention and in March 2016. Care homes were asked to provide the number ofevents foreachoutcomeonamonthlybasis. Inadditiontodataonthetotalnumberof residentsand total number of female residents each month, the care home location and other items ofbackgroundinformation,asdescribedinTable1.4.Notallcarehomescollectedthequantitativedatarequestedbytheevaluationteam.94%(n=17)ofcarehomesfromcohortone,89%(n=16)ofcohorttwohomes,62%(n=13)ofcohortthreehomesand55% (n=18)of cohort fourhomesprovidedsomedataabout residentoutcomesbetween July2013(oneyearpriortotheintervention)andMarch2016.Thequalityofthedatacollectedvariedbetween the care homes. Some care homes only provided pre-intervention data and some onlyprovidedpost-interventiondata,andsomeprovidedboth.Missingdataamongstthe64carehomesforthesixoutcomemeasureswere6.8%forcohortone,9.8%forcohorttwoand12.6%forcohortthree.41.1%carehomes (7outof17) fromcohortone,50.0%(8outof16) fromcohort twoand23.1%(3outof13)fromthecohortthreehomeswereabletoprovidedataforsix-to12-monthsbeforeandaftertheintervention.Cohortfourdatawasnot includedintheanalysisbecausetherewas a cut-off point for data collection of 8 April 2016. There was therefore insufficient time forenough post-intervention data points for this cohort because these care homes began theprogrammebetweenNovember2015andFebruary2016.

Outcome data were plotted as time-series graphs with a one-month interval. The rates ofpre-intervention and post-intervention incidents were compared using chi-squared test, i.e. totalnumber of events over total number of residents observed between pre-intervention period andpostinterventionperiod.AllstatisticalanalysiswascarriedoutusingSASstatisticalsoftware(version9.3).The analysiswas challenging for a number of reasons: (a) it is difficult to attribute change to thePROSPER intervention, not just because of the study design but because the intervention waschanging during the study, (b) an increase in reporting of incidents might be expected as aconsequenceofparticipationintheprogrammeandthiswouldhideanybeneficialimpact,(c)homesdonothavealongtrackrecordofcollectingdata,particularlyforpressureulcersandUTIs.Formanyof the care homes their methods of collecting data were often manual (sometimes extractedretrospectively from resident records), complicated andunreliable, and (d) it is difficult to reducehospitalutilisationasthisdependsonawiderangeoffactors.

Wethereforeexperiencedchallengeswithboththequalityandquantityofdataavailablefromcarehomes despite considerable efforts by the evaluation and implementation teams. Particularproblems were experienced with collecting pre-intervention data. The evalution team undertookmeasurestogetthebestpossiblequalitydatabuthadtorespecttheemergentnatureofthestudyandthatPROSPERwasprimarilyanimprovementprojectratherthanaresearchstudy.

36

ResultsfromallavailabledataforthestudyIf all the available before and after intervention data from the care homes are aggregated thenPROSPERwas associated with statistically significant reductions in the rates of falls and pressureulcers between the pre-intervention and post-intervention periods (Table 2.1). There was noreductioninhospitaladmissionsduetofalls,andastatisticallysignificant increaseinratesofUTIs,anyhospitaladmissionsandA&Eattendancesaftertheintervention.

Table2.1.Numberofeventsandeventratesamongthe64carehomes

Noofevents Noofresidents Rateofevents pvalue

Falls

pre 3058 12884 23.7% <0.01

post 4714 22564 20.9%

Pressureulcers

pre 644 12367 5.2% <0.01

post 858 22157 3.9%

UTIs

pre 484 10934 4.4% <0.01

post 1073 20900 5.1%

Hospitaladmissions

pre 297 11935 2.5% <0.01

post 704 21731 3.2%

A&Eattendances

pre 312 11932 2.6% <0.01

post 729 21839 3.3%

Hospitaladmissionsduetoafall

pre 122 12363 1.0% 0.16

post 252 21875 1.2%

37

Figures2.1-2.6showchangesinratesofsafetyincidentsovertime

Figure2.1.RateoffallsinthePROSPERcarehomes

Figure2.1.1Incidentratesbycohortbeforeandafterintervention

Figure2.1.2Runchartofchangesinincidentsbycohortovertime

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

38

Figure2.1.3Runchartoftotalchangesinincidentsacrossallcohorts

Figure2.2.RateofpressureulcersinthePROSPERcarehomesFigure2.2.1Incidentratesbycohortbeforeandafterintervention

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

39



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

40

Figure2.3.RateofurinarytractinfectionsinthePROSPERcarehomesFigure2.3.1Incidentratesbycohortbeforeandafterintervention


July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

41


Figure2.4.RateofanyhospitaladmissionsinthePROSPERcarehomesFigure2.4.1Incidentratesbycohortbeforeandafterintervention

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

42



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

43

Figure2.5.RateofA&EattendancesinPROSPERcarehomes



July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

44


Figure2.6.RateofhospitaladmissionsduetoafallinPROSPERcarehomesFigure2.6.1Incidentratesbycohortbeforeandafterintervention

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

45



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

46

Resultsofsub-groupanalysesofthedataGiventhecomplexityofthe improvement initiative, intermsofthenatureoftheinterventionandthe turnoverof residentswithin the carehomes, theevaluation teamconsidered it reasonable toaggregatedata fromall thecarehomes, irrespectiveofwhether theyhadcollectedbothpre-andpost-interventiondataandwhethertheyhadfocusedtoagreaterorlesserextentonspecificsafetyareas. Nevertheless, and in addition to thismain analysis, we carried out a number of subgroupanalysestoexploretheuncertaintyinherentinthedata.First,welookedattheoutcomesderivedfromcarehomesthatfocusedonspecificareasofsafetyratherthanattemptingtoimproveallthreeareas.Sixcarehomesfromcohortone;tencarehomesfrom cohort two and four care homes from cohort three focused on falls. Four care homes fromcohortone;eightcarehomesfromcohorttwoandfourcarehomesfromcohortthreefocusedonurinary tract infections. Four care homes (three in cohort one and 1 in cohort two) focused onpressureulcers.Theeventratesoffalls,pressureulcersandurinarytractinfectionsareshownin2.7-2.9. Similar to the results derived from all of the data, the intervention was associated withstatistically significant reductions in the rates of falls [21.7% 1277/5884) vs. 19.9% (2312/11608),p<0.01]andpressureulcers[7.1%(64/897)vs.1.6%(49/3015),p<0.01]betweenthepre-interventionandpost-interventionperiods.However,anincreasedrateofurinarytractinfectionswasobserved[5.0%(254/5091)prevs.6.7%(470/7054)post,p<0.01].Second,acasecouldbemadetoonlyanalysedata fromcarehomeswhichhadprovideddata forboththepre-andpost-interventionperiods.Sevencarehomesfromcohortone,eightcarehomesfromcohorttwo,andthreecarehomesfromcohortfourhadbothpre-andpost-dataforsixtotwo-monthseithersideoftheinterventionstartmonth.Re-analysisofthedataforthissub-groupofcarehomessuggestedthattherewasnostatisticallysignificantdecrease in incidentratesasaresultofthe intervention for any of the outcomes, i.e. the intervention effect seenwhen all the datawasanalysed disappeared – indeed there was a statistically significant increase in the rate of falls,urinary tract infections, A&E attendances, any hospital admissions and hospital admissions as aresultoffalls.TheresultsfortheseanalysesareshowninAppendix6.Third,wewereuncertainabout thevalidityofdatacollectedby threeof thecarehomesbecausetheyappearedtoreportnoincidentswhenwesuspecttheyweresimplynotreportingincidents.Wetherefore re-analysed the data having excluded these three care homes from the analysis. Nodifferencewasfoundbetweentheresultsofthissub-groupwhicharepresentedinFigures2.1–2.6usingallofthedata.

We found no differences in impacts on resident outcomes based on geographic area, home size,whethercarehomeshadoneassignedGP,levelofengagementwithPROSPERandcohortnumber.Thedata seem to suggest somekindof ‘cohorteffect’, forexample cohort three seem tomakeadisproportionatecontributiontoreducingfallsandcohortoneseemtohaveastrong influenceonthe observed changes in pressure ulcers. The reasons for these effects are not clear from thequantitativeorqualitativedatathatwecollected.

Finally,residentoutcomescollectedusingtheNHSSafetyThermometerwereavailableforsomecarehomes incohortsoneandtwo(n=18),beforethecarehomesdecidedtomakeuseoftheSafetyCross. In linewithQuality Improvementmethodology, these datawere derived froma sample of

47

residents over a 72-hour period, rather than from all residents. Figure 2.10 provided by Haeloshowed that there were no significant changes for the falls, pressure ulcers and urinary tractinfection. Overall, the safety incident rates were lower than the rates reported by care homes,confirming theevaluation team’sview that the carehomeswere less likely touse theNHSSafetyThermometer.

48

Figure2.7.RateoffallsincarehomesthatfocusedonreducingfallsFigure2.7.1Incidentratesbycohortbeforeandafterintervention


July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

49


Figure2.8.Rateofpressureulcersincarehomesthatfocusedonreducingpressureulcers


July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

50



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

51

Figure2.9.Rateofurinarytractinfectionsincarehomesthatfocusedonreducingtheseinfections



July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

52


July14 Jan15 Jul15

Incide

ntra

tes

53

Figure2.10:Rateoffalls,pressureulcersandurinarytractinfectionsintheNHSSafetyThermometer

4.6%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

31/03/201414/04/201428/04/201412/05/201426/05/201409/06/201423/06/201407/07/201421/07/201404/08/201418/08/201401/09/201415/09/201429/09/201413/10/201427/10/201410/11/201424/11/201408/12/201422/12/201405/01/201519/01/201502/02/201516/02/201502/03/201516/03/201530/03/201513/04/201527/04/2015

Prop

ortio

nofpatien

ts

Proportionofpatientswithafall

Falls Median

4.6%

0.0%2.0%4.0%6.0%8.0%10.0%12.0%14.0%16.0%18.0%20.0%

31/03/201414/04/201428/04/201412/05/201426/05/201409/06/201423/06/201407/07/201421/07/201404/08/201418/08/201401/09/201415/09/201429/09/201413/10/201427/10/201410/11/201424/11/201408/12/201422/12/201405/01/201519/01/201502/02/201516/02/201502/03/201516/03/201530/03/201513/04/201527/04/2015

Prop

ortio

nofpatien

ts

Proportionofpatientswithanypressureulcer

PU Median

3.2%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

31/03/201414/04/201428/04/201412/05/201426/05/201409/06/201423/06/201407/07/201421/07/201404/08/201418/08/201401/09/201415/09/201429/09/201413/10/201427/10/201410/11/201424/11/201408/12/201422/12/201405/01/201519/01/201502/02/201516/02/201502/03/201516/03/201530/03/201513/04/201527/04/2015

Prop

ortio

nofpatien

ts

ProportionofpatientswithanyUTI

UTI Median

54

2.4 CostsIn the original proposal the evaluation team said that itwould carry out a cost analysis to assesswhethersavingswererealisedasaresultofPROSPER.WecollateddatarelatingtotheoveralldirectandindirectcostsofrunningPROSPERandthestandardisedunitcostsforeachsafetyoutcome(seeAppendix7).Wethencarriedoutapreliminarycostanalysisbasedontheresidentoutcomeanalysispresentedintheprevioussection,usingalldatafromallcarehomes,whereavailable.Wereviewedtheliteratureandusedstandardnationaldatatoestimatethecostofindividualharms.

The analysis involved the same risks and limitations as that for resident outcomes, togetherwithadditional complications around accurate costing of falls, pressure ulcers and other harms. Carehomeswill treat theseharms indifferentways,withdifferentassociatedcosts to thesystem.Forthisreason,theanalysisshouldbetreatedwithextremecaution,andfurtherworkisrecommendedtodevelopamorerobustpicture.Thekeyfindings,presentedinTable2.2were:

• The overall cost of the PROSPER project was £282,000. This includes a significantcomponentofsetupanddevelopmentoftheintervention,thatwouldnotberequiredforfuturesimilarprojects

• The cost savings across all PROSPER cohorts associated with the reduction in falls andpressureulcerswasintherange£86,000-£143,000

• These cost savings were offset by the costs of the increased incidence of urinary tract

infections,A&Eattendances,andhospitaladmissions,whichwereintherangeof£121,000-£465,000. By far the biggest component of thesewere the costs of increased admissions(£100,000-£442,000)andthismaybeexplainedbyageneralrise inadmissionsfromcarehomesoverthetimeperiodofPROSPER,orthroughmoreaccuraterecordingofevents.

Table2.2Costsandsavings

PeriodTotal

residentsNoofevents

Eventrate

Expectednoof

eventsNodiff

Unitcosts(range)

Costsavings-

lowerend

Costsavings-

upperend

Falls(community) pre 12884 3058 0.237

post 22564 4714 0.209 5356 642 £115-£197 £73,830 £126,474

Ulcers(community) pre 12367 644 0.052

post 22157 858 0.039 1154 296 £43-£57 £12,728 £16,872

UTIs(community) pre 10934 484 0.044

post 20900 1073 0.051 925 -148 £27.8-£29.0 -£4,114 -£4,292

Admissions pre 11935 297 0.025

post 21731 704 0.032 541 -163 £611-£3403 -£99,593 -£442,708

A&Es pre 11932 312 0.026

post 21839 729 0.033 571 -158 £114 -£18,012 -£18,012

Totalsavings -£35,161 -£321,666

(PROSPERcost £282,597)

55

2.5 BeneficiariesThissectionsummarisessomeof themainbeneficiaries fromPROSPER.Data for thissectionweresourcedfrom:

• 127hoursofhomevisitsandobservations,includingdiscussionswithfamilymembers• 203individualinterviewswithcarehomestaff• 12discussiongroupswithcarehometeams• 26meetings,discussiongroupsandinterviewswithPROSPERteam• 23interviewswithotherstakeholders,includingGPs,NHScommissionersandprovidersand

membersoftheCouncil

ResidentsandrelativesAlthoughnotallcarehomeswereabletodemonstratechanges inresidentoutcomes(seeSection2.3),someindividualresidentswithincarehomesexperiencedimprovementsinthequalityoftheircarewhichmayhavehelpedthemavoidafall,pressureulcerorurinarytractinfection.ImpactonresidentoutcomescanbefoundinAppendix8.

Relatives also potentially benefitted in some care homes by being more involved in discussionsabout improvementsandaccess to information;perhapsgiving themmoreconfidence in thecareprovided. Ad-hoc informal interviewswere undertakenwith 13 relatives during site visits by theevaluationteam.One-quarteroftherelativesmentionedseeingdisplaysrelatingtoPROSPER.Onepersonmentionedtakinginformationprovidedbythehomeintoaccountwhenbuyingtheirmothernew slippers. One personmentioned feeling that it wasmore acceptable for them to pour theirmotheradrink,whereaspreviouslytheywouldhavereliedonstafftodothis.Whilstthesearesmallexamples,theypointtothepotentialoftheinterventionsthatcarehomesimplementedtohaveanimpactonthefamilymembersofresidentsaswellasresidentsthemselves.

CarehomestaffCarehomestaffreportedthattheybenefittedfromthetrainingorganisedbyPROSPER,particularlytrainingaboutsubstantive issues,suchasdementia, fallsprevention,pressureulcers,diabetesandmedicines.

A number of home managers and carers were positive about the support provided by theimplementationteam,saying itmadethemmoreconfident inapplyingnewapproachesandmorecreativeinthinkingaboutdifferentstrategies.

Staff in care homes that experienced reductions in incident rates reported feeling proud of theirachievements,especiallyiftheywerelowerthantheoverallPROSPERaverage.

More junior members of staff or those that may not traditionally be involved in improvementinitiatives,suchasdomesticstafformaintenancepersonnel,reportedfeelingmorepartoftheteamandrecognisedtheircontributiontoresidentsafetyandwellbeing.

56

CarehomesasorganisationsCarehomesareprivateenterprises.PROSPERhelpedthesebusinessesshare ideaswithothercarehomes.Thisissomethingtheymightnototherwisehavedoneinacompetitivesector.

Somecarehomes reported thatdisplayinggraphsof theirmonthlydatahelped them impress theCQCbecause itwasawayofdemonstrating that theywereactivelymonitoring for improvement.TheCQCdemonstratedaninterest inPROSPERasanexampleofarigorousimprovementinitiativebymakingspecificreferencetotheprogrammeintheirinspectionreports.

Care homes also benefitted organisationally because PROSPER supplied some tangible resources(suchasheelmirrorsand ferrules), allowingcarehomes togain theseconsumables for freewhiletheytestediftheyworkedintheircontext.

PROSPERalso arranged free training for staff thatmaynototherwisehavebeenavailable to carehomes.

WorkingwithpartnerorganisationsThePROSPER implementation team involvedawide rangeofpartners, particularly CCGsandNHScommunity services. Stakeholders from someof these organisations reported gainingmore readyaccesstocarehomes:

“HavingPROSPERgavemeawayintohomesthatIpreviouslyfounddifficulttointeractwith.

Theyseemedtoseememoreasahelperthanperhapsahindranceastheyhadinthepast.

AlthoughIthinkthereisstillsomesuspicion,themanagerscouldseemeinadifferentlight

asIwasrunningtrainingfortheirbenefit.Homesoftendofeellookeddownonorblamedfor

thingsbytheNHSsowe’renotgoingtochangeattitudesovernight,butitdidmeanIcould

interactinadifferentway.It’sastart.”

[Nursefromcommunityteam]

TherewasasenseamongstmanagersandfrontlinehealthteamsthatPROSPERhadhelpedtoforge linksbetweenhealthand social care,even if thiswasnotevidentonaday-to-daybasiswithinthecarehomesthemselves.

Itwassuggestedthat infuturePROSPERcouldbejointlybrandedasa localauthorityandNHSinitiative.OneNHSorganisationwasconsideringfundingPROSPER intheirarea.FrontlineNHSteamssuggestedthattheycouldplayamoreactiveroleindeliveringtrainingandfollowing-upon improvement progress with a jointly branded initiative. For instance, a falls team orcommunitynursesmaybeabletomonitortheextenttowhichcarehomesimplementchangesfollowingtraining,providingfurtheraccreditationforthosewhoachievecertainmilestones.

NHSstakeholdershadanumberofideasaboutwaystheycouldworkmorecollaborativelyandaddfurthervalueiftheinitiativewasrunjointly.HowevertherewasvariationacrossEssex,duetothenumberofdifferentcommissionersandNHSproviderorganisationsinplace.

EssexCountyCouncilEssexCountyCouncilappearstohavebenefittedfromPROSPERreputationallybybuildingstrongerlinkswithcarehomesandNHSorganisations.TheimplementationteamhadadifficultrolebecausecarehomestraditionallyassociatedtheCouncilwithcomplianceandsafeguardingratherthanwithimprovement.Bysupportingcarehomesandprovidingresources,PROSPERreportedlyhelpedsome

57

carehomes tohaveamorepositive viewof theCouncil. Thiswas, however, notuniversal.About10% of care homes reported feeling that implementation team used information gained duringPROSPER indiscussionsaboutsafeguardingorperformancemanagementandto ‘reporton’ them.These carehomes felt strongly thatPROSPER teammembershadnotdrawna linebetween theirimprovement facilitation role and their safeguarding / assurance roles. Specific examples wereprovided by care homes to illustrate their views.Whilst these are the perceptions of individualswithinthecarehomes,andimplementationteammembersmayhavedisagreed,theywerestronglyheldbytheminorityofcarehomesthatexpressedthem.Atthesametime,theCouncilteamwereclearthattheyhadadutyofcarethatcouldnotbeignoredbecausecarehomeswereinvolvedinanimprovementproject.

Itmightbeexpectedthatcarehomesthathadtakenpart inother improvement initiatives runbythe Council, such asMyHomeLife, would be more open to Council input, having built a positiverelationship previously.Whilst these care homes often said they heard about PROSPER frompastactivities and were more likely to want to engage, it is not clear whether this is because theirmanagers had a stronger commitment to improvement or whether their experience of pastinitiatives had made a difference. Rates of withdrawals were high in the initial cohorts amongstthosewhohadtakenpartinpreviousprogrammes.ThereforehavingarelationshipwiththeCouncilteamdidnotappeartobepredictiveofsuccessintheprogrammeorlongevity.

ThePROSPERapproachofworkingcollaborativelywithcarehomeshadthepotentialtohelpchangeCouncil culture over time. The implementation team reported that the approach adopted byPROSPERhadbeen rolledout intootherdivisionsof theCouncil. Some carehomes reported thatCouncilteammemberswereapproachingthemdifferently,recognisingtheworktheydidandtakingamore supportive, rather than regulatory, approach. Care homes stated that culture changewasneededboth in thecarehomesandacrosshealthandsocialcaresectors,andPROSPERmayhavehelpedthisprocess.

58

3. Helpfulandhinderingfactors

3.1 FeedbackaboutPROSPERcomponentsThissectionexamineswhathelpedandhinderedPROSPERimplementation.Dataforthissectionwassourcedfrom:

• reviewof523documents• 127hoursworthofvisitsandobservations• 203individualinterviewswithcarehomestaff,includingfromcarehomesthatwithdrew• 12discussiongroupswithcarehometeams• 26meetings,discussiongroupsandinterviewswithPROSPERteam

Drawingtogetherfeedbackfromallofthesesources,thethingsthatappeartohaveworkedwelltosupportimprovementsinclude:

• enoughcapacityintheimplementationteamtoprovideregularproactivesupport;

• personable and passionate improvement facilitators who understood and respected thecarehomecontext;

• maintaining regular contact andgroupmeetings for six- toeight-months, recognising thatchange takes time and that care homes have many competing priorities;

• providingsubstantiveideasandresourcestocarehomes;

• simple introductory training about quality improvement using examples and languagerelevanttocarehomes;

• simpletoolsthathelpcarehomestousedataforimprovement;

• providing ways for care homes to compare themselves with other care homes(anonymously)toprovidereassuranceandmotivation;and

• providingopportunitiesforcarehomestoshareideasandlearnfromeachother.

MoredetailsareprovidedinAppendix9.Table3.1summarisestheperceivedvalueofthevariouscomponentsofPROSPERbasedonfeedbackfromcarehomemanagersandstaff.ThecomponentsofPROSPERthatpeoplesaidtheyvaluedmostwere,inorderofpriority:

1. training about topics such as infection control, falls prevention, diabetes, dementia andmedication,especiallywhentrainingisofferedinthehome;

2. theSafetyCross;3. monthlymappinggraphstrackingprogressovertimeandcomparingwithothers;4. champions’daysandhavingchampionstosupportothersinthehome;

59

5. visitsfromthePROSPERteamtothehome;and6. beingencouragedtothinkinadifferentwayaboutpreventionandsharingideaswithother

carehomesthroughvisits,champions’studydaysandnewsletters.

60

Table3.1:Carehomes’perceivedvalueofvariouscomponentsofPROSPERKey:green-highvalue,yellow-mediumvalue,red-lowvalue

Component Type Perceivedvalue Comments

TrainingandvisitsQualityimprovementtraining

Originalplan Revisedtrainingdeemeduseful,butconceptsnotcoveredinenoughpracticaldepth.Manyhomestookawaythephilosophyoftestingnewideas,butdidnotuseformalPDSAcycles.

Othertraining Addition Trainingaboutmedicines,diabetes,falls,pressureulcersetc.deemeduseful,butconcernsaboutlackofparityassomehomeswerethoughttohavegreateraccessthanothers.

Championsstudydays

Addition Wellreceivedduetoinvolvingcarers,notjustseniorstaff.Practicalfocusvalued.

Communityofpractice

Originalplan Morecoproductionbyhomesrequested.Limiteduptakeinsomeareas.

On-goingsupportvisits

Originalplan Ongoingvisitsuseful,butthoughttobevariableandinfrequentbyearlycohorts.

ToolsNHSSafetyThermometer

Originalplan Notuserfriendlyandconcernsaboutvalidity.Homesdidnotfeelthattheoutputgainedwasagoodreturnonthetimeinvested.

SafetyCross Addition Visiblemethoddeemedusefulforrecordingandgettingawiderrangeofstaffinvolved.Concernsaboutthe‘cross’shapeandrequesttouseSafetyStickinstead.

MonthlyMapping

Addition Graphsdeemedusefulforseeingtrendsovertimeandcomparingtootherhomes.Designislimitedasnotbuiltforimprovement.

MaPSaF/Cultureiskey

Originalplan Wordingandformatnotuserfriendly.HomessawpotentialinassessingculturebutwantedPROSPERtofacilitateshortsessionswithstaff.

OtherresourcesToolkitofinformation

Originalplan Worksheetsdeemeduseful.Requesttorevisewordingtobemoreuser-friendly.

KnowledgeHub

Originalplan Limiteduptakeandlimitedpostingofsuccessstoriesornewtoolsbytheimplementationteam.Facebookpageplannedinstead.

Newsletter Originalplan Homesvaluedseeingsuccessstoriesfromothersbutwouldlikemoredetailtoallowthemtoadaptfortheirhomes.

Twitteraccount

Addition NohomesreportedusingtheTwitteraccounttosupportimprovement.

61

3.2 ImplementationsuccessfactorsDataforthissectionwassourcedfrom:

• reviewof523documents• 127hoursworthofvisitsandobservations• 203individualinterviewswithcarehomestaff• 12discussiongroupswithcarehometeams• 26meetings,discussiongroupsandinterviewswithPROSPERteam• 23interviewswithotherstakeholders,includingGPs,NHScommissionersandprovidersand

membersoftheCouncilIn addition to the specific interventions tested as part of PROSPER, other implementation factorsmayhaveinfluencedthereportedsuccessoftheinitiative.Theseincludethefocusonempoweringcarehomes,theknowledgeandsupportivenessoffacilitators,capacityintheimplementationteam,takingastaggeredapproachto implementationtoallowtheprogrammetodevelopovertimeandbuildingpartnershipsbetweenhealthandsocialcare.

FocusonpreventionandempowermentQualitative feedback suggested that a key reason PROSPER was so valued is that it focused onempowering care home staff to take control. The programme was not about implementingstandardisedtoolsorapproachesina‘topdown’manner,butratheraboutengagingcarehomesinthinking creatively and proactively about prevention. This approach was well received by carehomes.

“Ihaveabsolutelyloveditpersonally.Ithinkitisabrilliantinitiativeforhealthpromotionas

opposedtodealingwithsomethingwhenithashappened.Itissoproactive.Insocialcarewe

arealwaysfirefighting.Thisismoreproactive.Wealwaysseemtofocusoncomplaintsand

whatisgoingwronginsteadofbeingpositive.PROSPERisaboutfocusingonthepositivewhich

engagesstaffmore.”

[Managerinmedium-sizedhome,runbylargecorporation]

SupportivefacilitatorsThemotivationandenthusiasmofthefacilitationteamwascrucialforkeepingcarehomesengaged.Facilitatorswerereportedlyapproachable,responsiveandpassionate.

“[Facilitator] is amazing, so supportive.He is sopassionateabout it all. Thiswouldnothave

taken off so well if it was not for him. His enthusiasm is infectious. He attended the first

meeting at the home and responds to all emails rapidly. He visits regularly. I feel really

supported.”

[Managerfrommedium-sizedhome,runbycorporation]

The facilitators were not experts in improvement methodologies or in ways to reduce safetyincidents.TheylearntaspartofPROSPER.Thelackofexperiencedidappeartoinfluencetheextentto which improvement methodologies were understood and implemented, but the interpersonalskillsofthefacilitatorshelpedtomotivatecarehomes,regardlessofanylackofcontentexpertise.

62

Havingmembersof the implementation teamvisit regularlywasuseful.Carehomes that receivedregular visits reported more changes in culture and processes than those that were visitedinfrequently.CarehomesvisitedmorefrequentlywerealsomorefavourableaboutPROSPERoverall.

“[Facilitator]gavegoodtrainingandspenttimeonthefloor.Thisisimportanttomotivatestaff

becauseyouneedtogetpeopleonboard.Thestaffgavelotsofideasaftertheyheardafew

from[facilitator].Itisthecomingintothehomeandtalkingtostaffandworkingjointlywhat

makesadifference.Givingthemdirecttheinforatherthangoingthroughthemanager.

Acknowledgingtheideasofstaffisimportant.”

[Carerfrommedium-sizedhome,runbysmallorganisation]

“Thesuccessisduetoamixtureofboththeimprovementmethodsandhavingtheteamcome

intogetpeopleonboardfromthestaff,notjustthemanager.Thisishardtoachievein

practice.Thishomehashadlotsofstaffingissuesandsafeguardingissues.Wehavehadlotsof

inputfromcarebundles,skinteam,NHSsupport–soPROSPERisonepartofit.Comingand

visitingallthetimemakesadifference.”

[Managerfrommedium-sizedhome,runbycharity]

ImprovementteamcapacityIntheinitialstagesofPROSPER,theCouncilhadmanyotherprioritiesandtherewasreportedlynotenoughstaffcapacitytosupportcarehomesasplanned.Thismeansthatthefirstcohortdideithernotreceivealltheplannedsupportorassistancewasdelayed.

Therearearangeofreasonsforthis.Firstly,theprogrammewasnewanddevelopingsoittooktheimplementation team time to understandwhatwouldwork and the challenges facedby the carehomes. Secondly, the facilitation team themselves were not familiar with quality improvementmethods and tools, so they were learning about the content and may have found it difficult tosupport care homes. Thirdly, the facilitation team had other priorities, such as dealing withsafeguardingissuesandanewcontractingsystemforcarehomes,sostaffdidnotalwayshavealotoftimetodevotetoPROSPER.

Carehomesfromcohorttwoonwardsbenefitedfrommorecapacityintheteamandthattheyhadgainedmoreexperienceinapplyingimprovementmethodsinthecarehomesector.Howevertherewere still capacity issues, with facilitators unable to visit all care homes regularly and having toprioritisetheirresponsibilitiestotheindividualcarehomesandthewiderprogramme.Someoftheplanned implementation activities were given lower priority in favour of wider promotion anddisseminationoftheinitiative.

StaggeredimplementationapproachRather than recruiting one large cohort of care homes at the outset, the implementation teamdecidedtorollouttheprogrammeinsmallercohortsspacedapproximatelysix-monthsapart.Thisallowedtheteamtofocusits limitedresourceandtorefinethetoolsandtechniquesusedsotheyweremoreappropriateandfeasibleforcarehomes.

Somecarehomesnoted that theprogrammescheduleworkedwell,with training followedbyon-goingsupportandopportunitiestoshareideaswitheachother.

63

“Ithinkitfitstogethernicely.Firstyougototraining,thenyoupulloutyournumbers,thenyou

starttodochanges,thenseeifitworks.Forusitworks.Itallflowstogether.Butit’sonlynowI

knowthat,lookingbackonitall.Atthestarttheprocesswasn’tclear.Iftheycouldmaybedoa

diagramorexplainallthestepssoyouknowwhatyouareaimingtowards,thatmightmakeall

thethingsfittogetherbetterratherthanbeingamishmash.”

[managerfrommediumhome,runbyonefamily/group]

Otherssuggestedthattheprogrammewasquite‘bitty’andalwaysbeingaddedto,ratherthanbeingcohesive. There was a call for more clarity about what PROSPER is, what interventions will beinvolvedandwhattimehomeswouldbeaskedtocommit.

“IwouldliketosaythatPROSPERisagoodidea.Anythingthatisfreeandhelpsushomesis

goodinnit?Weputuptheposterandweusewhattheygiveus.Theyaregoingtogiveusmore

freethings.Idon’twanttosoundlikeacriticism,butitseemsitgoesfromonethingtothenext

butisnotacohesive,isthattherightword?Imeanawholecombinedthing.Itisverygoodand

welikebeingpartofitbutsomehowitisnotreallythatorganised…It’snotlikeitallgels

togethernicely.”[Administrativestaffmemberfromasmallhome,runbyasmallgroup]

“Theycouldhavesomesortofmanualorguidebooktellingyouwhatyouwillbedoingandhow

muchtimeitwillinvolve.Likeahelpmanualforacomputer.Thenyoucouldlookoverthe

wholethingandgetanoverview.Youcouldseewhatarecomingnextandthedatesof

training.Youcouldknowifsomeonemightvisit.That’swhatismissingIthink.Thereareall

thesethingsinittojuggle,andtheycouldmakeasimpleoverviewtotellyouwhattoexpectso

itisnottoooverwhelming.”[Managerfromasmallhome]

EngagementwithpartnersThe implementation team reported that the networking and liaison that they did with partnerorganisations was crucial to success. This includes working with the five CCGs in Essex andcommunityhealthtrusts.Thenetworkingledtothedevelopmentordistributionofsimpleresourcesand checklists (such as a falls, pressure ulcers and continence checklists). NHS organisations alsodeliveredtrainingandsignpostedhomestoadditionalservices,suchasfallspreventionteams,shopswhereshoefittersvisithomestomeasureandprovideappropriatefootwearandfreeprovisionofferrulesforwalkingframes.Asoutlinedinthe‘beneficiaries’section,NHSstakeholderswerekeentobemore involved in the programme in future and felt they hadmore to offer regarding traininghomes, supporting them to implement changes and tracking the quality and extent ofimplementationovertime.

BuildingonothersuccessesAs outlined earlier, Essex was an early implementer of MyHomeLife, a national organisationaldevelopmentandleadershipinitiative.HomesthattookpartgenerallyhadmorerapidsuccesswithPROSPER. This may be because homes that volunteered to take part in PROSPER were moreinnovative and willing to accept a challenge or that PROSPER was building on a history ofdevelopmentprogrammeswithinEssex.

Sincerecruitmentwaslargelybasedonexistinglinkswithcarehomes,itcouldbearguedthatthosethatwereresponsivetochangeweretargetedfirst.ThevalueofPROSPERmaybedifferentoritmay

64

have been slower to see change if a random sample of care homes had been selected. In latercohorts where therewasmore of amix of care homes, withdrawal rates remained at about thesamelevelaswiththeinitialcohorts,buttherewerepotentiallyalowerproportionofcarehomesactively engaging. For instance, a lower proportion of those from cohort four used theMonthlyMappingtool.

EvaluationasacomponentoftheinterventionFrom the outset, the evaluation of PROSPER was designed to be participatory and formative innature.Atthesametimetheevaluationteamneededtomaintainsomeobjectivity.

There was a positive interaction between the care homes, the improvement team and theevaluationteam,andmuchlearningabouthowtoundertakeparticipatoryevaluationeffectively.

Howevertherewerealsochallenges,including:

• differencesinapproachestodatarigour• differencesinhowtointerpretdataandcommunicateittostakeholders• managingthe‘politics’ofCouncilreputationandfuturecommitmenttotheprogramme.

TheevaluationteamsometimesfeltthatpreliminaryresultswereinterpretedprematurelyoroverlypositivelytoenthuseandkeepthecarehomesandtheCouncilengaged,whereasmorechallengingresultshadthepotential tobedismissedorvieweddefensively.Theevaluationteamalsobroughtissuestotheattentionoftheimplementationteamwhichtheymaynothavebeenopentohearingor which they had already acted upon. The improvement team sometimes questioned a lack ofpragmatismwithin theevaluation teamandona fewoccasions, criticised them for implementingnewideaswithoutfirstdiscussingthemwiththeimplementationteam.

Despitethesechallenges,havinganembeddedevaluationteamwaspositiveforPROSPERbecauseit:

• providedadditionalexpertiseinimprovementmethodologies;• ensured feedback and data was collected independently and could be used to refine the

initiativethroughout;• meantthatabespokedatacollectiontoolwasdevelopedforcarehomes;and• provided care homeswith anopportunity to share their views in a ‘safe’manner and get

expertsupportwithdataanalysis.

65

3.3 ImplementationchallengesAll complexprogrammesencounter challengesandPROSPERwasnoexception. Inaddition to thechallenges with specific intervention components outlined in Appendix 9, three other challengeswereidentifiedrelatingtotheturnoverofcarehomemanagers,accountingforthecomplexcontextinwhichcarehomesoperateandattributingsuccessestoPROSPER.

Dataforthissectionweresourcedfrom:

• 203individualinterviewswithcarehomestaff• 26meetings,discussiongroupsandinterviewswithPROSPERteam• 23interviewswithotherstakeholders,includingGPs,NHScommissionersandprovidersand

membersoftheCouncil• 10interviewswithcarehomesnottakingpartinPROSPER

CarehomemanagerturnoverThere isahighrateof turnoveramongstcarehomemanagersandstaff inEssexandelsewhere inEngland.24,25Itisestimatedthatupto40%ofcarehomestaffnationallymayleavetheirpostswithinthefirstyear.26ThiswasachallengeforPROSPERbecausemanagersandstaffmaysign-uptotakepartbutthenmoveon,orparticipateintrainingandthengetanotherposition.Thismeansthatcarehomes had to ‘start from scratch’ or the involvement in the programmewas delayedwhilst newmanagersgotestablished.Uptoone-thirdofcarehomesinPROSPERexperiencedachangeinseniormanagementorownershipduringthecourseoftheinitiative.

Maintainingthestabilityofcohortsaslearningsetswasalsodifficultbecauseasmallnumberofcarehomeswantedtojoinpartwaythroughacohort(ormayhavebeenrecommendedtojoinbyCQCorCouncilteams).Thismadelogisticsofboth implementationandevaluationdifficulttomanageandmeant that separate training was needed for individual care homes. This resulted in some carehomesfeelingthatotherswerebeinggivenspecialtreatmentwhenthey‘joinedlate’.

ContextCare homes are operating in a difficult economic and social context, with high staff turnover,negative publicity and many competing demands. Inspection requirements, changes in localauthoritycontracts,turnoverofseniormanagement,mergersandownershipchangesallimpactontheextenttowhichcarehomeshadthecapacitytoconcentrateonPROSPER.

When PROSPER was first implemented, care homes were coping with the introduction of a newcommissioning contract by the local authority, changes in the Deprivation of Liberty SafeguardspolicyandnewregulatoryinspectionsbytheCQC.Thecompetitivebusinessethosofthecarehomescouldalsosometimesactasadeterrenttoparticipation.

AttributingchangestoPROSPERAnotherkeychallengewasinattributinganychangesinoutcomestoPROSPERbecausetherewereanumberofparallelinitiativesthatcouldhaveimpactedontheobservedchanges.Somecarehomes

24 www.skillsforcare.org.uk/document-library/finding-and-keeping-workers/recruitment-and-retention-

strategy/workforceretentionfullresearchreport.pdf25 www.nmds-sc-online.org.uk/Get.aspx?id=28594526 www.nursingtimes.net/concerns-over-high-staff-turnover-in-care-homes/1806216.fullarticle

66

were getting regular visits from local authority or NHS teams separate from, and in addition to,PROSPER.

The lack of a comparison group in the evaluation made attribution difficult. It would have beenpossible to use a waitlist control design, but during the development and set-up phase this wasdeemed difficult because of a lack of alignment between the preferences of the evaluation andimplementationteams.Asacohortapproachwasused,waitlistcontrolswouldhavebeenpossiblebut was practically challenging and therefore rejected as an option. Better planning by theimplementation and evaluation teams early on (in the development phase) may have helped toconsiderthedifferentoptionsinmoredetail.

AreasforprogrammedevelopmentDuringinterviewsandhomevisits,allcarehomessuggestedwaysthatPROSPERcouldbeimproved.Suggestions, many of which were partially implemented and some of which are inevitablycontradictory,included:

Levelsofsupport

• targetingcarehomeswithspecificneedstogainmoreintensivesupport;• providingsupporttocarehomesforalongerperiodoftime,e.g.visitsorcalls;• providing equal levels of support for care homes, rather than some care homes being

perceivedtohavemorecontactandhelpthanothers;• being careful not tomake it seemas thoughPROSPER is asking care homes to givemore

thantheyreceiveinreturn,e.g.givingideasanddatatoPROSPER;• adaptingsuggestionsandinterventionstocarehomesratherthan‘imposing’interventions,

e.g.theskinbundle.Thisincludestryingtointegrateideaswiththeworkhomesarealreadydoing;and

• recognisingthatitcanbetimeconsumingforcarehomestotakepartinPROSPERintermsofthepaperworkinvolved,especiallyforsmallcarehomes.

Gainingbuy-in

• being clearer about what care homes will have to do and will gain from taking part inPROSPER;

• makingitmoreattractiveforcarehomestotakepart,suchassayingthat itcouldbeusedtowardsCPDformanagersortosupportCQCinspections;

• recruitingmorecarehomestotakepart;• havingaPROSPERteammembervisittooutlinewhattheprogrammeisinastaffmeetingto

helpstaffunderstand,ratherthanleavingthisforthemanagertodo;and• gettingtheregionalmanagersfromlargegroupsofcarehomesinvolvedinthetraining;

Promotionandjointworking

• raising awareness of the work care homes are doing through PROSPER amongst a widerrange of organisations locally, including employment forums, CQC, other Councildepartments,CCGsandGPs;

• championingissuesonbehalfofcarehomes,suchasaskingdistrictnursestodorootcauseanalysisofpressureulcers;

67

• gettinghealthteamsmore involvedintheprogrammeandworkingmorecloselywithcarehomesonprevention;and

• focusingonunblockinggapsandpromotingmorejoinedupworkingwithhealthservices;

“Weneedbetterrelationshipswithhealth.PROSPERcouldhelpwiththis,forexamplesothe

hospitaldon’tblameusforthings.PROSPERneedstohelpstrengthenrelationshipsbetween

thehospitalandcarehomes.PROSPERcouldactasavoiceforcarersandcarehomes.Get

hospitalstafftocomealongtoPROSPERtrainingsotheydon’tlookdownonus.Putnursesin

carehomesforone-daysotheycanseewhathappensandgivesolutions.”

[Managerfromalargehome,runbyacorporation]

Communication

• contactingcarersdirectlyratherthanalwaysgoingthroughmanagers;• providingregularreminderstocarersandmanagersaboutthingstotry;• givingfeedbackaboutwhatwasdoneasaresultofanycommentsprovided.Forexample,

when care homespoint out issueswith health or care services, care homeswould like toknowwhathappenedasaresultoftellingthePROSPERteam;and

• havingaFacebookpagetosharephotosandtips;

Training

• doingmoreneedsanalysisaboutwhattrainingisrequired;• including health staff in joint training, such as communitymatrons and district nurses, so

thateveryoneislearningtogetherandforgingbetterrelationships;• doingmorein-hometrainingsessions;• covering more topics at once during training sessions and recognising how topics are

interlinked;• runningadditionalsessionsinmoredeprivedareasofEssex;• runningtrainingupdatesregularly,e.g.everyquarter;• repeatingtrainingtopicstoreachalargergroupofstaff;• runningmorechampions’studydays;• including training about specific areas such as diet (not just hydration), dementia, and

palliativecare;and• focusingoninnovativeideasduringtrainingasmanysaidthetrainingwasatabasiclevel;

Othercontent

• providingideasabouthowtoengagewithresidentsandfamilymembers;• providingmoreeye-catchingposterswithpictures;• providingacontactsdirectoryforthelocalarea,e.g.wheretogotogetferrules, localfalls

preventionteametc.;and• focusinglessonthedriverdiagramsascarehomesoftensaidtheywere‘toocomplicated’;

Datacollection

• avoidingusingacrossshapefortheSafetyCrossbecauseofreligiousconnotations(usingastickorcirclecalendarinstead);and

68

• breakingdowntheMonthlyMappinggraphssocarehomescanseehowtheycomparetoothersintheirownlocality;bigandsmallandnursingversusresidentialhomes;

Usinginformationsensitively

• beingcarefulnottouseinformationaboutcarehomeslearntbythePROSPERteaminothermeetings,forexamplewhendiscussingsafeguarding;and

• beingcarefultoavoidtheimpressionthatcarehomesarebeing‘judged’.

AlmosteveryoneinterviewedsaidtheythoughtPROSPERshouldcontinueandthattheywouldliketoremaininvolved,especiallyifchangesweremadesoPROSPERofferedmoreon-goingvalue.

69

4. Implications

4.1 AchievingobjectivesOverall, our judgement is thatPROSPERhasbeena success. Individual carehomeshave reportedbenefitsfortheirstaffandresidentsandthiswascollaboratedbyindependentobservationsand,tosome extent, by quantitative before and after data. Table 4.1 highlights the extent to which theprogrammemettheoriginalaimsofPROSPER.

Table4.1:ComparingPROSPERaimswithoutcomes

Originalaim Progressafter21monthsToimprovesafetyandreduceharmforcarehomeresidentsacrossNorthEastandWestEssexbyimplementingamulti-facetedimprovementinterventionwitheducational,measurementandculturechangecomponents.

• Amultifacetedinterventionwitheducationalandmeasurementcomponentswasimplemented.Culturechangetoolswerenotprioritised.

• PROSPERexpandeditsreachtocoverallofEssex• Therewasasmallbutstatisticallysignificantreduction

inharmforresidentsusingallavailabledataforPROSPER.

• Therewerechangestohowsafetywasperceived.Co-designsolutionswithresidentsandtheirrelativesandintroducenewQItrainingopportunitiestobuildstaffcapabilitiestoaddresssafetyconcerns.

• Trainingwasintroducedformanagers,deputies,carersandotherstaff.

• TherewaslessemphasisfromPROSPERonco-designingsolutionswithresidentsandrelatives,thoughtherearesomeexamplesofthistakingplaceinindividualhomes.

ReducethepercentageofresidentsthatpresentatA&E.

• TherewasnostatisticallysignificantreductioninA&Eattendance.

Increasetheproportionofresidentswhoare‘harmfree’(asdefinedbytheNHSSafetyThermometer).

• TheNHSSafetyThermometerwasonlyimplementedwithcohort1andofferedtosomecohort2.Theresultsdidnotsuggestanincreasedproportionofresidentswhoare‘harmfree’.

Reducetheprevalenceoffallsandpressureulcersacrosscarehomes.

• TherewasasignificantreductioninfallsusingallavailabledataforPROSPERbutnotusingdataderivedonlyfromhomeswhichcollectedbeforeandafterdata.

• TherewasasignificantreductioninpressureulcersusingallavailabledataforPROSPERbutnotusingdataderivedonlyfromhomeswhichcollectedbeforeandafterdata.

Increasestaffunderstandingof,capacityandcapabilitytocreateasafeenvironmentforresidents.

• 67%ofhomesreportedachangeinsafetyculture.• 67%ofhomesreportedchangestocareprocesses.• Managersandstaffreportedincreasedknowledge

aboutsafetyissues.Understandtheconsequenceonhealthandsocialcarecosts.

• Weperformedapreliminarycostanalysisbutpoordataqualitymakesuscautiousaboutdrawinganyconclusions.

Establishanevidencebasefortheintervention.

• Adetailedevaluationwasundertaken.AchallengewasattributingchangestoPROSPERasotherprogrammeswererunning.However50-67%ofhomesbelievedPROSPERmadeadifference.

70

4.2 KeylessonsThissectionsummarisesthekeylessonslearnedfromPROSPER.

TheimportanceofcontextPROSPER reinforced that improving safety in care homes is a complex task requiring alignedinterventionsandastrong focuson theneedsof residentsandstaff.The interplaybetweenthreeelementswas key: an understanding of the context, the use of evidence-based interventions andeffectiveimplementation.

The care home sector is heterogeneous in ownership, size, resident characteristics, staffing levelsandaccesstosupport.Tobesuccessful,thisprogrammesuggeststhatinitiativesneedtoaddresstheinternalandexternalmotivationsofstaffandthesocialandtechnicalelementsofimprovement.Itisimportant for interventions to be tailored to, and relevant for, the care home context and torecogniseandtargetdifferentlevelsofthesystem,fromthedailycarereceivedbyresidents(micro-level),throughtocarehomestructuresandprocesses(meso-level),throughtothewidercontextinwhichcarehomesoperate(macro-level).

Whilstitisclearthatsomesafetyimprovementinterventionsaretransferablebetweensectorsandorganisations,improvementinitiativesneedtobeadaptediftheyaretobeacceptableandtoworkeffectively. Care homes have very different purposes, philosophies, staffing levels, structures,resident characteristics and workforce capabilities to the NHS and it cannot be assumed thatinterventions thatwork in healthcarewill automatically be relevant, acceptable or feasible in thecare home context. Theprocess of adaptation generates new learning that can be useful to bothsectors.Forexample,thecarehomesectoradoptsamoreuser-centredapproachtosafetythantherelativelyrisk-averseandprofessionallycentredmodelofthehealthsector.

RobustpreparationPlanning is key.We have learnt that more attention could have been put into the developmentphase of both the implementation and evaluation so that issues were ironed out early on andtoolkits, driver-diagrams and examples were collated ready for care homes. A six-month fundeddevelopmentperiodwasbuiltintoPROSPERbutinretrospectthiswascouldhavebeenusedmoreeffectively. The programme was not adequately staffed during the set-up phase so there weredelays in recruitment of care homes, providing information for them, and the development andpiloting of toolkits and data collection mechanisms. Similarly, the evaluation team experienceddelays in recruiting staff, gaining ethics approval and ensuring that the design and proposedimplementationwereoptimal.

LearningbydoingandsharingPROSPERfoundthatinitialtraininginqualityimprovementhelpedtoincreasethefamiliarityofcarehomeswith terminologyand tools, but teamsmay learnbest throughexperiential learning.Morespecifically, they seem to have a stronger orientation towards an oral and relational approach tolearning, rather than a written and technocratic one. This may be particularly relevant to staffgroups that are less familiar with more traditional approaches to learning. Regular support toimplementchangeinpracticehelpscarehomestoapplynewknowledgeandcontinuelearning.

71

CarehomestakingpartinPROSPERalsolearntbysharingideaswitheachother.ThePROSPERteamwereaconduitforthis,helpingtobreakdownbarriersandcompetitionbetweencarehomes.

Not only did care homes learn through PROSPER, but so too did the implementation team. Theteam’simprovementandfacilitationskillsdevelopedconsiderably.

MeasurementforimprovementisfundamentalCarehomeshaveatraditionofcollectingdataforaccountabilityandregulatorypurposes,butthereis not a strong tradition of collecting, analysing and interpreting data for improvement. PROSPERhelped to change this and carehomes respondedwell to the SafetyCross andgraphsofmonthlyincidentrates.Abouthalfofcarehomessaidthatbeingabletocomparechangesovertimebecamevaluableand theyhadasenseofownershipof theirdata.Carehomesalso found itmotivating tocomparetheirdatawiththeaverageofotherparticipatingcarehomes.

Couplingimprovementmethodsandsubstantive(content)knowledgeWhilst change requires a commitment to using evidence-based improvement interventions,introducing quality improvement tools alone is not enough. Developing substantive contentknowledge about pressure ulcers, falls and urinary tract infectionswas also important. Themosteffective balance between improvement process and content knowledge is not clear from theevaluation.

BerealisticabouttheeffectivenessandeffectsizeofimprovementinterventionsTheevaluationteamwasclearatthestartoftheprogrammethat,onthebasisofexperienceandthe published literature, whilst some adaptation in processes and culture as a result of theinterventionswereexpected,changesinsafetyeventsmightbemoredifficulttodemonstrate,andchanges in the use of NHS services were highly unlikely. This is a consequence of many factorsrelating to time, implementation, staffcapacityandcapabilityandthemethodsofevaluation.Theresultsoftheevaluationbearoutthiscaution.

IssueswithsustainabilityWhilstuseful for catalysing change in carehomes, theexpansionofPROSPER intoamoreholisticsupportprogrammeraisesquestionsaboutthepracticalitiesofspreadandsustainabilitywhentheoriginal fundingceases.Successseemstodependgreatlyontheskills,capacityandenthusiasmofthefacilitationteam.Theextenttowhichthiscouldbecontinuedorreplicatedisuncertainbutwillhopefullybeguidedbythefindingsofthisstudy.

A phased approach to cohorts worked well rather than a ‘big bang’ approach. The turnover ofmanagersissignificantinthecarehomesectorsousingthecohortapproachhelpedtoreducelosstofollow-upandgavecarehomesmultipleopportunitiestotakepart.

AttributionIt is unclear from the evaluation whether the positive outcomes are a result of introducingimprovement methods (as originally planned), the extensive holistic support that was offered, aconsequence of other external factors, or a combination of all of these. From an academicperspective,thisisanimportantquestionandwouldrequireamorerigorousevaluativedesignandmore resource to address. From a practice perspective, the pragmatic and relatively light-touchapproachtoevaluationaddedvalueandwasacceptable.

72

PartnershipsPROSPER reinforced the importance of health and social care working together. Although carehomesdidnotreportseeingimprovedjointworking,theadditionaltrainingofferedwouldnothavebeenavailablewithoutsupportfromhealthcarepartnersandhealthmanagers,andfrontlineteamssaidtheyfelttheyhadgainedmoreaccesstocarehomesviaPROSPER.

Inaddition,aparticipatorydesigntoimprovementprojectshasclearbenefits.Inparticular,theroleof an evaluation teamworking closelywith improvers, negotiating different bodies of knowledgeandexpertiseaddsvaluebutisalsochallengingtoenactinpractice

ChangetakestimePROSPERsuggeststhatjustchangingindividualstaffsafetybehavioursisimportantbutnotenough.For change to be sustained, it is important that this infiltrates the culture of the organisation soimproving safety becomes a shared norm. Helping leaders see the value of improvement andempowering a wide range of staff to do things differently can take time. PROSPER has beenassociatedwith some improvedoutcomes butmore substantive and embedded changemay takelongertoachieve,especiallyamongstcarehomesthatwerenotvisitedregularlyorfeltlessinvolved.

RecognisingvariationAroundone-quarterofcarehomesformallywithdrewfromPROSPER(24%)andonlyaroundhalfofcarehomesreportedthattheywereactivelyparticipatinginPROSPERatanyonetime.TheredidnotseemtobeanyconsistenttrendsinthecharacteristicsofcarehomesforwhichPROSPERwasmostsuccessful.Amixtureoflargeandsmallcarehomes,thoseownedbylargecorporations,familiesandcharities,andthose indifferentpartsofEssexallgainedsomebenefit.Onenotablesuccess factorappearstobethemotivationandexperienceofthecarehomemanager.Managerswhoareopentochange, open to accepting support and empowering the wider care home team tended to seegreatersuccessmorerapidlyfromthePROSPERapproach.

4.3 SustainabilityforEssexCountyCouncilEssex County Council has committed to providing financial support for PROSPER through to June2017. This will include funding for three full-time staff to run the programme, overseen by thecurrent project lead. These resources can be used flexibly, but the aim is to continue a level ofsupportforexistingcohortsandbringonthreeorfournewcohorts,eachwithabout20carehomes,overthenextyear.

This continuation of the programmewill be further supported by existing skills and resources atEssexCountyCouncilandinthecarehomes.Throughouttheearlypartoftheprogramme,Councilstafftookpartinqualityimprovementtrainingsotheseskillsarenowacorecapabilityofmanystaff,who can in turn provide mentorship and training. Communities of practice for care homes areestablishedandworkingwell, and thenetworkofPROSPER“champions”have theknowledgeandskillstosupportexistingandnewcarehomes.

EssexCountyCouncilwillbeassumingresponsibilityforhostingthedataprovidedbythecarehomestomonitorandtrackquantitativeimpact,arolefulfilledbytheevaluationteamduringthisproject.

73

4.4 DisseminationThroughout the project there was a strong focus on communicating PROSPER’s progress andfindings, based on a communications plan developed with the UCLP and Essex CouncilCommunicationsteams(seeAppendix10).

Local communication and disseminationmethods included newsletters, emails, events and onlinevideossenttoormadeavailabletocurrentorpotentialparticipantsintheprogramme.Additionalapproaches have been used to raised the programme’s profile outside Essex: Over the past two-years, ProfessorMartinMarshall has givenover 25presentationsnationally and internationally atconferences and events referencing PROSPER and its impact as an example of participatoryimprovementandevaluation.Theimplementationteam,ledbyLesleyCruickshank,haspresentedatover10eventsandconferences,includingattheInstituteforHealthcareImprovement(IHI)/BritishMedicalJournalInternationalForuminGothenburg,Sweden.

Multiple publications are planned to share the findings of the project, covering individualcomponents of the intervention (e.g., the updated MaPSaF tool), the overall impact of theprogrammeandtheevaluationmethodology.

Further dissemination will continue to occur through the Academic Health Science Networks(AHSNs). UCLPartners communication team will provide continued communications supportthroughmultiple channels, includingemails,websitesand socialmedia. Theywill also continue toraiseawarenessoftheprogrammethroughthenationalnetworkofAHSNs.Goodrelationshipshavebeen built with Eastern AHSN, who can also help support dissemination. UCLPartners and EssexCountyCouncilwillmakealltoolsandtemplatesavailableontheirwebsitessothatotherscanmakeuseofthem.

ThepositiveresponsefromCQCanditsreferencesininspectionreportsmayalsoactasanincentiveto get more care homes aware of and involved in the programme. UCLPartners is increasinglyworking with care homes and local government and will actively and directly promote theopportunitiesthatthePROSPERprojectoffers.Finally,thereiswiderlearningfromPROSPERwhichmightberelevanttoimprovementinitiativesinothersettings.

74

4.5ConclusionThis evaluation of PROSPER suggests that the programme has been largely successful and hasgenerated new learning about transferring improvement approaches from the health to the caresector. Theprogramme showed thatquality improvement approaches couldbe implemented in acarehomecontext,with careful adaptationand skilledand regular facilitationand support.Box5summarisestheimplicationsofPROSPERforcarehomes,forlocalitiesandforthewidersystem.

Box5:OverarchingimplicationsfromPROSPER

Implicationsforcarehomes• createspaceandtimeforcarehomestafftolearnaboutnewwaysofworking;• celebrateoftentocreatemomentumandbuildconfidenceinstaff;• workasateam,acrossallstaff,andwithresidentsandtheirfamilies–everyonehasideasto

contribute;• thelittlethingscount–usingjellytoboosthydration,decoratingwalkingframessothey

don’tgetmixedup;and• buildthecontentknowledgeofstaffaboutcareprocesses,itisasimportantasimproving

theirknowledgeabouthowtoimprove

Implicationsforlocalities• createanethosofco-productionandcollaborationwithcarehomestobuildanewkindof

relationship,workingasequals;• tailortheprogrammetothespecificcultureofcarehomes,includingtheirdesiretolearnby

doingandtheirstrongerorientationtowardsanoralandrelationalapproachtochange,ratherthanawrittenandtechnocraticone;

• developimprovementfacilitators,butusethelearningfromthisprojecttoshapethatrole,includingwhattheyshoulddoandshouldnotdotoaddvalue;

• adjustandadaptbasedonexperienceandfeedback’• findwaystomaintainenthusiasmandsustainparticipationandimprovement;and• berealistic-Itishighlychallengingtodemonstratereductionsinhospitalattendancesand

admissionsasconsequenceofspecificintervention,evenwhenthoseinterventionsappeartowork

Implicationsforthesystem• adoptaflexible,multi-facetedandcontext-sensitiveapproachtoimprovement,focusingin

particularonrapidfeedbackofdata(whichisrarelydoneincarehomes)andsharinglearning;

• provideopportunitiesforCareHomestodevelopasenseofidentityandprideinthehealthandsocialcaresystem;

• recognisethatsafetyimprovementapproachesandtoolsusedintheNHScannotbetransferredtothecarehomesectorwithoutadaptation;and

• supportthecarehomesectortomovebeyondregulatorydriversforimprovementtointernallymotivatedapproaches.WorkwiththeCQCtoreinforcethisthroughrecognisingthevalueofsimilarinterventionsandprogrammes.

Appendices

Appendix1.Evaluationmethods

This appendixdescribes themethodsused toevaluate thePROSPERprogramme.Theoverarchinglogicmodelandandanoverviewoftheevaluationapproacharedescribedinthemainbodyofthereport.

ProgrammedocumentationTheevaluation team read and incorporated insights from523 reports of homevisits,PDSA cyclesandotherprogrammedocumentsfromtheimplementationteamandcarehomes.Thisinformationwasusedascontextandtohelpshapesomeoftheinterviewquestions.

CollectionandanalysesofmonthlyincidentratesThe outcome data and care home characteristics were collected by The Evidence Centre using amonthly data collection form. Data was collected for a period between one-year prior to theinterventionmonthandMarch2016.Carehomeswereaskedtoprovidethenumberofeventsforeachoutcome(falls,pressureulcersandUTIs)onamonthlybasis.Other information includedthetotalnumberofresidentsandtotalnumberoffemaleresidentseachmonth,thecarehomelocationandotherbackgroundinformationasdescribedinTable1.4.Althougheffortwasmadetoencouragecare homes to submit the data, not every care home was willing to share their data with thePROSPERteam.TableA.1 shows thenumberof data collectionpoints fromeach carehomeamongst the 64 carehomesthatprovideddata.Fivecarehomes(onefromcohortone,twoeachfromcohortstwoandthree)provideddataforlessthantherequiredsix-monthperiodbeforeoraftertheintervention.TableA.1.Numberofdatacollectionpointsforeachcarehome

Therateofeventswascalculatedasnumberofeventsdividedbynumberofresidents inthecarehomes,asobservedateachmonth.

HOME_IDCohort 1 (from Jul 2013) 22 26 27 29 30 31 34 35 37 40 47 48 50 55 59 60 79 Total

No of data points 18 27 27 33 13 20 32 4 18 33 21 31 11 11 27 20 9 355% 5.07 7.61 7.61 9.3 3.66 5.63 9.01 1.13 5.07 9.3 5.92 8.73 3.1 3.1 7.61 5.63 2.54 100

Cohort 2 (from Jan 2014) 28 32 36 38 39 41 42 43 44 52 53 54 57 58 75 77 TotalNo of data points 27 27 26 25 30 14 27 5 21 14 14 21 17 3 12 14 297

% 9.09 9.09 8.75 8.42 10.1 4.71 9.09 1.68 7.07 4.71 4.71 7.07 5.72 1.01 4.04 4.71 100

Cohort 3 (from Jul 2014) 61 62 63 65 66 67 72 73 74 78 92 95 103 TotalNo of data points 9 21 19 3 9 7 13 10 17 8 16 13 3 148

% 6.08 14.2 12.8 2.03 6.08 4.73 8.78 6.76 11.5 5.41 10.8 8.78 2.03 100

Cohort 4 (from Jan 2015) 33 49 76 81 83 84 85 86 87 90 91 93 96 97 98 99 101 102 TotalNo of data points 27 25 12 13 11 8 15 1 9 15 11 6 2 2 13 4 1 3 178

% 15.2 14 6.74 7.3 6.18 4.49 8.43 0.56 5.06 8.43 6.18 3.37 1.12 1.12 7.3 2.25 0.56 1.69 100

77

Outcome data were plotted as time-series graphs with a one-month interval. The rates ofpre-interventionandpost-interventionwerecomparedusingchi-squaredtest (i.e. totalnumberofevents over total number of residents observed between pre-intervention period andpost- intervention period). All statistical analyses were carried out using SAS statistical software(version9.3).CostanalysisDirectcostsandcostsavingsrelatedtotheoutcomeswerecollectedforthestudy.ThetotalcostsofthePROSPERprojectandthecostsforeachcomponentincluded1)Programmesetupcosts;2)Staffcosts;3)Supplyoftechnicalskillcostsand4)Travelcosts.Foreachoutcome,costsavingwasdefinedas the costper additional event avoided. Thiswasestimatedusingpublisheddata at thenationallevel.

We carried out a preliminary cost analysis based on the resident outcome analysis using all datafromallcarehomes,whereavailable.Wereviewedtheliteratureandusedstandardnationaldatatoestimate the cost of individual harms. The total cost saving was summarised as overall savingsbecausewewereunabletoattributecoststoanyoneoutcome.Wethereforeusedthesumofcostsavings from each outcome (i.e. the sum of cost savings from A&E attendances or hospitaladmissions resulting from falls, pressure ulcers or urinary tract infections). For each outcome thecostsavingwascalculatedasnumberofeventsavoidedattheendofthe6monthfollowupXcostforeacheventavoided.Numberofeventsavoided=expectednumbers–observednumbers,whereexpected numbers = rate of event in the sixmonths prior to the intervention X total number ofresidentsinthesixmonthsaftertheintervention.

Furtherinformationaboutthequantitativeanalysisofthedataisprovidedinsection2.3.

CarehomevisitswithstaffinterviewsandobservationTheevaluationteamundertookobservationsandinterviewsduringvisitstocarehomestoseehowthe programme was impacting in people’s work environments and to track changes in a smallnumberofhomesovertime.ThepurposeofthesitevisitswastoobserveprocessesandPROSPERmaterials inuseand to speakwithawide rangeof staff in theirownworkingenvironmentaboutimpactsoftheprogrammeandhelpfulandhinderingfactors.ThevisitstookplacebetweenAugust2014andApril2016.

Ten care homes were visited (18% of the 57 homes in cohorts 1, 2 and 3), four on repeatedoccasions. These ten care homeswere selected as follows: two care homes from the first cohortweresuggestedbytheimplementationteamattheoutset,basedontheirknowledgeofthehomes,includingbeingindifferentpartsofthecounty,beingofdifferentsizeandownershiptypeandbeingatdifferentlevelsofdevelopment.Aperceivedwillingnesstotakepartinrepeatedvisitswasalsoafactorinselectingthesetwohomes,whowerevisitedfourtimeseachoverthetwoyearperiod.Twofurtherhomeswereselectedfromcohortonebytheevaluationteamforanindividualvisitbasedondifferingextentsofengagementintheprogrammeandperformanceonkeyindicators.Threehomeswere selected from cohort two and three from cohort three, all by the evaluation team. One ofthesehomes fromeachof cohort two and three received two repeated visits and the rest in thesubsamplewerevisitedonce.Foralltenhomes,samplingwaspurposefulinordertoensureamixoflargerandsmallerhomes;avarietyoffamily,charityandcommercialownership;spreadacrossthegeographicarea,andmorefavourableorchallengingperformancebasedonCQCreports,PROSPER

78

indicators and feedback from the local authority. The evaluation team attended training sessionsandcelebrationeventswherehometeammemberscouldbeobservedtofurtherselecthomesforvisits.

Twoevaluatorstookpart ineverysitevisit.Sixtyhoursworthofobservationswereundertaken. Inhomesvisitedmorethanonce,anevaluatorspenttimesittinginsharedareassuchasloungesandstaff rooms, observing processes. In every visit, the evaluatorsmade field notes during and aftereach visit and shared these with the wider evaluation team for input and questions. The notescirculateddidnotidentifythenameofhomesnorindividualstaffmembers.

103one-to-onediscussionswithmanagers,deputiesandotherstaffwereconductedduringhomevisits. The approximate proportion of interviewees was 45% owners, managers or deputies, 13%otherseniorcarers,10%nursingstaff,18%carers,12%domesticstaff(kitchenandcleaning)and2%others such asmaintenance staff. Theseproportions are approximatebecause theexact role titlewasnotknowninsomeinstancesorstaffmayhavehadmultipleroles.Theaimwasnottorepresenttheexactproportionsofstaffworkingincarehomes,butrathertohaveamixofjuniorandseniorpersonnelandcarersanddomesticworkers.

Allfacetofaceinterviewswithstaffconductedduringsitevisitsweresemistructured.Participantswereselectedbytheevaluationteaminpartnershipwiththehomes,basedoninvolvementinthePROSPERprogrammeandroles.Athemeguidewithquestionsaboutperceivedimpactsanduseoftoolswasused,inlinewiththeevaluationobjectives,butthequestionsweretailoredtotheroleandlevelofengagementofeachinterviewee.ThisincludedvaryingthelanguagestylewhereEnglishwasan additional language or where educational level was low. Interviews lasted from between tenminutesand1.5hours,againdependentonroleandlevelofengagement,aswellasstaffavailability.The interviews were not taped because care home managers and steering group memberssuggested that thismaybeoffputting,especially for junior staffwhomaybeconcerned that theircommentswould be heard bymanagement or by the local authority. However noteswere takenduringinterviewsbytheexperiencedinterviewer,includingverbatimquotes.Inmostinstances,oneevaluator asked questions and interacted with participants whilst another transcribed as muchverbatimasfeasible.Interviewnotesweretypedandsharedwithparticipants.

Twelvediscussiongroupswereheldincarehomesduringsitevisits,wherecarersanddomesticstaffhad an opportunity to share their views. Two discussion groups took part in each of the homesreceivingrepeatedvisitsandfourothersessionswererunduringoneoffvisits.Focusgroupsusedasimilarrecordingapproachtoonetooneinterviews.Participantsforfocusgroupswereselectedbyhomes,dependentonstaffavailabilityonthevisitdate.Groupsranged insize fromthreetoeightpeople. Discussions lasted no longer than an hour. Staff were able to come and go during thediscussionasneededtoprovidecarewithinthehome.ThediscussiongroupswereusedmainlytounderstandanychangesmadeinhomesasaresultofPROSPERandanychallengesandhowthesewereovercome.Inmostcasesmanagerswereaskednottoattendthediscussionssolessseniorstafffeltabletospeakfreely.

Adhocdiscussionswithasmallnumberoffamilymembers(13)andresidents(numbernotcounted)wereundertakenduringhomevisits, thoughthiswasnotaplannedpartof theevaluationdesign.Questionsforfamilymembersandresidentsfocusedonhowmuchtheywereinvolvedwithplanningimprovementsandanychangestheyhadseenovertime.

79

TelephoneinterviewswithseniorcarehomestaffThesitevisitswithinterviewsandobservationaimedtogaindetailedfeedbackandobservechangesinasmallnumberofhomesover time.Togain feedback fromawiderrangeofhomes, telephoneinterviewswere conductedwith care homemanagers and senior staff to collect their reflections.Everyparticipatingcarehomeandtenhomesthathadchosentowithdrawduringtheprojectwereinvitedtotakepartininterviews.Participatinghomesreceivedaninvitationthreetimesduringthetwoyearprogrammetoallowchangestobeexaminedovertime.In total, 203 telephone interviews were undertaken with staff from participating homes. Thiscomprised102staff from67carehomes (74%ofhomes fromall cohorts),withhalfof thesestaffbeinginterviewedmorethanonce.Owners,managers,seniorcarersandPROSPERchampions(staffactively engaged in the programme) were the focus of these interviews because these were theteammembersmostcloselyengagedwith theprogramme.66%of the telephone interviewswerewithowners,managersordeputiesand37%werewithfrontlinestaff.Everyparticipatinghomewasapproachedbytelephoneandemail tonominatestaff toparticipateand book time for calls. The evaluation team asked that a manager or deputy plus a PROSPERchampion (frontline teammember)benominated fromeachhome, inaddition toanyothers thatthehomesmaywishtopropose.Giventhat103inpersoninterviewswereconductedwithawiderrangeof staff during site visits, itwasdeemedappropriate to limit telephone interviews to thosemostengaged.Telephonecallsweremore feasible in termsof costand time for theseadditional interviews thanvisiting all 90 participating homes to speak with people in person. This approach was also moreconvenient forhomesas staff couldnominateanappropriate timeand rearrangewith littleornonoticeifneeded,inlinewiththeoperationalrequirementsofthehomes.InthemajorityofcasestheevaluationteamhadmethomestaffduringPROSPEReventsor training,so therewasadegreeoffamiliaritybeforethetelephoneinterviewstookplace.A semi structured interviewprocesswasused,with abrief setof questions andopportunities forstaff tocommentonother issuesofpriority for them. Interviews lastedbetweentenminutesandonehour,dependingontheavailabilityofindividualstaffandtheextentthattheywantedtomakecomments.Wherepermissionwasgranted,telephoneinterviewswererecordedandtranscribed.Ininstances where permission to record was not granted (two thirds of cases), the interviewertranscribed verbatim notes during the interview. The interviewer was experienced with thisapproachandwasable to transcribequotes indetail.Briefnotes fromeachcallwereprovided tointervieweestocheckpriortofinalisation.Tenhomesthatchosenottotakepartorwhowithdrewearlyafterbeginningtheprogrammewereselected for a telephone interview. Managers from all of these homes consented to a shorttelephoneinterview.Asemistructuredinterviewschedulewasused,focusedonthereasonsfornotparticipating and suggestions about solutions to any barriers encountered. The transcriptionapproachwasthesameasforallotherinterviews.

80

ObservationofprogrammeactivitiesTheevaluationteamobserved12trainingsessions,communityofpracticeevents,champions’studydaysandotherprogrammeeventsoveran18monthperiod.Atotalof67hoursofobservationtookplace.Fieldnoteswerepreparedduringandaftertheevents.Thepurposeofattendingwastohearfeedbackfromhomesaboutimplementationandimpactsandtoobservetheimplementationteaminaction.

Feedbackabouttheeventstothe implementationteamwasprovidedatmonthlyevaluationteammeetingsand inwritingsothat independentreflectionsabouteventscouldbeusedtohelpshaperevisions to further training and events. Where quotes or stories from care home staff werecollected during events, these homes were followed up in writing and by telephone to confirmaccuracyandtheacceptabilityofusingthesestoriesforevaluationpurposes.

OnlinesurveyAbefore and after surveywasused to assessmorequantifiable changes in carehomemanagers’perceptionsofhomesafetyculture.Uponregistering to takepart in theprogramme,managersordeputiesfromhomesincohortsonetothreewereinvitedtocompleteanonlinesurvey.Theywereinvited again in March 2016. The invitation was sent by the evaluation team based on emailaddresses provided by the local authority. One email reminder and one telephone reminderwasprovided.Bothsurveysweredesignedbytheevaluationteam.The‘before’surveyaskedaboutcomponentsofsafetycultureexplicitlydrawnfromtheManchesterPatientSafetyFramework(MaPSaF),inrecognitionthatthistoolwasselectedtosupportculturesafetymeasurementandreflectionforthePROSPERprogramme.The‘after’surveyincludedthesamequestionsplusadditionalquestionsaboutPROSPERcomponentsandimpacts.Thequestionswerepilotedwithasampleofsixhomesfromoutsidetheareatotestvalidityandeaseofuse.Carewastakentouselanguageappropriatetotheeducationallevelofstaff.Matchedbeforeandafterdatawereavailablefor51carehomes(89%ofthetotalhomesincohorts1to3).TheStatisticalPackagefortheSocialSciences(SPSS)wasusedtocompareanswersbeforeandaftertheprogramme.Statisticaltestsofdifferencesinscoreswereundertaken,usingtwo-sidedt-testsatthe95%levelofconfidence.Followupoccurredbetweeneight-and20-monthsaftertheinitial survey, depending on the cohort start date. The cohort numberwas used to crosstabulatefindings to account for different durations of follow-up. Open ended feedback was coded andanalysedusingtheconstantcomparativemethodandreportednarratively.Quoteswereextractedtoillustratekeythemes.

AnalysisofManchesterPatientSafetyFrameworkdataInitially the implementation team planned to help all homes use the MaPSaF tool to guidediscussions about safety culture at several points in time during the programme. Theimplementation teamplanned to compare the consensus scores derived from this discussion toolovertimeasawayofindicatingchange.OnequarterofhomesincohortonecompletedMaPSaFusinga‘surveyapproach’designedbytheimplementation team. The evaluation team reviewed these data to explore preliminary feedbackabout safety culture.However, the implementation teamdidnotprioritiseusingMaPSaFasakey

81

componentofPROSPERandforthisreasonitwaslesseasytotrackchangesovertimeusingthistoolthanwasoriginallyplanned.

InterviewsandfocusgroupwiththeimplementationteamThroughout the programme there were discussions between the implementation team andevaluation team. The evaluation team attended four implementation ‘project team’ and steeringgroup meetings and the implementation team attended all of the evaluation team meetings toprovideon-goingfeedback.There were 11 additional interviews with the implementation team by telephone and in person.These discussions were unstructured, in response to developmental issues in the programme atspecific points in time. Notes were kept during discussions and key themes were circulated forchecking by participants. In large part, feedback from the implementation team came from twoseniorteammembers.A discussion group to reflect on successes and lessons learnt was also held with the widerimplementation team towards the end of the programme. This included all junior and seniorimplementationteammembers.Thiswasfacilitatedbyonememberoftheevaluationteamusingasemi structured framework.Noteswere kept and circulated for feedback following thediscussiongroup.

InterviewswithotherstakeholdersTwenty three interviews were conducted with other stakeholders, including GPs, clinicalcommissioning group,NHS community services, and hospital teams and seniormembers of EssexCountyCouncil.These interviewswereconducted inperson(5)orbytelephone(18).Stakeholderswereidentifiedvianominationsbythelocalauthorityorbyhomes,inadditiontoinformationgainedindependentlybytheevaluationteam.Theycomprisedrepresentativesfromorganisationsthatmaybeinfluentialforthesustainabilityoftheprogramme(commissioningteams),organisationsthathadhelpedwiththedeliveryofPROSPERtrainingandcomponentsandorganisationsthatmaywork inpartnershipwithcarehomestosupportresidents.Purposefulsamplingwasused.Potentialintervieweeswerecontactedinitiallybyemailtoarrangeanappropriatetime.Allbutoneperson approached (who had since left their post) responded. Interviews used a semi structuredschedule.Interviewswererecordedwherepermissionwasgrantedorverbatimnotesweretakenbyanexperienced interviewerwherepermissionwasnotgranted (50%ofcases).Notesweresent toparticipantsforreview.

CompilationandanalysisofqualitativedataAllofthequalitativedatadescribedabovewascollectedbyaspecialistindependentevaluationteamwithexperienceinthesemethodswhohadworkedincarehomespreviously.Atotalofthreeevaluatorswereinvolvedincollectingthesedata.Oneevaluatorwasinvolvedinallaspectsofthedatacollection,withtwosupplementaryresearcherssupportingsitevisits,interviewsandtranscriptionwhererequiredandasdescribedabove.

Qualitativedatacollectiontookplaceonarollingprogrammethroughouttheimplementationperiod.Attendanceatprogrammetrainingandeventsoccurredapproximatelyeveryquarterfortwoyears,dependingonthescheduleoftheimplementationteam.Beforeandaftersurveystookplace

82

atthecohortinceptiondateandinMarch2017.Telephoneinterviewstookplaceinthefirstquarterof2015,thirdquarterof2015andfirstquarterof2016.SitevisitsbeganinAugust2014.Repeatedvisitstookplacebiannually.Homesthatwerevisitedoncehadvisitstimedaftertheendofthe‘intensive’phaseofprogrammeimplementation,whichvarieddependingonthecohortofeachhome.Feedbackfromtheimplementationteamwasprovidedmonthly.Interviewswithotherstakeholderswerelargelycompletedneartheendoftheimplementationperiod,inthefirstquarterof2016.

Alargeamountofdiversedatawerecollectedfromprogrammedocuments,homevisitfieldnotes,eventfieldnotes,focusgroups,interviewnotesandtranscripts.AllnoteswereinputintoNVivosoftwaretosupporttheextractionofkeythemes.Themeanalysiswasundertakenguidedbytheunderpinningevaluationresearchquestions.Theconstantcomparativemethodwasusedtoidentifyrecurringthemeslinkedtotheprogrammetheoryandevaluationquestions.Fieldnotesandinterviewtranscriptswereannotatedwiththemecategoriesandsubcategoriesandquoteswereextractedtoillustratepointsandprovideaflavouroftheperceptionsandstrengthoffeelingofparticipants.

Initiallythethemeanalysisfocusedonadescriptiveprecislinkedtotheevaluationquestions.Astrendsemerged,theanalysisexploredmorecomplexideas.Theevaluatorsexploredwhetherthemesdifferedbasedoncohortnumber,participantroleandseniority,geographiclocation,homeownershiptypeandothercharacteristicsoftheintervention,participantsandhomes.TheextenttowhichviewpointsdifferedbasedonreportedengagementwithPROSPERandquantitativeresultswasalsoexplored.

Thepreliminaryanalysiswasundertakenbytwoevaluators,whoreviewedallmaterialindetail.Writtenpreliminarythemesanalysiswasthencirculatedtothewiderevaluationteam,alongsideaccesstoallrawmaterialssothatmultipleanalysisperspectivescouldbeincluded.Throughouttheprogrammetheevaluationteammetmonthlyorbimonthlytoreviewemergingthemesandconsiderotheravenuesforexploration.

Duringtheprogramme,asteeringgroupprovidedasensecheckaboutthemesemergingandraisedfurtherquestionstobeexaminedduringthequalitativeanalysisprocess.

TableA.2summarisestheevaluationmethods.TheEssexCountyCouncilethicscommitteeapprovedtheprogrammeandresearchgovernanceapprovalwasgrantedbyUCL.TableA.2:MethodsusedtoaddressPROSPERevaluationquestions

Evaluationquestion Methods1.WhatwasthePROSPERintervention?Whatwasinvolvedintheinterventionandhowwasitimplemented?

• ObservationofPROSPERactivitiesincludingtraining,homevisitsandteammeetings

• Reviewofprogrammedocumentation• DiscussionswithPROSPERteammembers• Telephoneinterviewswithcarehomemanagersandstaff• PROSPERteamrepresentedatevaluationteammeetings

2.WhatimpactdidPROSPERhave?Doestheinterventionimpact • Telephoneinterviewswithcarehomemanagersandstaff

• Visitstohomes

83

onsafetyculture? • Observationofcommunityofpracticeevents• AnalysisofManchesterPatientSafetyFrameworkfeedback• SurveyofhomemanagersbeforeandafterPROSPER

Doestheinterventionimpactonsafetyprocesseswithincarehomes?

• Telephoneinterviewswithcarehomemanagersandstaff• Visitstohomes• Observationofcommunityofpracticeevents• AnalysisofManchesterPatientSafetyFrameworkfeedback• SurveyofhomemanagersbeforeandafterPROSPER

Doestheinterventionimpactonresidentoutcomes?

• Beforeandafteranalysisofmonthlyincidentrates• AnalysisofNHSSafetyThermometerdata,whereavailable

Doestheinterventionimpactoncosts?

• CostmodelbasedonquantitativeanalysisandadditionalNHSdataaboutA&Evisits,admissionsandambulanceconveyances

3.WhatinfluencedtheimpactofPROSPER?Whathelpsorhinderstheintervention?Whatwouldhelpitembedandbesustained?

• Telephoneinterviewswithcarehomemanagersandstaff• Repeatedcasestudysitevisitswithselectedhomes• Observationofcommunityofpracticeevents• DiscussionswithPROSPERteam• DiscussionswithNHSandotherstakeholders• Interviewswithmanagersofhomesthatdidnottakepart

84

Appendix2.Ethicsapproval

85

Appendix3.ExamplesoftoolsusedinPROSPER

ThisappendixprovidesvisualisationsofthetypesoftoolsusedinPROSPERandthefeedbackgiventocarehomestogiveaflavourofthematerial.

NHSSafetyThermometer

86

SafetyCross

87

Monthlymapping

What do these graphs show? Congratulations on doing a good job of compiling your data and using it to make improvements. These graphs show how your home has been progressing to improve care. They show how many falls, pressure ulcers and urinary tract infections there have been compared to the total number of residents in your home. The lower the proportion, the better. The aim is to get your graphs as close to zero as possible. No one else will see these graphs. They are just for you to use. You might want to share with staff or put up on the wall to help spark discussion about what to focus on next. For example if you are doing well with one thing (e.g. low rates of falls) then maybe you will want to focus on making something else even better. What do the trends mean? These aren’t perfect figures, but the graphs show trends. You can see whether the graphs go down after you start to make changes. We have not inserted the exact numbers because the important thing is the trend over time. Hopefully you will be going down, rather than up. It’s normal to go up and down quite a bit. Comparing with other homes You can also look at how you are doing compared with other homes taking part in PROSPER. The ‘Average Essex’ line on the graphs is the average for the 60 homes that have provided numbers so farcify the blue line for your home is below the orange ‘Essex Average’ line, that means you are doing better than a lot of other homes. If the blue line for your home is mainly above the orange ‘Essex Average’ line, that means maybe there is more work to do in that area. You could use the PROSPER meetings or the PROSPER knowledge hub website to find out what other homes have tried and think about whether it would work for you too. It is not a competition at all. Every home is different and every homes starts from a different place. The important thing is that you can use the information you have to think about what might need to change to help residents and staff

MaPSaF/CultureisKeytoolCultureisKey A B C D E1.Whatisouroverallcommitmenttoquality,residentsafetyandcontinuedimprovement?Howhighaprioritydowegivetoresident,staffandvisitorsafety

Nobodyseemstobebothered

aboutthegeneralqualityof

careprovided.Thereisnot

enoughtimeorstafftodo

thingsproperlyorimprove

things.Notsureifpoliciesand

proceduresexist.Resident,

stafforvisitorsafetyislow

priorityandnotreallythought

about.

Qualityisonlythoughtabout

whenwehaveinspection

visits.Qualityauditsonlytake

placewhensomethinggoes

wrong.Policiesandprocedures

aretheretotickabox,nobody

paysattentiontothem

Resident,staffandvisitor

safetyisanafterthought.

Managementdothingswhen

askedbyotherslikeCQCorthe

LocalAuthority.Wearenot

askedforourinputinhow

thingsaredone.Quality

auditingtakesplacebut

nothingseemstochange.

Policiesandproceduresexist

andareupdatedbutnot

alwaysfollowed.Resident

safetytakesfairlyhighpriority.

Qualityofcareisvery

importantinhowwedothings.

Everyoneisinvolvedin

improvingqualityandincluded

inthedecision-making.

QualityAuditsareusedto

improvedqualityandlearn

frommistakes.A“no-blame”

cultureisencouraged.

Resident’sviewsandwell-

beingtakespriorityover

company’sself-protection.

ThereisarealfocusonQuality

anditisputbeforeanything

else.Thehomeisacentreof

excellenceandcontinuously

lookingtoimprove.Everyone,

includingstaffandresidents,

areinvolvedandawareof

potentialsafetyrisks.Resident

safety&well-beingis

constantlyoneveryone’s

minds.

2.Howdowelearnfromthecausesofresidentsafetyincidents?Whodoweinvolvefollowingthesafetyincident?

Whenthingsgowrongitis

covereduporignored.

Informationaboutanincident

isgatheredbutnothingis

done.Nochangesaremade

afteranincident.Thereisa

lackoftraining,awarenessand

riskmanagement.

Thehome/companyseesitself

asavictim.Thereisa

defensiveblamecultureso

thereisnopointinreporting

incidents.Aquickfixsolutionis

oftenputinplacebutthereis

nosupportforthoseinvolved.

Incidentstendtohappen

again.Noonetakes

responsibility.

Itisrecognisedthatsystems

alsocontributetoincidents,

notjustindividuals.The

home/companysaysithasa

fairculture.Ananonymous

reportingsystemisinplacebut

noonefeelsconfidentin

reportingincidents.Thereisno

ownershipoftheincident.

Thereisalackof

communication.

Everyoneacceptsincidentscan

beacombinationofindividual

andsystemmistakes.Wefeel

confidentreportingincidents

andareactivelyinvolvedwith

residentsandrelativesto

preventincidentsfrom

happening.Lessonsarelearnt

fromincidents.Residentsafety

ispromotedaspriority.

Investigationsareseenasa

chancetolearn,witha

commitmenttosharefindings

bothinternallyandexternally.

Resident’sviewsareaskedand

thereisastrongfocuson

improvement.Systemsare

regularlyreviewedwithahigh

levelofopennessandtrust

witheveryoneinvolved.

89

CultureisKey A B C D E3.Howdowecommunicatesafetyissueswithinthehomeanddoweinvolvekeypeoplewhereappropriate?

Peoplearescaredtospeakup

whenanincidenthappensor

couldhappen.

Thereisastrongblame

culture.

Wedon’ttellpeopleoutside

thehomeofsafetyissues.

Thereisadefensive

cultureofit’snotourfault.

Wemightchangethings

whensomethinghappens

butwedon’ttryandstop

itfromhappeninginthe

firstplace.Basictrainingis

given.

Thehomehaspoliciesand

proceduresbutnotsure

theyarefollowed.We

understandour

responsibilitiesbutnotthe

homesresponsibility.

Informationiscollectedon

incidents,fallsetc.butwe

don’tknowwhyorwhatit

isusedfor.

Thereisanopenandfairculture.Weare

encouragedtobeinvolvedinallaspects

ofsafety.Weareencouragedto

challengepoorpractice.Incidentsare

usedforlearning.Communicationis

goodandweincluderesidentsand

otherssuchasDistrictNurses,GP’s.

Weworktogetherwithothers

fromoutsidethehomesuchas

DistrictNurses,TVN’s,GP’sand

SocialCaretokeepresidents

safe.Anoblamecultureis

promoted.Residentsare

centraltoallcommunication

andsafety.

4.Howdoesthehomemanagestaffeducationandtrainingaboutsafetyissuesandriskmitigation?

Wearejustseenasbodies.

Traininghasalowpriority

unlessitismandatory.We

feelunsupportedby

managementanditisthought

wealreadyhavetheskillsto

dothejob.Nobody

understandsorcaresthatthe

lackoftrainingorstaffing

levelsincreasesrisk.Thereis

littleornosupervisionor

support

Staffinglevelsarelow

and/orwerelyheavilyon

agencystaffwhichcan

increaserisk.Trainingis

availablebutdowntous

toreadorwatchDVD’s

andaskquestionsifwe

don’tunderstand.Focusis

onsavingmoneyand

budgetsratherthan

residentchoice,safetyor

positiverisktaking.We

haveminimumsupport

andsupervision.

Trainingreflectsthe

residentsneedsand

focusesonprevention.

Trainingisnotalways

relevant.Proceduresare

seenasatoolfor

managementtocontrol

staff.Weallhavea

personaldevelopment

plan.Wearesomewhat

supportedbyseniors&

managerswithsome

supervision.

Somecommitmenttomatchpeopleto

postsandinductionistailoredto

resident’sneeds.Trainingiswell

plannedandweareencouragedto

develop.Trainingisavailablefor

everyone.Managementattemptsto

understandandmanagesafetyincidents

withgenuineconcernaboutour

wellbeing.Staffinglevelsaregoodand

helpsmanagerisk.Wearesupported

withregularsupervision

Commitmentisrecognisedand

rewarded.Greatconfidencein

management.Weare

motivatedandsupportedto

develop.Education&trainingis

recognisedasimportant.

Everyoneunderstandsriskand

safetymanagement.Learningis

adailyoccurrence.

Managementarefairandtreat

usconsistently.Wearewell

supportedandsupervised.

5.Howdowesharegoodpracticeandlessonslearned?Howdoweworkeffectivelyasateamtoimprovesafetyissues?

Coveringupofbadpractice

withacultureoffear.People

workinisolationwithno

responsibilitytakenfortheir

ownactions.Sharingand

communicationdoesnot

happen

Policiesandprocedures

areinplacebutwearenot

toldwhatthislookslikein

practice.Littlemotivation

andcommitmentto

establishasafer

environment.Wedon’t

workwithpeopleoutside

thehomelikehealthor

socialcare.

Systemsareinplaceto

collectdatabutarenot

wellused.Management

tellsuswhattodo;weare

notaskedwhatcouldwork

inpractice.Information

notsharedbetween

othersoutsidethehome

suchashealth.

Informationisnot

analysed.

Systemsandprocessesareinplaceand

areregularlysharedwithus.Good

practiceandlearningisownedand

sharedbyusall.De-briefshappenand

lessonsarelearned.Systemsand

processesaresharedwithothers

outsidethehomessuchashealthand

socialcare.

Sharingbestpracticeincreases

motivationandencourages

confidencetoofferasafer

service.Bestpracticeis

reviewedandchangesare

implemented.Wefeel

confidenttochallengeand

shareinformation.Asafe

environmentisseentobecost

effective.Weworktogether

withhealthandsocialcare.

90

CultureisKey A B C D E6.Howdoresidentsandvisitorstakeresponsibilityfortheirownsafety?Howdowesupportandrespectresidents’choice&controlofpositiverisktaking?

Choiceisnotallowed.Culture

ofnotallowingrisk,fearof

theconsequencesorarenot

insured.

Staff/residents/visitorsare

notinvolvedindecisions

aboutrisk

Somepoliciesand

informationareavailable

butwithnoforward

planning.Policiesareonly

referredtowhenan

incidentoccurs.Personal

freedomsarerestricted.

Risksarediscussedbut

withnofollowupactions.

Risksareonlylookedat

whensomethinghappens

withsomeinvolvementof

residents/visitors.

Informationiscollected

andthennotused.Risk

managementisseenasa

tickboxexerciseandlacks

meaning.

Residentsafetyispromoted.

Concerns/complaintsandfeedbackis

welcomed.Everyoneisencouragedto

beonboardwithpositiverisktaking

witha‘can-do’attitude.Personcentred

goalsarepromotedtogive

independence.

Careisalwayspersoncentred

withevidenceofpositiverisk

takingandbestinterest

discussions.Everyoneis

involvedinthereviewofsafety

issues.Visitorsareencouraged

togivefeedbackandhavea

roleinsafety.

PDSAworksheetWORKSHEET: PDSA Cycle Progress Sheet

This sheet is used to monitor progress as you complete your PDSA cycle. After completing this progress sheet, you will have a record of what you did, the information you studied, and what you will use from your PDSA cycle

Complete this part after you decide on the test or observation, including a plan for collecting data.

PLAN

•

Complete this part as you carry out your cycle. Keep notes on what happens. Before completing this, be sure you are clear on your plan, then move to the do, study and act steps.

DO

•

Complete this part after you have finished your testing and observations, having gathered your data and reflected on what happened. Include expected and unexpected results.

STUDY

•

Record what you will use from this cycle or what you will do differently next time. What other tests or cycles will you do?

ACT

•

Home/Floor/Unit

Start Date:

End Date:

92

Appendix4.Impactsonsafetyculture

This appendixprovides furtherdetails aboutpotential impacts fromPROSPERon safety culture incarehomes.Informationwascollectedforthissectionusingthefollowingmethods:

• reviewof523documents• 127hoursworthofvisitsandobservations• 203individualinterviewswithcarehomestaff• 12discussiongroupswithcarehometeams• 26meetings,discussiongroupsandinterviewswithPROSPERteam

ChangesinhowsafetyisdefinedIn interviews and discussion groups, about one-third of care homes said that they already had agoodsafetyculturebeforePROSPER,andsomeof thesefelt thatPROSPERgavethemthetools tostrengthentheiractions.Two-thirdsofcarehomesreportedmorefundamentalchangesinthewaythey thought about safety. There was a shift towards defining safety as preventing harm forresidentsratherthanavoidingissuesraisedduringinspectionsandsafeguardingconcerns.Thecarehomes that weremost engaged in PROSPER (andmost frequently visited by the implementationteam)reportedseeingthebenefitsforimprovingthequalityoflifeofresidentsandforempoweringstaff.Thesecarehomesweremorelikelytobeginengagingwithawiderrangeofstaffandfamilies,seeingeveryoneaspartoftheimprovementteam.

Althoughdifficulttoquantify,observationsandinterviewssuggestthatPROSPERsupportedashiftinhowmanagersandcarersinsomecarehomesdefinedsafety.AtthebeginningofPROSPER,safetywasoftendefined in termsof reducing risk for regulatorypurposes, suchasavoidingsanctionsbyCQC or local authority safeguarding investigations. However, as they gained more exposure toPROSPER,seniorcarehomestaffweremorelikelytodescribesafetyasbeingabout improvingthewellbeingofresidentsandworkingcollaborativelyasateam.Thequotesbelowfromonecarehomehelptodemonstratethisshift.

“Safety isabout reducingour riskof safeguardingproblemsandmaking sure thatweareok

when the CQC come. It is aboutmaking sure people don’t fall sowe don’t get in trouble or

don’t get blamed for pressure ulcers, even though they come in with them or come from

hospitalwiththem.It isaboutcoveringourselvesandtryingtodothebestfortheresidents.”[Managerofsmallhomerunbysmallorganisation,atthebeginningofPROSPER}

“Safetyisallabouttryingtomakelifeasgoodaspossibleforourresidents.Weworkforthem

andwewanttogivethemahomewithdignityandrespect.Wewantthemtohavequalityof

life.Itisaboutjudgingtherisksandgettingabalance.Thisisahome,notahospital.Wecould

avoidallfallsbyaskingpeopletostayseatedallthetimebutthentheywouldnothavegood

quality of life. So for me safety is about working together, all of us as a team, from the

domesticstothecarerstothecooksandmanagers,tohelppeopleavoidnastythingsbutstill

haveagoodqualityoflife.Topreventasmuchaswecanandthinkinnewwaysaboutdoinga

goodjobforourresidents.”[Samemanager,aftertakingpartinPROSPERfornine-months]

93

ChangesinawarenessManagersandcarerssaidtheyweremoreawareofsafetyissuesasaresultofPROSPER.

“I’vefoundthatithasmadeusallmoreaware.Ithasbroughtthingstolightthatweweren’tso

awareof,forexample,footwear,glassesandfindingreasonsforfalls.Itgivesyouachecklistto

thinkthrough.WewouldliketothinkwewouldbegoodwithoutPROSPERbutwewouldn’t

havethatchecklistapproachandtothinkofthereasons–ithashighlightedthethingsthatwe

shouldbeawareof.”

[Managerfromasmallhome,runbyasmallgroup]

“It’shelpedmystaffanawfullot.We’remoreawareofhowtopreventfalls.Weconcentrated

onUTIs.Nowtherearealwaysjugsofjuicearound.Thereisabigboardintheloungewithtips,

thecrosses,ourgraphsandnewsletters.Thissparksdiscussionwithstaffandrelatives.”

[Managerfromamedium-sizedhome,runbyacorporation]

“It’sverygood.Ithasimprovedthequalityofcareatgroundlevel.Ourfiguresshowreductions

infallsandpressuresores.Itremindsustogobacktobasics.Allthecarersandnursesseethe

SafetyCrossesintheunits.Thisraisesawarenessofwhatneedstobedoneeveryday.We’re

empoweringthecarerssotheyareinvolvedinwhatwe’redoing.”

[Deputymanagerfromalargehome,runbyacorporation]

ChangesinhowdataarevaluedAkeysuccessofPROSPERwasinhelpingtoshiftattitudesaboutdataonsafetyincidents.Nineoutoftencarehomesreportedlookingatdatamoreregularlyandusingittothinkaboutwhythingsarehappeningandwhatcouldbedonedifferently.

“WedoPROSPERineverystaffmeeting.Welookatcomparisonsoffallseachmonthandthink

aboutwhyithasgoneupanddown.WehaveonlybeendoingitsinceFebruary.Weactually

foundmorefalls–soweblamedPROSPERformakingusfocusonitmore(laughter).Butstaff

havereallytakentoitandcomeupwithnewideas.Lookingatthedataallthetimehasreally

changedhowwethinkofthingsbecausewearefocusingonpreventionnow.”[Managerofmedium-sizedhome,runbycorporation]

“PROSPERhelpstochangetheculture.Staffarespendingmoretimelookingatthings.Staffare

thinkingoutoftheboxafterseeingmonthlygraphs.”

[Careratsmallhome]

Anumberoflargercarehomesandthoserunbycorporationssuggestedthattheyalwayscollecteddata,butmanagerstendedtoreportittoacentralisedplaceratherthanusingittoguidetheirownimprovement activities in the home. PROSPER helped these care homes takemore ownership oftheirdataanduseitforimprovement.`

````

94

“TheactualmeasuresthathavecomeaboutasaresultofPROSPERhavebeengood.We

monitoredalotoftheKPIsanyway,suchasfallsandpressureulcers,butithasfocusedthe

mind,notjustformanagement,everybodyisthinkingaboutitnow.Ithastakenitfrombeing

onlythemanagerfocusingondatatogettingouttomorestaff–everybodyreallyhasastake

initnow.”

[Managerofmedium-sizedhome,runbycorporation]

SmallercarehomesweremorelikelytosaythattheydidnotcompilerecordstotrackchangesovertimebeforePROSPER.Recordsaboutpressureulcerswouldhavebeenkept inan individual’scareplan(ofteninpaperform),butnotdrawntogethertogiveapictureforthehomeoverall.Fallsmayhavebeenlistedinanaccidentregister,butlittlewasdonewiththedatatomapitovertime.

The Safety Cross andMonthlyMapping tools introduced by PROSPER helped care homes have asimplewayoflookingatpatternsandcelebratingsuccesses.

Another shift in attitudes towards data was evident in expanding the number and type of staffmembersinvolvedincollectingandusingit.BeforePROSPER,ifdatawerecompiledthiswaslargelydonebythecarehomemanagers.AfterPROSPER,carehomesbegantoseethevalueindelegatingresponsibilitytoawiderrangeofstaff.Thus,inabouthalfofcases,deputymanagers,seniorcarersorPROSPERchampionswereresponsibleforcollatingdatamonthly intographs.CarerswereoftengivenresponsibilityforfillingintheSafetyCross,whichwasdisplayedprominentlyonboardswithinthecarehomes.

ChangesintherangeofteammembersinvolvedAswell as expanding those involved in data collection, another key change in culture occurred interms of the variety of staff engaged in looking at this information, brainstorming ideas forimprovementandimplementingtheminpractice.

“Ithashelpedmetohaveloadsofchampions,tosupportmeasamanager.Weareinvolving

residentsandrelativestoo.Thishashelpedmeasamanagertoseedifferentwaysofdoing

things,soitisnotjustthemanagerbeinginchargeanymore,we’reallpartofittogether.”

[Managerofmedium-sizedhome]

PROSPERwasthoughttobevaluableforempoweringstaff.

“Itgivescarersresponsibilityandacknowledgementsotheycanbeempoweredtotrytheir

ideas,sonotjustlisteningtothesenioronduty.”

[PROSPERchampioninmedium-sizedhome]

Duringdiscussiongroups,carersspokeabouthowtheyfeltmoreempoweredandengaged.Theyfeltthattheyhadakeyroletoplayinimprovement,ratherthanamorenarrowlydefinedrolewashingorcleaning.CarerssaidthatthishadoccurredasaresultofPROSPER.

“IthinkIdoanimportantjob.Iamjustinthekitchenbutyouknow,that’simportanttoo.But

nowIreallylikemyjobmore.Latelymeandthegirlshavebeengivingourideas.Wealllisten

toeachotherandwetrynew ideas.Themain thingwe learnt is ifyoutryand fail, just try

somethingelse.Keeptryingtomakethingsbetter.WealllookonthePROSPERboardinthe

95

loungeandwecomeupwithnewthings.Theytriedmyidea…anditmademesortofproud.”

[Kitchenworkerinmedium-sizedhome,runbycorporation]

Thischangewasmostapparentincarehomesthatembracedtheconceptof‘PROSPERchampions’wherebycarersareallocatedresponsibilityforhelpingotherstafflearnaboutprevention–orwherechampionstakeonacertainrole(asinthecaseof‘jellychampions’deliveringjellytoresidentsdailyas a way of reducing the risk of dehydration). Managers usually choose staff to be champions,perhapsbecausethepersonhasaspecialinterestinhydration/fallsorbecausetherewasadesiretodevelop that person’s leadership and facilitation skills. Champions attended PROSPER champions’dayseveryfour-tosix-months.Thisadditiontotheprogrammereportedlycameaboutbecauseonehome implemented champions themselves. The PROSPER team felt this was a good idea andencouraged this approach in other care homes,making the role officially recognised and runningspecificstudydaysforthesepeople.Otherstrategiesthatcarehomesusedtoengageawiderrangeofteammembersincluded:

• discussingideasforchangeregularlyinteammeetings;• displaying information about PROSPER and resident safety in staff rooms, entry foyers or

residentloungestoencouragediscussionbetweenstaffandwithresidentsandfamilies;• havingadisplayboardwherestaff,relativesandresidentscanpostcomments;• asking everymember of the team to be part of solutions, such as encouraging domestic

workers,cooksandcarersatdifferentlevelstohelpresidentsdrinkmore;and• havingasuggestionsboxinthestaffroomorpublicareas.

“Wecanalwaysdothingsbetter.[PROSPER]challengesmanagersandcarerstodothings

better.Itputstheballinstaff’scourt.OneofthethingsItookfromthestudydaysandput

intoactionstraightawaywastoputanideasboxupinthestaffroom.Withthissortofthing

youmuststartatthegrassroots.Carersneedtoownitandsoeveryonemustunderstandit.”

[Managerofmedium-sizedhomerunbycharity]

Asmallnumberofcarehomes(fewerthan5%)reportedmoreengagementwithfamiliesasaresultofopenlypromotingtheirsuccesses.

“PROSPERispromotedintheloungesopeopleaskaboutit,sowetellfamiliesaboutit.Itisa

wayofopeningupthefloorfortalkingaboutthings.Atnextrelatives’meetingwewilldiscuss

PROSPERtoo.”

[Managerofmedium-sizedhome,runbycorporation]

Asmallnumberofhomes(fewerthan5%)saidthatdisplayingPROSPERmaterialshelpedthemhavediscussionswithfamilymembersofresidents,sothatfamilywereseenmoreaspartoftheteam.

“Itisagoodwaytoopenupdiscussionswithresidentsandfamily.Forexamplewhymum

needsanewpairofslippers.”

[PROSPERchampionfromalargehome,runbyacorporation]

96

Manyparticipatingcarehomesarepartoflargergroupsorcorporations.Thesecarehomesreportedsharing the lessons they are learningwith other care homeswithin the company, thus helping torollouttheideasandtools.

“We’repartofasmallgroupofcarehomesbuttheothercarehomesarenotpartof

PROSPERbecausethey’reoutsidetheboundary.Wetellthemwhatwearedoingandshare

allthematerials.Wehavearegularmanagers’meetingsofallthecarehomesandweshow

themwhatwearedoingandgiveoutcopiesofeverything.Theytakeitonboardanduseit

too.”

[Managerofasmallhome,runbycharity]

“[Corporation]isreallybigsoifwefindsomethingthatworkswecascadethattoallthecare

homes.Wecandoitreallyquick.Likewetestedanideahereandthenreportedittoour

regionalmanager.Shetoldalltheothercarehomesandnowwe’realldoingit.Whenwefind

somethingthatworks,wetellallourothercarehomes.”

[Deputymanagerofalargehome,runbycorporation]

Sharingwithothercarehomeswasalsothoughttobeparticularlyusefulforexpandingtherangeofideasandengagement.Thoughcarehomesarebusinesses,homemanagersand,whereavailable,proprietorsand regionalmanagersdidnot seem to feel that commercial sensitivity stopped themfromsharingideaswithothers.Althougheachcarehomewasstrivingtobe‘thebest’,beingpartofPROSPER seemed to help care homes share ideas to this end, rather than keeping things tothemselvesforcommercialadvantage.

“We’vefounditreallyusefultolistentofeedbackfromotherhomes.Forexample,wefound

outwehadalocalfallspreventionteam.Shethendidassessmentsonourresidentsandwe

madechangesinthecareplans.Wewouldn’thaveknownaboutthatserviceifotherhomes

didn’tmentionit.”

[Seniorcarerfromsmallhome]

ProactivefocusonpreventionAnotherchangeincultureisevidencedbycarehomesreportingthinkingmorelaterallyaboutreducingrisks.Ratherthanbeingreactive,therewasmoreofafocusonproactiveprevention,asreportedbyabouttwo-thirdsofcarehomes.

“Sometimeswethoughtfallswereunavoidable,butthischangedhowwethoughtabout

things.Atfirstwestartedofftryingtoreducefallsatnights.Thatworkedbutthenitincreased

thenumbersoffallsonthelateshift,sonowwe’relookingatdoingPROSPERthroughoutthe

wholecarehome.Itmadeusseewecouldpreventthingsinsteadofjustcopingwhenthings

happened.Wehavesomepersistentfallersthatinterventionsaren’tworkingwith,butthe

changeswemadehavepreventedsomeresidentsfromfalling.Staffaremorevigilantthan

theywerebefore.Theyareactivelylookingoutforwaystopreventfalls.AtstaffmeetingsI

gavetheminfoandputupnoticesandpostersintheoffice.It’slikewe’reallonhighalert

now.”

[Carerfromalargehome,runbyacorporation]

97

“Ithasreallyhelpedmeasamanagerkeepafocusonareasthatwemightnotfocuson.I

foundithelpedyouthinkaboutotherthingsthatyouwouldn’thaveconsideredlikefootwear,

distancebetweentablesanddrinks.Youmightnotthinkofthosethingsasaffectingfalls,butit

helpedmethinkoutsidetheboxabit.”

[Managerfromasmallhome,ownedbyasmallgroup]

Aboutone-thirdofmanagersandseniorcarerssaidPROSPERhelpedthemfeelmoreconfidentthattheyweredoingthebesttheycouldtosupportresidents.Althoughtherewassomefrustrationthatchangesmaynotalwayshappenasquicklyas theywouldwish, therewasasensethatcarehometeamswereworkingtogethertominimiseriskandpreventincidents.

“Ihaveconfidencenowthatwearereallydoingthebestwecaninthecircumstances.Wecan

evidenceittoo,showingwhatwearedoing.Itgivesmepeaceofmindtoknowpeopleare

beingbettertrained.Ithaschangedhowwethinkaboutthings.”

[Manageratlargehome]

EmbeddingtoolsandtechniquesAbout10%ofcarehomesreportedextendingtheuseofimprovementtoolsintootherareas,ratherthanfocusingonlyonthefalls,urinarytractinfectionsandpressureulcerpriorityareasofPROSPER.The tools most commonly adapted for other uses in care homes were the Safety Cross, graphsmappingincidentratesandmakingquicktestsofchange.

“Nowwe’reingrainedinit,wewillkeepdoingit.Wecollectdataforourselvesnow.Weare

usinggraphsnowforotherthingsbutIwouldneverhavethoughtofthatbefore.Itneedsto

bemadepartofcompanypolicyorwayofworking.Maybewhenpeoplesignuptotakepart,

itshouldbeatacompanylevelsothecompanyundertakestoworkinthatwayandkeepit

goinginfuture.”

[Managerfromamedium-sizedhome,runbyacorporation]

“Weareusingthemethodologynowforotherthingstooandapplyingittootherareas.Likewe

usedthevaluesideasinameetingwithstafftotrytoimprovecommunication.”

[Seniorcarer,smallhomerunbycharity]

Onecarehome reported that they liked the ideaofexamining safety culturebutdidnot find theMaPSaF tool user-friendly. Therefore they devised their own staff survey and plan to repeat thisovertimetomonitorchangesinstaffsatisfactionandsafetyculture.

Fivecarehomesreportedthattheyhadbeeninvitedtospeakateventsorworkcollaborativelywithothercarehomesororganisations,andtheysawthisasabenefitofbeinginvolvedinPROSPER.

“PROSPERshouldkeepgoing.Itwillchangeculturegradually.Itishelpingtoeducate

staffmoreandwithworkingtogetheracrosshealthandsocialcare.Thecompany

recognisesPROSPERassomethingusefulandwegetgoodpublicityfromtakingpart.It

hasopenedupdoorsforusintheactivitieswedowithotherhomes.Ithaslotsofknock

oneffects.”

[Managerofmedium-sizedhome,runbylargecorporation]

98

OtherviewsNotallcarehomesreportedchangesasaresultoftakingpartinPROSPER.Aboutone-thirdofcarehomesstatedthattheydidnotseeanychangesasaresultoftakingpartintheprogrammeorthattheywouldhavemadechangeswithoutPROSPER.Thesecarehomes tended tohavebeenvisitedlessfrequentlybythePROSPERimplementationteam.Theywerealsolargelyfromcohortoneandtwo.Thismaybebecausecohortthreeandfourhomeswerestillengagedintheprogrammewheninterviewedwhereascontacthaddroppedoffwithmanycohortoneandtwohomes.Itmayalsobea functionof theprogrammebeing strengthenedover time.Cohortonehomes inparticularwereaskedtousetoolssuchastheNHSSafetyThermometerandMaPSaF,whichtheyfounddifficultandoflimitedvalue.

“Wearen’tdoinganythingdifferently.Wellweareusingthecomputerforthenumbersand

wefilledinthesurvey,butthathasnotchangedwhatwedo…Wealwayswantedtomake

theriskssmallforourpeople.Wealwaystriednewthings.Nothinghaschangedintheway

wethinkaboutthingsbutitisgoodtohavethetraining.”

[Managerofsmallhome,runbysmallorganisation]

“Idon’tthinkPROSPERhelpedus.Wehadalmostnocontact.Theyhaverungusonceinthe

lastsix-monthsandnotvisited.Itseemslikewearealwaysgivingtothem;givingthemour

dataandourideasofhowwe’reimprovingandwe’renotgettinganythingback.Wehaven’t

changedhowwethinkordoanythingbecauseofthembecausetheyhavenotbeenintouch

withus.”

[Managerofalargehome,runbylargecorporation]

99

Appendix5.Impactsonsafetyprocesses

This appendixprovides furtherdetails aboutpotential impacts fromPROSPERon safetyprocessesand behaviours in care homes. Information was collected for this section using the followingmethods:


This appendix repeats some of the concepts highlighted in Appendix 4 regarding safety culturebecausebehavioursarerelatedtosafetycultureandattitudes,sothechangesareinterlinked.

TestingwaystoreduceharmsApartfromcarehomesthathadstartedtheinitiativewithinthepasttwo-months,mostcarehomestakingpart inPROSPERsaid theyhad triedanactivity to reduceharms.Overall,betweenhalfandtwo-thirds of care homes interviewed at any one time said they had tested newways to reduceharms.

“Weweretalkingabouthowwecouldadapttheredtraysusedinhospital,youknowwhere

peoplewithredtrayshavetobegivenmoredrinksoracertaintypeoffoodorwhatever.Well

thenwestartedusingreddoiliesforhydration,toremindustogivethosepeoplemoredrinks.

We’restartingtorolloutthatideathroughthewholehome.Butwedidn’twantitto

stigmatisesomepeoplesoweusecoloureddoiliesbutwegiveeveryoneadoily,notjustsome

residents.Redisstillusedforextradrinks,butothercoloursareusedtoosootherresidents

aren’tleftout.”

[Managerofmedium-sizedhome,runbyalargecorporation]

“Wehada‘comedinewithme’competitionaspartofnutritionweek.Residentswentand

visitedotherhomes.Fivehomestookpart.Weallgotgoodpublicity,whichispriceless.This

wasuniquebecauseitinvolvedfivedifferentcorporatecompaniesthatcametogetherfor

residents.Weputasideseeingeachotherascompetitorsandjustfocusedondoing

somethingfunandhelpfulforresidents.”

[Managerofmedium-sizedhome,runbyalargecorporation]

BoxA1listsexamples:

100

BoxA1:ExamplesofchangestoimprovesafetytestedbycarehomestakingpartinPROSPER

Staff

• assigningsomestaffmembersas‘champions’toremindotherstaffabouttipsandkeepharmreductionattheforefrontofpeople’sminds;

• having‘jellychampions’toencourageresidentstohydratemorethrougheatingjelly;• takingpartintrainingaboutreducingharmssuchasfallspreventionorinfectioncontrol;and• incorporatingnewideasintomovingandhandlingtraining;

Eventsandactivities

• runningan‘AliceinWonderlandteaparty’topromoteliquidintakewhichbecameacommunityevent,withparticipantsfromothercarehomesandotherprofessionalsinvited;

• ‘HumptyDumptyday’topromotefallsprevention;• undertakingactivitiesaspartofnationalcampaignssuchasdignityawarenessweekand

hydrationweek;and• providingactivitiestopromotegreaterstrengthandflexibilitysuchaschairlinedancing,

chairtaporchairballeteveryweek;

Tools

• usingcompactmirrorstohelpcarerscheckforpressureulcers,includingonpeople’sheels;• usingcoloureddrinkcoastersorcupstohelpidentifypeoplewhomayneedtodrinkmoreso

staffandrelativesareremindedtoofferpeopledrinksmorefrequently;• usingchartsanddisplayboardstocreatefriendlycompetitionbetweenunitsorwards;• markingtimeofday,residentnameandotherdetailsontheSafetyCrosstohelpidentify

trendsovertime;• usingskininspectionsheets,fall’schecklists,commentsboardswithhelpfultipsandother

aidstoremindstaffaboutgoodpractice;and• havingprizesawardedbytheparentcompanyfor100-dayseventfree.

UsinginformationforimprovementAsoutlinedinthesectionaboutsafetyculture,usinginformationforimprovementwasasignificantchangeinmanycarehomes.Carehomesreportedusingmoretoolstolookatdataaboutincidentsandspendingmoretimeanalysingtheinformationandplanningactionsasaresult.

“Weweretrackingthingsbeforebutwerenotusingtheinformationtoreallydosomethingto

getbetteroutcomes.Wehadinformationdatingbackfive-yearsbutneverdidsimplethings

withitliketellrelativeswheretobuygoodslippers.Abigchangenowisthatwearenotjust

collectingnumbers,weareusingthemtomakethingsbetter.”

[Managerofmedium-sizedhome,runbylargecorporation]

101

“Themonthlychartsmakeadifferencebecauseitshowsthetrendsandwherewearegoing

sowetakemoretimetoreflectonthings.Wedoauditsanywaybutthishelpsustiethings

together.Forexample,isfallsrelatedtoUTIs?Wesawthatthenumberoffallscoincidedwith

thenumberofurinaryinfectionsowearelookingatthereasonsbehindthingsnowbecause

ofthosemonthlygraphs.”

[Carerfromsmallhome,runbysmallorganisation]

Carehomessaidtheyfounditusefultodisplayinformationvisually,suchasusingtheSafetyCross,MonthlyMappinggraphs,posterswithinformationaboutreducingharmorotherdisplays.

For example, one home created a visual display with the Safety Cross and their falls reductionchecklist in thedining room. Staff said that this helped to remind themabout key actions. It alsomeant some relatives got involved,particularly in termsofbuyingdifferent typesof slippers. Thishome used the Safety Cross to record the number of falls each day and the time of day theyoccurred.Thishelpedtheteamunderstandwhenfallsweremostlikelytohappenandtakestepstoaddressthem.Thestaffnoticedthatfallsoftenoccurredimmediatelyafterbreakfastsokitchenstaffstartedtospendmoretimewithresidentsoncebreakfasthadbeenserved.Thehomereportedthatthishelpedtoreducethenumberoffallsatthistimeofday.Thiswassupportedbyareviewofthefallsrateskeptbythehome.

Overall, carehomes reported recording thingsmore robustly as a result ofwhat they learnt fromPROPSER. A number noted that their improved record keepingmight be useful when presentingtheirworktoregulatorsandcommissioners.

“Carersaremorevigilantaboutcheckingpressureulcersandrecordingwhatwearedoing.

Wecanprovewearedoingit.WithCQCinspectionsthiswillbegood,sotheycanseewhat

we’redoing.Beforewemighthavedonethingshereandtherebutweneverrecordeditwell.

Nowwekeepgoodrecordsofeverythingandlookbackonitanduseittoplanmore.”

[Seniorcareratmedium-sizedhome,runbycorporation]

“Wewillcontinuetodothemonthlydatatableandadaptittoincludemedicationsaswell

(likehowmanypeopleareonantibioticsorsupplements).Andwewillkeepdoingbodymaps

forpressureulcers.Thatallhelpswithouraccountabilitywhenwelookatwhoarethegood

carersandwecanalsoprovetoinspectorsthatwehavebeendoingtheworknow.”

[Managerofsmallhome,runbysmallorganisation]

TheCQCinspectionreportsforanumberofcarehomesmentionedPROSPERfavourably.

However on some occasions, care homes reported proudly displaying their PROSPER informationand recordsof testsof change toCQC inspectorswhowereperceivednot tobe interested– andwhodidnotcommentoninvolvementinPROSPERinthehome’sinspectionreports.

102

Appendix6.Resultsfromthecarehomeswithbothpre-andpost-datafor6-and12-months

___________________________________________________________________________Sevencarehomesfromcohortone,eightcarehomesfromcohorttwoandthreecarehomesfromcohortthreehadpre-andpost-datafor6-to12-monthsineithersideoftheinterventionstartdate.TableB.1showsthenumberofeventsandrateforeachoutcome.TheresultsofthesixoutcomesbycohortareshowninFiguresB1-B6.TableB.1.Numberofeventsandeventratesamongthecarehomeswithbothpre-andpost-data

Noofevents Noofresidents Rateofevents pvalue

Falls

pre 1841 8734 21.1% <0.01

post 2117 9070 23.3%

Pressureulcers

pre 464 8383 5.5% 0.49

post 501 8669 5.8%

UTIs

pre 309 8304 3.7% <0.01

post 412 8669 4.8%

Hospitaladmissions

pre 170 8375 2.0% <0.01

post 258 8502 3.0%

A&Eattendances

pre 175 8323 2.1% <0.01

post 290 8502 3.4%

Hospitaladmissionsduetoafall

pre 74 8925 0.8% 0.07

post 97 8875 1.1%

103

FigureB1:RateoffallsinthePROSPERcarehomes

FigureB1.1Incidentratesbycohortbeforeandafterintervention

FigureB1.2Runchartofchangesinincidentsbycohortovertime

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

104

FigureB1.3Runchartoftotalchangesinincidentsacrossallcohorts

FigureB2:RateofpressureulcersinthePROSPERcarehomes


July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

105



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

106

FigureB3:RateofurinarytractinfectionsinthePROSPERcarehomes



July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

107


FigureB4:RateofanyhospitaladmissionsinthePROSPERcarehomes


July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

108



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

109

FigureB5RateofA&EattendancesinPROSPERcarehomes



July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

110


FigureB6:RateofhospitaladmissionsduetoafallinPROSPERcarehomes


July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

111



July14 Jan15 Jul15

July14 Jan15 Jul15

Incide

ntra

tes

Incide

ntra

tes

112

Appendix7.PROSPERandUnitcosts

__________________________________________________________________________________

The cost of the PROSPER and the unit costs for hospital and resident outcomes are presented in

TablesC.1andC.2.Intotal,thecostsofthePROSPERwere£282,596.52.

In addition to the direct costs, we acknowledged the indirect costs which included the costs of

1)generalmanagementandsupportservicessuchasfinanceandhumanresources;2)informalcare;

and3)unpaidcarer’stimeandcosttothecouncilofcommissioningemptybedswhiletheresidents

areinhospital.Wealsoacknowledgedthepotentialindirectsavingsincluding1)numberofcall-outs

foralliedhealthprofessionals/pharmacists/social care;2) reductionof total staffworkload;and3)

reductionofmedicationprescriptions.Wehavenotattemptedtoquantifyanyofthese.

TableC.1:OriginalbudgetedcostsofthePROSPERproject Programmemanagement,QIMethodologytraining,supportandliaisonwithHealth.

• ProjectLeadx1FTE(Band4@£22,500+23%oncosts)x2.5years £73,522.50

• QIOfficerx2.5FTE(Band4@£28,500+23%oncosts) £89,932.50

3-monthsupportprogrammeperhomex80-100homesdeliveredthroughfourcohortsatquarterlyintervalsovera1-yearperiod.

• QIAssistantx2FTE(Band3@£22,212+23%oncosts) £54,641.52

QIMethodologytraininghalfdaysession-approx.5homespersession,10-20people

• Venue&supportingmaterials/toolkit@£750persessionx20sessions £15,000.00

LearningEventstomaintainmomentum

• CommunityofPracticeeventsx3(venue,catering&supportingmaterials)@£3000.00perevent

£9,000.00

• Prosperchampions’studydaysforcarestaffx8,2perquadrantarea(venue,catering&supportingmaterials)@£3000.00perevent

£24,000.00

Additionalresources

• Compactmirrors,doily’s,ferrules,championbadges,bags,lanyards £15,000.00

• CapitaRecruitmentcosts3posts(internalsecondments)@£500 £1,500.00

Total £282,596.52

113

TableC.2:UnitcostsforhospitalandresidentoutcomesOutcome Hospitalcosts Reference

Falls Costsperpersonfalling£1720-8600 (Craigetal,2013)27

Pressureulcers £1214(category1)to£14108(categoryIV). (Dealeyetal,2012)28

UrinaryTractInfections(UTIs)

£2359perpatient.

http://www.hospitaldr.co.uk/blogs/features/unplanned-admissions-medical-technology

Anyhospitaladmissions

Averagecostperepisode:Electiveinpatientstaysnationalaverage£3403Non-electiveinpatientstays(longstays)£2716Non-electiveInpatientstays(shortstays)£611

PSSRUCosts-2014(http://www.pssru.ac.uk/project-pages/unit-costs/2014/)

A&Eattendances

TheaveragecostofanA&Eattendanceis£114Theaveragecostofanoutpatientattendance£108Theaveragecostofanexcessbeddayis£237

ReferenceCosts(Depart.Health)2013https://www.gov.uk/government/publications/nhs-reference-costs-2013-to-2014

Communitycosts

Falls £115-197 TianYetal.201329,

Pressureulcers £43to£57Ulcerwithoutcomplications Costingstatement:PressureulcersImplementingtheNICEguidelineonpressureulcers(CG179)(NICE,2014),https://www.nice.org.uk/guidance/cg179/resources/costing-statement-248688109

UTIs £27.8-£29.0 Turneretal,201030

27CraigJ,MurrayA,MitchellS,ClarkS,SaundersL,BurleighL.ThehighcosttohealthandsocialcareofmanagingfallsinolderadultslivinginthecommunityinScotland.ScottMedJ.2013Nov;58(4):198-203.doi:10.1177/0036933013507848.28DealeyC,PosnettJ,WalkerA.ThecostofpressureulcersintheUnitedKingdom.JWoundCare.2012Jun;21(6):261-2,264,266.29TianY,ThompsonJ,BuckD.Exploringthesystem-widecostsoffallsinolderpeopleinTorbay,TheKingsFund2013http://www.kingsfund.org.uk/publications/exploring-system-wide-costs-falls-older-people-torbay30TurnerD,LittleP,RafteryJ,TurnerS,SmithH,RumsbyK,MulleeM;UTISgroup.Costeffectivenessofmanagementstrategiesforurinarytractinfections:resultsfromrandomisedcontrolledtrial.BMJ.2010Feb5;340:c346.doi:10.1136/bmj.c346.

114

Appendix8.Impactsonresidentoutcomes

ThisappendixprovidesfurtherdetailsaboutimpactsfromPROSPERonresidentoutcomes.

Table D.1 shows the impact of PROSPER on falls by care home characteristics. There were nostatistically significantdifferences forallof thecarehomecharacteristics, inpartdue to thesmallsamplesize.

TableD.1.ImpactofPROSPERonfallsbycarehomecharacteristics Noimpact

Post-rate-prerate>=0

Impact

Postrate-prerate<0

Numberofcarehomes 10 8

Mediannumberofresidents(interquartilerange) 39(35-53) 36(31-38)

Averageproportionoffemaleresidents 76% 73%

LocationMidEssexNorthEastEssexSouthEssexSouthWestEssexWestEssex

10%50%30%0%10%

0%75%0%0%25%

OwnershipOneprivateowner,e.g.familyOwnedbysmallgroup,e.g.twoorthreehomesOwnedbylargecorporation

10%30%60%

25%25%50%

HowmanyhomesownedintheUKbysameowner12-56-1011-2021+

0%40%0%10%50%

0%

37.5%12.5%12.5%37.5%

TypeofcareprovidedResidentialcareNursingcareBothresidentialandnursingcare

80%10%109%

71.4%0%

28.6%

Averageproportionofresidentsover80 78% 82%

Averageproportionofresidentswithdementia 51% 60%

Average%residentsfullyfundedbylocalauthority 43% 52%

Average%ofresidentsinhomeforoversix-months 74% 77%

EachresidenthastheirownGPOneGPpracticethatcoversthewholehomeAsingleGPpracticeformostresidentsbutsomeresidentshaveadifferentGP

50%20%30%

50%12.5%12.5%

115

TableD.2 shows theoutcome rates at the three timepoints i.e. pre-six-months; intervention andpost-six-monthsamong18carehomeswithbothpre-andpost-interventiondata.Againwedidnotobserveanyreductionsinfalls,pressureulcers,UTIs,allhospitaladmissions,A&Eattendancesandhospitaladmissionsduetoafall.

TableD.2.Ratesofthesixoutcomesinthepre-six,interventionandpost-six-months.

116

Appendix9.FeedbackaboutPROSPERcomponents

ThisappendixprovidesfurtherdetailsaboutcomponentsofPROSPER.Ithighlightsaspectsthatwerefoundtobemoreor lesseffective in facilitatingchange. Informationwascollectedforthissectionusingthefollowingmethods:


Trainingandfacilitation

TraininginqualityimprovementapproachesInitially PROSPER quality improvement training was offered by an external team made up ofimprovement specialists from healthcare. Care homes reported that this training was difficult tounderstand,notpracticalenoughandnottailoredtothecarehomes’context.Asaresult,PROSPERadapted the training and from cohort two onwards it was run by the PROSPER implementationteam.

Animportantchangetothetrainingwasthatitincorporatedexamplesfocusedoncarehomes.Thefacilitators alsoused ideas testedbyprevious cohorts, so they couldprovide real examples aboutwhatothercarehomestriedandwhattheoutcomeswere.Theterminologyused isalso linkedtothe care home context. For instance PDSA cycles are likened to risk assessments, whereby carehomes consider an issue, think about how to manage it, review progress and decide on furtheraction.Linkingtheconceptstothingscarehomesarefamiliarwithreportedlyworkedwell.

Another change was that training materials were redeveloped to be more user-friendly. Initiallyblankdriverdiagramworksheetswerepresentedandcarehomeswereexpectedtocreateadriverdiagramfortheirspecificaim.Howevercarehomesfoundthisdifficultandtime-consumingtodo.Therevisedtrainingprovidedsampledriverdiagrams,developedasapartnershipbetweenPROSPERandNHScommunityteams.

APDSAworksheettemplatewasdevelopedtohelpcarehomestaffseehowtoundertakerapidtestsof change in practice. These concrete examples and worksheets were well received and wereincludedinaresourcetoolkitwhichwasmadeavailableonline.

117

Howevermanycarehomes incohorts threeand foursaid theydidnotuse thedriverdiagramsorPDSA cycles.Thesecarehomesdid focuson testingnew ideas,butnotusing the formal structureandterminologyofqualityimprovementmethods.Thismaynotbenegative,butitsuggeststhatthetraining and support moved away from being about quality improvement methods towardssupporting improvement more generally. Although the implementation team noted that cohortstwo, three and four received the same training, but just withmore practical examples, feedbackfrom 203 interviewswith care homes and 12 discussion groups suggested that therewas a shift,fromcohort twoonwards, away fromemphasising improvementmethodologies in amore formalsense.ThismeansPROSPERmayhavebecomemoreabout improvingcare inageneral senseandlessaboutintroducingspecificimprovementmethodologies.

In terms of the training, all cohorts suggested that staff from care homes could co-facilitate theintroductorytraininginfuture.Thiswasastrongmessage,putforwardbyabouttwo-thirdsofcarehomes.

“Itwouldbebetteriftheyhadpeoplefromhomesrunningthetrainingorrunningitside-by-

sidewiththem.Ormaybesomeonefromthehospital,theCouncilandoneofus.Itmightmake

itmorerelevantandeasytounderstand.Imeanitwasrelevant,butifitwasamanageror

someonewhoworkedinahomebefore,welltheycangiveexamplesthatmakesenseanduse

wordslikewhatweuse.”

[Seniorcarerfromsmallhome,runbysmallgroup]

TrainingaboutreducingharmsInadditiontotrainingaboutqualityimprovementmethods,PROSPERworkedwiththeNHStooffertrainingaboutsubstantivetopicssuchasinfectioncontrol,diabetes,medicinesandpressureulcers.Someof this trainingwas runasworkshopswhere carehome representatives came together inacentralvenue.Atothertimes,NHSfacilitatorsvisitedcarehomesandoffered10-30minutetrainingsessions, perhaps repeated so different carers could comeoff the floor to attend. These in-homesessionswerewellreceived.

“PressuresoretrainingfromPROSPERwasgood.Theydiditin30minutesshiftswithtraining.

Likeevery30minutestheyrantrainingfor10minutesorsomething.Peoplecameinfromtheir

daysoff.Thattypeoftrainingshouldbecontinued.Itisreallygoodcomingtohomes.”

[Managerfrommedium-sizedhomerunbycorporation]

“Theshortsharptrainingofferedtocarershasbeengood.Carershavespottedthatdistrict

nursesaren’tdoingwhattheyshouldnowbecausethey’vehadthattraining.”

[Carerfromsmallhome,runbysmallorganisation]

118

Theseadditionalstudydaysandin-hometrainingwerenotpartoftheoriginalPROSPERprogrammedesign. They were well received but meant that PROSPER became about more than introducingquality improvementmethodologiesandtoolstocarehomes. Infact in interviewsconductednearthe endof theprogramme,homemanagers and carers constantly referred to this additional freetrainingasbeingthemaincomponentofPROSPERandthethingtheyvaluedmost.InthefinalsetofinterviewsnotonehomementionedPDSAcycles.Offeringtrainingaboutdiabetes,medicinesandsoonwasvaluableforcarehomes,butdilutedthefocusonqualityimprovementapproaches.

The rationale for introducing trainingabout substantive topicswas that carehomesneeded ideasaboutwhat improvementstomake,ratherthansolelyhowto introduceandmonitorchange.Thisrelates to trainingaboutpressureulcers, fallsandurinary tract infectionsas thesewere the threeprimaryfocusareasofPROSPER.Hydrationisariskfactorforallthreetopics,sothiswasapriorityfortraining.Othertopicssuchasdiabeteswerepartlychosenduetoavailabilityand interest fromtheNHS;thoughcarehomesalsosaidtheyfoundthisuseful,eveniftheydidnotrequestit.

Thistrainingreceivedmuchmoreofafocusthanothercomponentsthatwereoriginallyplannedtobeprioritised,suchasusingasafetyculturetoolandshowingcarehomeshowtotrackchangesintheirincidentratesovermanymonths.Atoolwasusedforthisbutitwasnotemphasisedasaday-to-day improvementtoolbythefacilitationteam.Thustheadditionalsubstantivetrainingappearstohaveovershadowedtheoriginal intentoftheprogramme–totestthefeasibilityof introducingqualityimprovementmethodstocarehomes.ThispresumablyreflectsthedifferentcomponentsofDeming’sclassic‘profoundknowledge’–specialistcontentknowledgeaboutthetechnicalelementsof any system being improved needs to be present alongside general knowledge about how toimprove.

There was also a perception from some care homes that the training was not being rolled outuniversally across Essex, with PROSPER care homes more likely to benefit than others. This issomewhatoutsideofthecontrolofPROSPER,asthetrainingisofferedbyNHSpartners.TherearefiveCCGswithintheEssexCountyCouncilboundaryandfourcommunityhealthproviderswhohaveallofferedslightlydifferentthingstoPROSPER,meaningthereisalackofconsistencyacrossareas.Howeverthisreinforcestheimportanceofbeingcarefulaboutwhatispromisedtocarehomesandthe need to follow-up regularly with partners to ensure that care homes in all areas are able tobenefitfromadditionaltrainingtoavoidfeelingsofunfairness.OriginallyPROSPERplannedtofocuson twoareas inEssexonly.Expandingbeyond thishasbenefits fordissemination,butalsomeansthatmoreeffortneedstobeputintomanaginginconsistenciesinprovision.

HomevisitsThePROSPERimplementationteamprovidedon-goingsupportvisitsforaboutsix-months.Initiallyitwas envisioned that visits would be monthly, but there was variation in this. Some care homesreceived significantlymore visits thanothers. The rangewas zero to five visitswithin a six-monthperiod. Most homes received two or three visits in the first six-months, with some care homesreceivingfollow-upafterthisandothercarehomesreportinglittlefurthercontact.

Carehomes that received fourormorevisitswerehappywith the levelof contact.Homesvisitedless often called for more frequent follow-ups. Homes that had ended their ‘intensive’ first six

119

months also said they felt left on their own, with little communication about whether visits hadstoppedaltogether.

“Yes,letmeremember.She[officer]cametwo-monthsago,nothree.Yes,three-months.We

canringherifwewant.Isupposetherehasnotbeenthatmuchcontact.Wehavehadtwo

visitsoverthespaceofsix-months.You[evaluator]comemoreoften[laughs].”

[Managerfromsmallhome,runbyasmallgroup]

“Theypromisedtocomeeverymonthbuttheyareonlycomingeverythree-months.Itcouldbe

abitmoreregularifyouaskme.Idon’twanttosoundungratefulbutitdoesfeellikeweare

justlefttogetonwithit.Ithinkitdoesimpactonhowfastwegobecauseweareflounderinga

bitandwithoutsomeonetokeepitinyourminditisabithard.”[Managerofamedium-sizedhome,runbyacorporation]

Incohortfour,twostaffwerehiredsolelyforPROSPERandcouldvisithomesregularly.Cohortfourhomes reported being satisfiedwith the level of contact from the PROSPER team,muchmore sothaninitialcohorts.

Duringevaluationteamvisitstocarehomesandinterviews,homemanagersandstaffgavepositivefeedback and widespread praise about the friendliness, enthusiasm and approachability of thePROSPERimplementationteam.Somecarehomessaidthattheyinitiallybegantheprogrammeandhavecontinuedwithitduetothepersonablestyleoftheimprovementfacilitators.

“ThethingthathasmademestickwithitisIlike[facilitator].Wearebusy.Wehavealoton.

Wearedoingwellalready.Sheisgoodthough.Shekeepsyouonboardandmakesyouwantto

takepart.”[Managerofalargehome,runbyalargecorporation]

“Ilike[facilitator].Sheisnice.SheisnotthathelpfulandIthinksheknowsonlyasmuchasus,

butsheisveryunderstandingandwillhaveachatwithyouandthat.It’sthepersonalthingeh,

itmakesadifference.Shecomesouteverytwo-orthree-months.”[Managerofamedium-sizedhome,runbyacorporation]

There was variation in the type of support provided to care homes, with up to seven differentfacilitators in the teamatanyonetime (including theproject lead).Eachcarehomewasassignedonefacilitator,andthefacilitatorshadvaryinglevelsofknowledge,stylesandtimeavailable.Somecarehomes felt that theyhada facilitatorwhowasknowledgeablewhereasothers thought therewasscopeforupskillingthefacilitator.

“Igotthesensethattheyarelearningastheygoalongtoo.TheyaredoingtheirownPDSA

cycles[laughs].That’sallfineanddandy,butiftheywantustogetwiththeprogrammeina

shortspaceoftimeandmakeallthesechanges,itwouldhelpiftheycouldknowmoreabout

howtousetheThermometer,howtodoPDSAlikewiththosediagrams.Idon’tknow,Ijust

thinkifthey,Imeanwe’regratefulforanyhelp,don’tgetmewrong,butIdidn’treallyfeel

helped.Okwait,hereisabetterwayofsayingit.Ifeltsupportedbecausethepeoplearenice.

Theyaretryingtheirbest.ButIdidn’tfeelhelpedassuchbecausetheydonotseemtobe

experts.Theydonotseemtoknowwhattheywantustodo.Theydonotknowabouthowto

120

havelessfallsorpressuresores.Wealreadyknowwehavesores.Wedon’tneedhelptofind

thatout.Weneedhelptoknowhowtostopit.Theydon’tknow.”

[Managerofalargehome,runbyalargecorporation]

Towardstheendof2015,thePROSPERteamintroducedtwoband3supporterswhotookonakeyfacilitation role with the care homes. These team members did not have a lot of training orexperience in quality improvement and were also working at a lower banding than other teammembers.However,theseteammemberswereverywellregardedbythecarehomes.Carehomessaidtheyfeltthatthesefacilitatorswerefriendly,helpfuland‘onourlevel’.

CommunityofpracticemeetingsPROSPERrancommunityofpracticeeventsandstudydaysforPROSPERchampionstogivehomesanopportunitytocometogethertoshareideas.

ThefirstcommunityofpracticeeventwasheldinNovember2014forcohortone.Thesecondeventwas held in May 2015 for care homes from both cohorts one and two. In late 2015, smallercommunityofpracticeeventswere run in thenorth,eastandwestofEssex.Carehomes fromallthreecohortsoperatingatthetimewereinvited.

Thefirstcommunityofpracticeeventcombinedtrainingwithsmallgroupwork.Whilstcarehomesappreciated the event, they suggested that more time could have been spent networking andallowing them to share their successes. The PROSPER team took this on board for future events,focusingmoreonsharingideasbetweencarehomes.

Stakeholders from the local authority andNHSwere present at the first two large community ofpractice events. A member of the implementation team facilitated the smaller community ofpracticeeventsandtherewerenodelegatesapartfromcarehomestaff.

Whereas carehomemanagerspredominatedat the first event, the second communityofpracticeevent had participation from carers, managers and deputies. In addition, there were regionalmanagers representing large corporations,UCLP, aGP,NHS community teammembers and somemembersofthePROSPERimplementationteam.About40peopletookpart.

Having carers at the event helped to showcase that the philosophy underpinning PROSPER wasbeginningtospreadinsomecarehomes.CarerssaidthatoneofthekeyoutcomesofPROSPERwasbuildingacadreofenthusedandempoweredcarers.

“Itiseducatinguscarersaboutwhattolookfortoavoidpressureulcersandfallsorto

identifyUTIs(urinarytractinfections).ThemainthingthatPROSPERisdoingisempowering

staff.Wehavehadafairbitoftrainingandresourcesbeforebutwhatmakesthisdifferentis

thatitisaboutus.Beforewe(carers)wouldneverreallygettocometoaneventlikethiswith

allthemanagersandotherpeopleandbetreatedlikewehavesomethingtosay.Whenwe

tellaboutwhatwedid,peoplelistenandthatmakesusproud.Havingmoreeventslikethis

wouldbegood,especiallyaimedathelpingcarersunderstandevenmore.”

[Carerfromamediumsizedhome,runbyacorporation]

121

“Beforewefeltabitintimidated,youknow,becauseI’mnotanurseorasenior(carer)or

anything.IwasworriedIdidn’tknowenough.ButsincebeingaPROSPERchampionmy

confidencehasgrewmore.Carersaren’tjustcomingintowashandcleannow.Wecando

otherthings.Wearehelpingtoimprovethings.Weareworkingasateamnow…Comingto

somethinglikethis(communityofpractice)justshowsthatwecantakepart.Itisscarytobe

inaroomlikethiswitheveryone,butthenwhenothercarersstarttalkingyoufeel,yeah

(punchestowardsceiling),likewecanmakeadifference.Wegotlotsofcoolideastotake

backandtryout.”


Atthiseventwasacall forsomespecificactivities targetingcarers–bothtospreadthewordaboutquality improvementmethods, toprovidepractical skills inmotivatingothers tochangeandtogainfurther informationaboutsubstantivewaystoreducefalls,urinarytract infectionsand pressure ulcers. The implementation team took this suggestion on board and set up‘champions’ study days’ targeting carers. These events were very well regarded, withparticipantssayingtheyfeltmotivatedasaresult.Carersroutinelysaidtheyhadimplementedthingstheylearntatthechampions’studydays.

“Wecameawaybuzzingfromthat[champions’studyday].Itwasgoodbecauseweheard

whatothercarehomesaredoingandwefeltgoodaboutwhatwe’redoingtoo.Therewas

lotsoffunandlaughter,butwewerelearningaboutseriousthings.Itwasallonourlevel.Not

toohighandmighty.Itwasrealpractical.Thingstouseeveryday.Itoldalltheothergirls

whenIgotbackandwearetryingthingslikedecorating[walking]frames.”


Smallercommunitiesofpracticeraninlocalareaswerenotalwayswellattended,butwerewellreceivedbythosewhodidattend.Thesetendedtobefacilitatedbytheband4supportworkersandtofocusonhelpingcarehomessharetheirideasandsuccessesinaninformalenvironment.Whilstpeopleappreciated someof the ideas shared, they seemed toprefer the largereventswheretherewasmoreofa‘buzz’.

Key things that helpedmake PROSPER community of practice events and champions’ study daysusefulincluded:

• including a mix of taught content and opportunities for care homes to share theirlearningandideas;

• havingallCouncil improvementofficerswhoaresupportingcarehomeswithPROSPERattendingtheeventssocarehomeshaveanopportunitytoseetheirassignedofficeronoccasionsotherthanhomevisits;

• using introductory ice-breaker activities and fun practical activities to createengagementattheevents;

• usingextensiveexamplesfromcarehomesandsimpleterminology.Jointfacilitationbythe implementation team and care home staff was requested by homes to promote

122

more ‘ownership’ andparticipation and thiswas implemented in someof the smallerlocalisedcommunityofpracticemeetings;

• focusingonbuildinga‘communityofpractice’ratherthanrunningstandardeducationalevents/workshops;

• invitingawiderangeofcarehomestaff,ratherthanonlyseniorstaff;

• providinghand-outsandresourcesforparticipantstotakeawaywiththem;and

• providingenoughnoticeofevents.

123

ToolstosupportusingdataInaddition toprovidingvarious typesof trainingandsupport,PROSPERalsoprovidedcarehomeswithtoolstosupporttheiruseofdata.Avarietyoftoolsweretested.ThesedatatrackingtoolswerethesecondmostpopularandvaluedaspectofPROSPER,afterthesubstantivetraining.

NHSSafetyThermometerThefirsttoolthatPROSPERtestedtohelpcarehomesmonitortrendsinsafety incidentratesovertimewastheNHSSafetyThermometer.Appendix3containsacopyofascreenshotfromthisonlinetool.Thedevelopersofthetoolhadstatedthatitwasbeingusedin402carehomesandrequirednofurther adaptation for this setting. This was one reasonwhy it was included in the original workprogramme.

Carehomeswereasked to log into the tooleachmonth toprovide incident ratesovera set timeperiod,e.g.72hourspermonth.Thistoolwastestedwithcohortoneandtwo,butwasthenphasedoutduetofeedbackfromthecarehomes.

Whilstcarehomescouldseethevalueofusinganonlinetooltotrackprogress,theywerehinderedbytechnicalissueswithsettingupandusingthesystem,concernsaboutwhethertheinvestmentoftimeneededtoinputdatawasworththeoutputandqueriesaboutthevalidityoftheinformation.

“Ireallydon’tlikeit.ImeanIlikeitfinewhenyougetonanditissimpleenoughtousebutI

thinkitisawasteoftime.Thewholeprocessforsettinguptakesages.Youhavetoemailthen

getanemailbackfromtheguythenanotheremailwithpasswordsandtheneventuallythey

setyouup.Itseemsalong-windedthing.Itsometimesisdownornotworking.Itjustseemsa

fiddlytechnicalthingthatisnotworthit.”[Managerfromamedium-sizedhome,runbyasmallgroup]

Inparticular,carehomesreportedthattheoutputgainedwasnotworththeinvestmentoftimetoinputinformationabouteachindividualresident.

“Wecollectdata[aboutincidents]anywayandhaveitforthewholehomeforthewhole

monthoryear.Idon’tseeanyaddedvalueinlookingatjustone-dayor72-hoursorwhateverit

isandthencountingepisodesforthat.Thetoolsshouldbeadjustedfortheneedsofcare

homes,notjustusingsomethingofftheshelfthatdoesnothingforus.Maybeforasmallhome

thatdoesn’tusuallycollectdatathenitwouldbeagoodintroductorytool,butwe’reata

differentstagetothat.Itfeelslikewearegoingbackwardsandjustmakingmoreworkforthe

sakeofit.”

[Managerofalargehome,runbyalargecorporation]

“Ittakesalotoftime…yesitisonlyonceeveryfortnightbutyouhavetoenterintheresidents

everytime.Soyouhavetoputin[number]individualdetails,thensayifeachonehadafallor

soreorwhatever.Idon’tevenknowwhatwegetoutofit.Isuppose[Council]needitfortheir

recordsorsomething.”[Administrativestaffmemberfromamedium-sizedhome,runbyacorporation]

124

Anotherissuewasthatcarehomesdidnotnecessarilyfeelthattheyreceivedsupportinhowtousetheinformation.

“Atthemomentitseemslikewearecollectingallthisdata,butthennotdoinganythingwith

it.Idon’tknowwhattheywantustodo.Imean,Iknowwehavetoputindataeverytwo

weeks,butthenwhat?Itseemsabitpointless.Noonehasexplainedanything.It’sabit

frustratingtobehonest.”

[Managerfromasmallhome,runbyasmallgroup]

SomecarehomesqueriedthevalidityoftheNHSSafetyThermometermethodology,inparticularthe approach of using samples rather than census data. In care homes there may be limitedturnover of residents, a relatively small number of residents and of incidents. TheNHS SafetyThermometertookasampleofafewdayspermonthratherthanacensusofall incidents.Onlyonefallorpressureulcerisenteredperperson.Thismeansthatmaybedifficulttodemonstratechangesovertime.

“Idon’tthinkitisvalid.Whydoasample?Tomesamplingisopentobiasandselectingwhat

youwantthenumberstotellyou.Idon’tthinkwewillgetanythingoutofitbydoingthat.We

aredoingitbutkeepingarecordforalldays,notjustoneortwo-daysamonth.”

[Managerofasmallhome,runbyasmallorganisation]

“Idon’tthinkitisgoingtotellusanything.Wejustputintheworst-casepressureulcer,notall

ofthem.Weonlylookat72hours.Ijustdon’tgetwhatitissupposedtotellus.Itisbiased.Itis

nothelping.Itisjustmakingmorework.”[Administrativestaffmemberfromalargehome,runbyalargecorporation]

Thesecommentsperhapshelpstoexplainwhyalthoughtwo-thirdsofcarehomesfromcohortonesigneduptousetheNHSSafetyThermometer,actualusagewasinconsistentandsporadic.Lessthanone-quarterofthecohortonehomesinputdataregularlyandthesehomesgenerallydidnotusethechartsgeneratedbytheThermometerforimprovementpurposes.ThesamewastrueforcohorttwosoPROSPERdecidednot to use this tool from cohort threeonwards. Cohort one and twohomeswere still encouraged to use the Thermometer if they wished, but not one of these care homescontinuedinputtingtheirdataafterthe‘requirement’todosoended.

CareHomeshavesuggestedthattobemoreusefultheThermometercould:

• measure days between events rather than events per month (since the number ofeventsmaybelow);

• inputthetotalincidentsofevents,ratherthanonlyafewdayspermonth;• inputtotalsratherthanthedetailsofindividualresidents;• ifindividualresidentdetailsarerequired,carrytheseoverautomaticallyeachmonthto

avoidhavingtoretypeallinformation;and• disaggregateagegroupingstoallowtodifferentiatepeopleinolderagebrackets

125

SafetyCrossInresponsetotheissueswiththeNHSSafetyThermometer,PROSPERintroducedtheSafetyCross.This is a cross-shaped calendar listing days in the month. Staff colour in green to represent anincidentfreedayandredtorepresentthedateofanincident.Somecarehomesdividethedayintoshifts.Othersmarkoffthetimeofincidents.SomecarehomesusedtheSafetyCrossforonetypeofharm; others used three different Safety Crosses for falls, urinary tract infections and pressureulcers.

The Safety Cross was introduced by the UCLP team to one care homewho joined cohort one inSeptember2015attherecommendationofCQC.Thisapproachwasusedasanexampleatthefirstcommunity of practice in November 2015, with a photograph of the Safety Cross on the wall,surroundedbyafall’schecklist.Fromthatpoint,thePROSPERimplementationteammadecopiesofthetoolavailableforthecarehomestouseiftheywished.SeveralcarehomesbeganusingitfromthispointandPROSPERincludeditinthetoolkitandtraininggiventocohorttwo.Thetoolwasnotnecessarily designed as an alternative to the NHS Safety Thermometer, as it does not providelongitudinaldata.

TheSafety Crosswas extremelywell received. After the substantive training, thiswas the secondmosthighlyvaluedPROSPERintervention.Alimitationisthatitdoesnotallowtrackingofprogressformorethanone-monthatatime,butasignificantbenefitisthatitisavisualpromptandinvolvesavarietyofstaffratherthansolelymanagers.ManycarehomesreportedplacingtheSafetyCrossonthewallinaprominentarea,eitherinthestaffroomorpublicareas.Whendisplayedinpublicareas,thishasreportedlysparkeddiscussionswithrelatives insome instances. In largerhomes,differentunitsorfloorsmaycompetewitheachotherinafriendlymannertogetthe‘bestresults.’

Aboutone-thirdofhomesfromthefirstthreecohortsqueriedwhythediagramwasacrossshape.There was some initial discomfort about something that could be perceived to have religiousconnotations.ThePROSPERimplementationteamreportedthatonecarehometurnedthecrossonitssidesoitwasmorelikethegreencrosscodeforthisreason.Othersuseda‘walkingstick’versionorothershape.DuringevaluationinterviewsandvisitstherewererequestsforPROSPERtochangefromsupplyingcrossshapestoanothershapeinstead.

Asmallnumberofcarehomeshavesaidthatbeingsovisiblemayhavenegativeimpactsifpeoplefeeltheneednottodiscloseincidents,butoveralltheSafetyCrosswaspositivelyviewed.

126

GraphsmappingincidentratesTo help track changes in resident outcomes over time, care homes were invited to provide theevaluation team with data about monthly rates of falls, pressure ulcers, urinary tract infections,emergencydepartmentattendancesandhospitaladmissions.Thiswassupposed tobepartof theNHS Safety Thermometer intervention, but care homes were not using the tool in this way. Theevaluation teamthereforeproducedadata table,whichcarehomesused tocollate their incidentrates.Thiswasusedbysomecarehomesasaninterventiontool,sotheycouldtrackchangesovertime.

Due to issueswith theNHS Safety Thermometer, the evaluation team developed a simple onlineversionof thedata table toolwhichautomaticallygraphedhome’s resultsandcomparedwith theaverageofallcarehomestakingpartinPROSPER.ThePROSPERimplementationteamrolledoutthis‘MonthlyMapping’ tool fromcohort threeonwards.However the toolwasdesignedasa trackingdeviceratherthanafullimprovementtool.

Allbutonehomefromcohortone(94%),89%ofcohorttwo,62%ofcohortthree,and55%ofcohortfourusedthetool.Cohortoneandtwowereactivelyencouragedbytheevaluationteamtousethetool,includingwithinpersonandtelephonesupport,whereascohortthreeandfourreceivedemailpromptingonlyfromtheevaluationteam.

Homemanagers generally valued receiving graphs of their incident rates, either directly from theevaluationteamoronline.

“Thankyouforthegraphs.Itisveryhelpfultoseethedatainblackandwhite,ohhowIloathe

winter!It'sinterestingtonotethesimilarpatternsacrossthecounty,don'tyouthink?WhichI

believeisinpartduetoseasonaltrendsandviruses,etc.,andanescalationofUTIs,lackof

fluid,confusionandfalls,inourcaseanyway.Onecananticipatethetrendsalmosttotheday.

AllIneedtodonowistoreducetheodds!!!Yourevidenceisapowerfultooltotakebacktothe

teamandanincentivetodobetter.”

[Managerfrommedium-sizedhome,runbysmallorganisation]

“OurCQCinspectorhasappreciatedourboardwhereweputyourgraphs,thankyouforyour

help,graphsreallyshowtheprogressinthehome(allupsanddowns).Iamdoinganalysisof

reasonsoffallsandallincidents/accidentsinthehomenowtohaveaclearpicture(for

examplealadywithaverylimitedmemoryandunderstanding-veryadvanceddementia-

camebackfromhospitalafterthreebloodtransfusionsandbeingveryweakandunsettledhad

twofallswithintwohours(sheimmediatelywenttoone-to-onecareforfewdaysuntilshe

recovered)-buttherewasaclearreasonoffalls-Irecordeditandrecordallreasonsofall

accidents/incidentstodosomeanalysiseverymonth...itmighthelpaswell...carryonwith

graphsandideas....itreallyhelpsustoimproveourworkandalsohelpstochangethinkingof

staff-itisverytransparentandmakestafftostopandthink"why",notonlytoworklike

robots....:).”

[Managerfrommedium-sizedhome]

127

“Whenyouseeitallvisuallylikethisitpaintsaninterestingpicture.WhenIfilledinthetableit

wasmerenumbersbutthisbringsittolife.Ihavedistributedaroundthehome."

[Managerfrommedium-sizedhomerunbylargecorporation]

An important component is that the graphs allowed carehomes to compare themselveswith theaverageincidentrateofothercarehomestakingpartinPROSPER.

“Havingthemonthlychartshashelpedusdecidewhattofocusonandweseeweneedto

reducefallssowearefocusingonthat.Wearedoingbetterthanotherhomesinsomeother

areassofeelreassuredaboutthat.Ilikethecomparativenatureasweareabitcompetitive

andwewantourhometobegood.Thatreallymotivatesyou.”

[Managerofmedium-sizedhomerunbysmallorganisation]

"IwasparticularlypleasedtoseethatweareunderthePROSPERaverageforpressuresores

andfalls.Interesting.Ineedtocatchupwithlastmonth’sdatainputbutitdefinitelyshowsus

wheretofocusandwhattoworkon.Itmakesmemoreinterestedinlookingatthedata.Thank

youforbeingsohelpfulandgivingusagoodresource.Thisisprobablythemostusefulthing

wehavegotoutofPROSPER."

[Managerofmedium-sizedhomerunbycorporation]

"Iwasfascinatedtoseehowmyhomecompareswiththeothers.Itisclearwherewehave

moretodoandwherewearedoingok.It'squiteencouraging."

[Managerfromsmallhomeownedbysmallgroup]

TheevaluationteamhasmadetheMonthlyMappingtoolcontinuouslyavailabletoallcarehomesthatwishtocontinueusingitafterPROSPERends.Basedonthesuccessofthetool,EssexCountyCouncilarealsodevelopingtheirownversion.

Safetyculturetool–MaPSaF/CultureisKeyTheManchesterPatientSafetyFramework(MaPSaF)wasoriginallydesignedforuse inhealthcare,asapromptforteamdiscussions.Thetoolasksteamstodiscusswheretheteamandorganisationliesonarangeoffactorsrelatingtosafetyculture,andtoreachconsensusthroughdiscussion(seeAppendix 3 for an example). PROSPER held a development session with care homes prior tolaunchingtheprogramme.TheaimwastorefinethewordingoftheMaPSaFtool,tomakeitmoreacceptableandappropriateforthecarehomecontext.

TheoriginalPROSPERproposal stated thatMaPSaFwouldbeusedwhencarehomes first join theprogrammetohelpidentifystrengthsandweaknessesineachhomeandtoguidethedevelopmentoftestsofchange.Thisdidnotoccur.

PROSPERdidsomedegreeoftestingofMaPSaFwithcarehomesbutthiswasnotprioritisedasanintervention.Initialtestsshowedthatthetoolinitsusualformatwasnotsuitableforthecarehomecontext.Althoughcarehomessaidtheycouldseethevalueoftheconcept,thelanguageofthetoolandtheimplementationprocessmaynotlenditselfwelltocarehomes.Carehomesfeltthattheycouldnotsetasidethreehoursforafullteamdiscussion,becausestaffarerequiredonthefloortoprovidecare.Theimplementationteamsaidthatshorterdiscussionsaboutselecteddimensionsofthetoolcouldbeused,butthecarehomesstillfoundthisproblematic.

128

“Ifoundithardtogetpeopletodoit.Atfirsttheywantedustohaveameetinganddiscussit,

canyoubelieveit?Wedon’thavetime.Itwouldtakethreeorfourhours.Didtheyexpectus

topaythewholeteamtositaroundandtalkaboutasurvey?Theysaidwecoulddiscussone

ofthetopicsateachmeeting,kindofonceamonthorsomething.We’renotdoingit.Wejust

filledintheformsandsentthemback.Wehaven’theardanythingback.”[Managerfromamedium-sizedhome,runbyalargecorporation]

“IthinkthisisanotherNHSthing.I’mnotsureitisrelevanttousreally.Youaresupposedto

haveameetingallthetimetodiscussitandagree.Idon’tthinkthatisreallygoingto

happen.It’sjustfittingitin.Wewouldallhavetostaylateorcomeinontheweekend.”[Seniorcarerfromalargehome,runbyalargecorporation]

Inthefirstcohort,theimplementationteamaskedcarehomestousethetoolasatypeofsurveyforstaff and then to discuss aspects at teammeetings. This was because the implementation teamrecognisedthatitwasunrealistictoexpectcarehomestosetasideoneormorehourstoundertakeafullMaPSaFdiscussion.Theythoughtitmaybeusefulforstafftolookatthedimensionsofthetoolthemselves,sendanonymouslycompletedformstoPROSPERforcollationandthenusetheresultsto guide future discussions in team meetings. This was done after discussion with the tooldevelopers.

However,theformusedwastheevaluationsheetfromtheMaPSAFtool(whichisusuallygivenoutat the beginning of group discussions). The implementation team did not adapt the language orformatofthetooltobeusedasasurvey,whichwouldhavebeenusefuliftherewasadesiretotestthis approach.Whilst the tool developerswere consulted, they are not experts in the care homecontext or in gaining feedback using a survey-type approach. This was not discussed with theevaluation team, who do have expertise in survey design. Perhaps as a result of the lack ofadaptationforthismethodofcollation,stafffoundtheprocessandthelanguagedifficult.

Althoughtencarehomessubmittedbatchesof‘surveys’foranalysis,resultsofthecompilationtooksometimetoreachcarehomes,andtherewas littleunderstandingofwhattodowiththeresultswhentheyarrived.

“Yeswehavetofillinaform.Laterwemighttalkonthetopics.Itisabitlikeatest.Itispretty

hard.Idon’tknowwhy.Wejustdowhatwearetoldandmakeittrytosoundgood.”

[Carerfromalargehome,runbyalargecorporation]

“Idon’tknowwhatit’sfor.Theysaiditistohelpus.[Manager]talkedaboutitinthestaff

meetingbutwejustalltickedwhatwethoughtsoundedgood.Ihaven’theardanythingback

fromit.”[Carerfromasmallhome,runbyasmallgroup]

Forthesecondcohort,PROSPERarrangedaworkshoptotrainmanagersinhowtofacilitatethetool.Theworkshopwaswell received, thoughparticipants said theywouldprefer for it tobe runafterintroductory training about other improvement approaches. However, whilstmanagers had goodintentionsaboutusingthetool,nohomeuseditthemselves.

129

In two instances PROSPER facilitators ran meetings or drop-in sessions in care homes using theMaPSaF tool. This is the wayMaPSaF was designed to be used. The care homes that receivedfacilitationusingthetoolwereverypositiveaboutthis,sayingthatithelpedthemseethingsfromadifferentperspective.ThissuggeststhatifarevisedMaPSaFtoolwasimplementedasplannedwiththecarehomes,itmayhavebeenuseful.HoweverthePROSPERteamineffectsetthetoolasideanddidnotuseitwithcohortthree,thoughtheyrestartedusingitinlimitedtestingmodeincohortfour.This was largely due to capacity within the team, but the team did choose to prioritise otheractivitiesthatwerenotoriginallypartoftheproposedinterventionovertestingoneoftheoriginalinterventiontools.

Late on in the project the implementation team re-introducedMaPSaF with a limited number ofcarehomesincohortfour,changingthewordingandstyleofMAPSAF,renamingthistheCultureisKey tool. A carer who had used the survey as part of cohort one became part of the PROSPERimplementationteamandhelpedtofurthersimplifythelanguage.Thetoolwassimplifiedontoonepage,withlanguagedesignedtoresonatewithcarehomestaff.PROSPERaimstotestthetoolwithuptotencarehomesincohortfourandthisprocessison-goingatthetimeofsubmittingthisreport

130

Resources

OnlineresourcesPROSPER set up an online forum (Knowledge Hub) for participating care homes. The aimwas tosharetoolsandsuccessstories.Howeveronly10%ofparticipatingcarehomesregisteredforthesite(nine carehomes). The implementation teamdidnotupload resources that the carehomeswerenotalreadyemailed.Nocarehomestendedtopostresourcesorcomments.

No care home reported actively using the site for discussion, sharing or networking. No specificbarrierstousewerenoted,thoughcarehomemanagersstatedthattheyhavelimitedtimeanddidnotseeanyadditionalinformationaddedtothesiteotherthanwhattheyalreadyhad.Inordertomaketheplatformmoreofaliveforum,newresources,slides,summariesofcommunityofpracticeeventsandlinkstootherwebsitesmayhaveneededtobeaddedmoreregularlysocarehomeshaveareasontojoinandcontribute.

AtwitteraccountwassetupforPROSPER,butnohomereflectedonitsusefulness.

OneparticipatingcarehomesetupaFacebookpage tohighlight someof their successes throughPROSPER.Residentsreactedwell tothis,askingfortheirphotostobeplacedonline.Asaresultofthissuccess,thePROSPERteamarenowplanningtosetupaFacebookpagefortheprogrammeasawhole.ThisisoneexampleofhowthePROSPERteamcontinuedtogainideasfromcarehomes,sothesharingofideaswastwo-way.

ToolkitofworksheetsAtoolkitofworksheetsand informationsheetswascompiledabout reducing falls,pressureulcersandurinarytractinfections.Somecarehomesstatedthattheyusedthechecklistsandinformationsheets,andsomedisplayedthematerialswithinthecarehome.Thefall’spreventionchecklistwasparticularlypopular.

However other care homes commented that the wording of the toolkit is sometimes difficult tounderstandandhardtodisseminatetocarers.Carehomesrequestedarevisionofthetoolkitsothatitmatcheswiththeeducationallevelandpreferredterminologyofthetargetaudience.

MonthlynewsletterFromJanuary2015,PROSPERbegandistributingashortcolourfulmonthlynewsletter. Initially thiswasonepage,butcarehomesaskedformoresothenewsletterbecametwotothreepages.Carehomes said they valued seeingphotographsand success stories fromother carehomes.Howeverthey wanted more practical details about what they were doing and how they have dealt withbarriers,sotheycouldadapttheideasfortheirowncarehomes.Anumberofcarehomesreportedseeinganideainthenewsletterandadaptingitfortheirownhome.

131

Appendix10.Communicationplan

ThisAppendixreproducesthePROSPERcommunicationsstrategydevelopedduringthesecondyearoftheprogrammeandsignedoffbytheprojectteaminOctober2015.Thedeliveryofthedifferentactivitiesdescribedinthestrategyisoutlinedintherelevantchaptersofthereport.

PROSPERcommunicationsplan–October2015

1. Context

ThisstrategydocumentoutlinestheapproachtocommunicationstakenforthePROSPERprogramme.

PROSPERseekstodeveloptheskillsthatarecentraltodeliveringimprovedsafetyforresidents.ItskeyaimistoimprovesafetyandreduceharmforcarehomeresidentsacrossEssex.

Undertheoverallaim,specificobjectivesareto:

• Worktogetherwithstaff,residentsandtheirrelativestofindwaysofintroducingnewqualityimprovementtrainingopportunitieswhichwillenablestafftoaddresssafetyconcerns;

• ReducethepercentageofresidentsthatpresentatA&E;• Increasetheproportionofresidentswhoare‘harmfree’(asdefinedbytheNHSSafety

Thermometer);• Reducetheprevalenceoffalls,pressureulcersandurinarytractinfectionsacrosscare

homes;• Increasestaffunderstandingof,capacityandcapabilitytocreateasafeenvironmentfor

residents;• Improvecollaborationandlearningbetweenhealthandsocialcare;and• Establishanevidencebasefortheintervention.

Thistwo-yearpilotprojecthasbeendevelopedinpartnershipwithEssexCountyCouncil,UCLPartnersandEssexresidentialcareandnursinghomes.

PROSPERisfundedbyTheHealthFoundation,anindependentcharityworkingtoimprovethequalityofhealthcareintheUK,andisthefirstsocialcareinitiativetobefundedbythecharity.

Sofarintheproject,muchworkhasgoneintocommunicatingwithlocalstakeholders–specificallycarehomes–toencourageasmanyhomesaspossibletotakepartintheprogramme.Communicationswillcontinueinthisveinforrecruitmenttothefinalcohort,butwillbebolsteredbycommunicationswithbroaderstakeholderstotrytosecurethefutureoftheintervention.

Programmevision:ThePROSPERprogrammebecomesbusinessasusualforcarehomesinEssexandacrossEngland.

Immediatechallenge:

132

• RollingouttheprogrammetothefinalcohortofcarehomesinEssexbeforetheendoftheprogramme.

• ObtainingfundingtocontinuetheprogrammeinEssexbeyondthecompletionofthepilotinthesummerof2016

• Achievingbehaviourchangeinstaff,carehomeownerstoimproveoutcomesforcarehomeresidents

Long-termchallenge:

• AchievingbehaviourchangeincommissionerstoprioritisePROSPER• DiffusingimprovementsresultingfromthepilotacrossEngland

2. Programmetimeline

Milestones/Comm’sopportunities

November2015

• Cohort4commencement

January2016

• HealthFoundation3rdprogressreportdue(15thJan)• Pan Essex Dementia Friendly Network launch (20th Jan)

February2016

• Prosperchampions’studydaystotakeplace• DignityinCareDay(1stFeb)

March2016

• Prospercommunityofpracticeevents

PROSPER 2016

1. Set up

4. External Evaluation

2014 2015

2. Intervention

3. Improvement network

MaPSaF

Write upDatacollectionandreview

Programmesetup

OngoingmeasurementandreviewImplementmeasurementtools

Capability&trainingprogrammeCohort1

Ongoingimprovement cyclesDesign&testsolutions

1 1

GatherBaseline



1


1


133

• NHSEnglandThinkKidneyAKIincarehomesprojectworkingwithProsperhomes–attendingcommunityofpractice

• NHSEnglandNutritionHydrationWeek(14th–20thMarch)

April2016

• CommencementofCohort5• InternationalForumonQualityandSafety,Gothenberg–posterdisplay

May2016

• DementiaAwarenessweek(18th-24thMay)

June2016

• TheProspers–awardceremony• HealthFoundationfinalreportandevaluationdue• Endofaward• Nationalcarehomeopenday(17thJune)

3. StakeholdersEasiertoreachgroups(orestablishedrelationships)• Residentsandrelatives• Carehomestaffinpilot• Carehomeownersinpilot• LocalgovernmentcommissionersinEssex• NHScommissionersinEssex• HealthFoundation• PatientgroupsandlocalHealthWatch• CommunityHealthProvidersinEssexHardertoreachgroups(orrelationshipsnotestablished)• Carehomestaff(outsideofpilot)• Carehomeowners(outsideofpilot)• Localgovernmentcommissioners(outsideofpilot)• NHScommissioners(outsideofpilot)• CQC• DepartmentofHealthandlocalgovernment• Researchcommunity(HSRUK,CLAHRCs,AHSNs&focusedacademicnetworks)

Adifferentapproachtocommunicationswillneedtobetakenwithstakeholdersdependingonwhethertheyhavealreadybeenengagedwiththeproject.Forexample,staffthathavetakenpartinthepilotmaybemoreinterestedinseeingtheirprogressandcontributiontothesuccessofthepilotinorderforthepositivebehaviourchangetocontinue.Staffnotalreadyengagedintheproject,wouldbeinterestedinwhatthebenefittothemisaswellasthebenefittopatients.

134

4. Keymessages

Thefollowingkeymessagesaresuitablefortheprojectasawhole,butwillneedtobeadaptedforkeyaudiences.Theproofpointsbeloweachkeymessageareusedtobolstercommunicationsbyprovidingevidenceforthemessage.

1. ThePROSPERprogrammeisencouragingapositivechangeofcultureincarehomesin

Essexthroughafocusoneducation,measurementandopenconversation• AsofMarch2016,abouttwo-thirdsofhomesreportedanemergingchangeinsafety

culture

• Carehomesarerobustlymeasuringoutcomesforresidents,andtrendsintheseoutcomesarebeingcloselymonitored

• Carehomesarereportingashifttowardsdefiningsafetyaspreventingharmforresidentsratherthanavoidingincidentsduetoinspectionsandsafeguardingconcerns.

2. PROSPERisenablingcarehomestofocusonproactivepreventionofincidentstoimproveresidentoutcomes

• Carehomesarereportinggreaterfocusonproactivepreventionandmonitoringofsafetyincidentswithastrongengagementwithusingcomparativedata

3. PROSPERisempoweringstaffinEssexcarehomesthrougheducationandimprovedknowledgeofhealthissues

• Homesreceivetraininginqualityimprovementmethods,aresourcetoolkit,toolstohelpmonitorchangeinsafetycultureandresidentoutcomes,opportunitiestomeetandsharelearningwithotherhomesandsupportvisitsfromcouncilimprovementfacilitators

• Carershavereportedfeelingthattheyhaveakeyroletoplayinimprovement,ratherthanamorenarrowlydefinedrolewashingorcleaning.

• Staffareencouragedtotrialandtestnewapproachesandmeasuretheresults,alongaPDSAcycle

• Thereisearlyevidenceofapossiblechangeinthedynamicbetweensomehomemanagersandtheirstaff,withsomestafffeelingmoreempoweredtoleadchangeandinnovatethanhasbeenthecaseinthepast.

4. ThePROSPERmodelcanbereplicatedandadaptedinanyareaofthecountry• ThemodelofengagementwithcarehomeshasbeenadoptedbyThinkKidneyinthe

developmentoftheirAKIresourcesforcarehomes.12Prosperhomesarepartofthetestpilot.Prosperprojectteamispartoftheworkinggroupdevelopingtheresources.

135

5. Outlineofcommunicationsapproach

Principles:

• PlaytoUSPsofthestudy:improvementincarehomeshighlytopical,solutionslieinideasandactionsoffrontlinestaff,highlevelofinterestinhowhealthandsocialcarecanworkandlearntogether,commitmenttoevaluation

• Useprojectpartnersasagentsfordisseminatinglearning-weneedtoidentifychampions

• Influenceusingestablishednetworkswhereverpossible

• Usemultiplemethodspersistentlytogetmessagesacross

6. Communicationsaimsandobjectives

Aim:RaiseawarenessofthePROSPERprogrammeto:

• EncourageuptakeincarehomesinEssex,• Spreadanddiffusethelearningfromthepilottosupportthecasefornationalroll

out.

Addressingtheimmediatechallenge(outlinedonpage5):

Targetaudience Whatwewantthemtothinkabout

Method

Residentsandrelatives

- TheirlovedonesarereceivingthebestcarewiththehelpofthePROSPERmodel

- Newsletters- Socialmedia- Traditionalmedia

Carehomestaff

- Howtheyaddvalueasemployees- Whattheycandotoimproveoutcomesforresidents

- Howtheycanproactivelyinstigatechange

- Theimportanceoflearningnewskills

- On-siteconversations,storiesaboutsuccessandchallengeselsewhere

- Newsletters(printed)- Seminars,workshops- Socialmedia

Carehomeowners

- Systematicimprovementmethodsaddvalueandachieveresults

- Smallchangesmakedifferences- Staffareanimportantuntappedresourceforimprovement

- CQCvaluesimprovementprogrammes

- Dataandprocessauditshelpinformcontinuousimprovements

- Theimportanceofcontinuingtheprogramme’sapproach

- Newsletters- Personalconversationsatlocallevelandwithnationalcorporations

- Commissioning/providerevents

136


Method

Addressingthelong-termchallenge(outlinedonpage5):


Method

NHScommissioners

- GPpracticesandambulanceservicesneedsupporttoworkwithcarehomes

- Reductionin999calloutsandinappropriatehospitaladmissionsmaybepossibleandisanimportantaim

- Thereismuchthathealthcouldlearnfromimprovementinthecarehomesector

- Personalconversations- CCGnewsletter- ViaagendaitemonHealthandWellbeingBoards

- ThroughAHSNnetworks

CQC

- Engagementwithsystematicimprovementshouldbevaluedaspartoftheregulatoryregime

- Regulationcanencouragespreadofimprovementlearning

- HomespromotingwhattheyhaveachievedduringCQCinspections

- Regionalforummeetings- Localityinformationsharingmeetings

DepartmentofHealthandlocalgovernment

- Improvementmethodscanbetransferredbetweensectorsbutneedadapting

- Buildingcapacityandcapabilityforimprovementrequirestimeandinvestment

- Learningfromourengagementwithcarehomes

- National,localconferences- Nationalwebsites- OpenlettertoAlistairBurt/JeremyHunt

Researchcommunitythrough;HSRUK,CLAHRCs,AHSNs,PatientSafety

- ImportanceofevaluatingnewQI/safetyapproachesinnewsectors,learningfromtransferringmethodsbetweensectors

- Peerreviewedpublications- Publicationsinprofessionaljournals

- Conferenceandseminarpresentations

137

Collaborative&FocusedAcademicNetworks

- BenefitsofembeddingevaluationintoQIprogramme,usingparticipatorymethods

- Blogs- ThroughAHSNandCLAHRCnetworks

- Spreadingbestpracticestakeholderevent

Media(local,national,trade)

- PROSPERisapositiveprojecttoimproveoutcomesforpeopleincarehomes

- Pressreleasestolocalmediaonimprovementsandactivitiesinspecifichomese.g.“noframeisthesame”

- SharelearninginHSJresourcesarticle

Inadditiontotheaboveactivities,thefollowingtoolswouldaidthepromotionofthePROSPERprogramme

• Creationofaninformationtoolkit,tobeusedbytheprojectteamtofacilitatecommunicationswithrangeofaudiences,suchasAHSNsandPatientSafetyCollaboratives.Toolkittoinclude:

o Executivesummaryoftheprojecto Casestudiesofgoodimprovemento Presentationslidedeckshowingoutcomesandlessonslearned

• Creationofaninfographic/animationtoexplaintheprojectsimplyforpotentialcommissioners,carehomes,staffandpublic.Tobepublishedonpartnerwebsitesandusedateventsandpresentations

• Developmentofcarehomestaffasspokespeoplefortheprogramme.Gaininginsightsfromstaffwilladdauthoritytoconversationswithpotentialcommissionersandcarehomes.Staffcouldspeakateventsorvideoscouldberecordedoftheirexperiences

Eventsandawardstoconsider:

Conferencesandawardscanbeaveryeasywaytoraisetheprofileofaprogramme(ashasbeenexperiencedalreadyforPROSPER).ItislikelythatUCLPartnerswillhavesomeformofexhibitionpresenceattheseconferences,butspeakerandworkshopopportunitiesshouldbeinvestigated.

• NHSConfed–15-16June2016www.nhsconfed.org/conference/join-us-in-2016

• Health+Care(CommissioningShow)–29-30June2016www.healthpluscare.co.uk/

• PatientSafetyCongressandAwards–6-7July2016www.patientsafetycongress.co.uk/

To