Provision, uptake and cost of cardiac rehabilitation programmes

Health Technology Assessment 2004; Vol. 8: No. 41Provision, uptake and cost of cardiacrehabilitation programmes: improvingservices to under-represented groupsAD Beswick, K Rees, I Griebsch, FC Taylor,M Burke, RR West, J Victory, J Brown,RS Taylor and S EbrahimOctober 2004Health Technology AssessmentNHS R&D HTA ProgrammeHTA

HTAHow to obtain copies of this and other HTA Programme reports.An electronic version of this publication, in Adobe Acrobat format, is available for downloading free ofcharge for personal use from the HTA website (http://www.hta.ac.uk). A fully searchable CD-ROM isalso available (see below).Printed copies of HTA monographs cost £20 each (post and packing free in the UK) to both public andprivate sector purchasers from our Despatch Agents.Non-UK purchasers will have to pay a small fee for post and packing. For European countries the cost is£2 per monograph and for the rest of the world £3 per monograph.You can order HTA monographs from our Despatch Agents:– fax (with credit card or official purchase order)– post (with credit card or official purchase order or cheque)– phone during office hours (credit card only).Additionally the HTA website allows you either to pay securely by credit card or to print out yourorder and then post or fax it.Contact details are as follows:HTA DespatchEmail: orders@hta.ac.ukc/o Direct Mail Works Ltd Tel: 02392 492 0004 Oakwood Business Centre Fax: 02392 478 555Downley, HAVANT PO9 2NP, UK Fax from outside the UK: +44 2392 478 555NHS libraries can subscribe free of charge. Public libraries can subscribe at a very reduced cost of£100 for each volume (normally comprising 30–40 titles). The commercial subscription rate is £300per volume. Please see our website for details. Subscriptions can only be purchased for the current orforthcoming volume.Payment methodsPaying by chequeIf you pay by cheque, the cheque must be in pounds sterling, made payable to Direct Mail Works Ltdand drawn on a bank with a UK address.Paying by credit cardThe following cards are accepted by phone, fax, post or via the website ordering pages: Delta, Eurocard,Mastercard, Solo, Switch and Visa. We advise against sending credit card details in a plain email.Paying by official purchase orderYou can post or fax these, but they must be from public bodies (i.e. NHS or universities) within the UK.We cannot at present accept purchase orders from commercial companies or from outside the UK.How do I get a copy of HTA on CD?Please use the form on the HTA website (www.hta.ac.uk/htacd.htm). Or contact Direct Mail Works (seecontact details above) by email, post, fax or phone. HTA on CD is currently free of charge worldwide.The website also provides information about the HTA Programme and lists the membership of the variouscommittees.

Provision, uptake and cost of cardiacrehabilitation programmes: improvingservices to under-represented groupsAD Beswick, 1 K Rees, 1 I Griebsch, 2 FC Taylor, 3M Burke, 1 RR West, 4 J Victory, 5 J Brown, 2RS Taylor 6 and S Ebrahim 1*1 Department of Social Medicine, University of Bristol, UK2 MRC Health Services Research Collaboration, Department of SocialMedicine, University of Bristol, UK3 Bristol Heart Institute, University of Bristol, UK4 Wales Heart Research Institute, University of Wales College of Medicine,Cardiff, UK5 United Bristol Healthcare NHS Trust, UK6 Department of Public Health and Epidemiology, University ofBirmingham, UK* Corresponding authorDeclared competing interests of authors: nonePublished October 2004This report should be referenced as follows:Beswick AD, Rees K, Griebsch I, Taylor FC, Burke M, West RR, et al. Provision, uptakeand cost of cardiac rehabilitation programmes: improving services to under-representedgroups. Health Technol Assess 2004;8(41).Health Technology Assessment is indexed in Index Medicus/MEDLINE and Excerpta Medica/EMBASE.

NHS R&D HTA ProgrammeThe research findings from the NHS R&D Health Technology Assessment (HTA) Programme directlyinfluence key decision-making bodies such as the National Institute for Clinical Excellence (NICE)and the National Screening Committee (NSC) who rely on HTA outputs to help raise standards of care.HTA findings also help to improve the quality of the service in the NHS indirectly in that they form a keycomponent of the ‘National Knowledge Service’ that is being developed to improve the evidence ofclinical practice throughout the NHS.The HTA Programme was set up in 1993. Its role is to ensure that high-quality research information onthe costs, effectiveness and broader impact of health technologies is produced in the most efficient wayfor those who use, manage and provide care in the NHS. ‘Health technologies’ are broadly defined toinclude all interventions used to promote health, prevent and treat disease, and improve rehabilitationand long-term care, rather than settings of care.The HTA programme commissions research only on topics where it has identified key gaps in theevidence needed by the NHS. Suggestions for topics are actively sought from people working in theNHS, the public, consumer groups and professional bodies such as Royal Colleges and NHS Trusts.Research suggestions are carefully considered by panels of independent experts (including consumers)whose advice results in a ranked list of recommended research priorities. The HTA Programme thencommissions the research team best suited to undertake the work, in the manner most appropriate to findthe relevant answers. Some projects may take only months, others need several years to answer theresearch questions adequately. They may involve synthesising existing evidence or designing a trial toproduce new evidence where none currently exists.Additionally, through its Technology Assessment Report (TAR) call-off contract, the HTA Programme isable to commission bespoke reports, principally for NICE, but also for other policy customers, such as aNational Clinical Director. TARs bring together evidence on key aspects of the use of specifictechnologies and usually have to be completed within a limited time period.Criteria for inclusion in the HTA monograph seriesReports are published in the HTA monograph series if (1) they have resulted from work commissionedfor the HTA Programme, and (2) they are of a sufficiently high scientific quality as assessed by the refereesand editors.Reviews in Health Technology Assessment are termed ‘systematic’ when the account of the search,appraisal and synthesis methods (to minimise biases and random errors) would, in theory, permit thereplication of the review by others.The research reported in this monograph was commissioned by the HTA Programme as project number99/21/02. As funder, by devising a commissioning brief, the HTA Programme specified the researchquestion and study design. The authors have been wholly responsible for all data collection, analysis andinterpretation and for writing up their work. The HTA editors and publisher have tried to ensure theaccuracy of the authors’ report and would like to thank the referees for their constructive comments onthe draft document. However, they do not accept liability for damages or losses arising from materialpublished in this report.The views expressed in this publication are those of the authors and not necessarily those of the HTAProgramme or the Department of Health.Editor-in-Chief:Professor Tom WalleySeries Editors:Dr Peter Davidson, Professor John Gabbay, Dr Chris Hyde,Dr Ruairidh Milne, Dr Rob Riemsma and Dr Ken SteinManaging Editors:Sally Bailey and Caroline CiupekISSN 1366-5278© Queen’s Printer and Controller of HMSO 2004This monograph may be freely reproduced for the purposes of private research and study and may be included in professional journals providedthat suitable acknowledgement is made and the reproduction is not associated with any form of advertising.Applications for commercial reproduction should be addressed to NCCHTA, Mailpoint 728, Boldrewood, University of Southampton,Southampton, SO16 7PX, UK.Published by Gray Publishing, Tunbridge Wells, Kent, on behalf of NCCHTA.Printed on acid-free paper in the UK by St Edmundsbury Press Ltd, Bury St Edmunds, Suffolk.G

Health Technology Assessment 2004; Vol. 8: No. 41AbstractProvision, uptake and cost of cardiac rehabilitation programmes:improving services to under-represented groupsAD Beswick, 1 K Rees, 1 I Griebsch, 2 FC Taylor, 3 M Burke, 1 RR West, 4 J Victory, 5J Brown, 2 RS Taylor 6 and S Ebrahim 1*1 Department of Social Medicine, University of Bristol, UK2 MRC Health Services Research Collaboration, Department of Social Medicine, University of Bristol, UK3 Bristol Heart Institute, University of Bristol, UK4 Wales Heart Research Institute, University of Wales College of Medicine, Cardiff, UK5 United Bristol Healthcare NHS Trust, UK6 Department of Public Health and Epidemiology, University of Birmingham, UK* Corresponding authorObjectives: To estimate UK need for outpatientcardiac rehabilitation, current provision andidentification of patient groups not receiving services.To conduct a systematic review of literature onmethods to improve uptake and adherence to cardiacrehabilitation. To estimate cost implications ofincreasing uptake of cardiac rehabilitation.Data sources: Hospital Episode Statistics (England).Hospital Inpatient Systems (Northern Ireland). PatientsEpisode Database for Wales. British Association forCardiac Rehabilitation/British Heart Foundationsurveys. Cardiac rehabilitation centres. Patients fromgeneral hospitals. Electronic databases.Review methods: The study analysed hospitaldischarge statistics to ascertain the population need foroutpatient cardiac rehabilitation in the UK. Surveys ofcardiac rehabilitation programmes were conducted todetermine UK provision, uptake and audit activity, andto identify local interventions to improve uptake. Datawere also examined from a trial estimating eligibility forcardiac rehabilitation and non-attendance. A systematicreview of interventions to improve patient uptake,adherence and professional compliance in cardiacrehabilitation was conducted. Estimated costs ofimproving uptake were identified from national survey,systematic review and sampled cardiac rehabilitationprogrammes.Results: In England, Wales and Northern Ireland nearly146,000 patients discharged from hospital with primarydiagnosis of acute myocardial infarction, unstable anginaor following revascularisation were potentially eligiblefor cardiac rehabilitation. In England in 2000, 45–67%of these patients were referred, with 27–41%attending outpatient cardiac rehabilitation. If alldischarge diagnoses of ischaemic heart disease wereconsidered, nearly 299,000 patients would bepotentially eligible and in England rates of attendanceand referral would be 22–33% and 13–20%respectively. Rates of referral and attendance weresimilar in Wales, but somewhat lower in NorthernIreland. It was found that referral and attendance ofolder people and women at cardiac rehabilitationtended to be low. It was also suggested that patientsfrom ethnic minorities and those with angina or heartfailure were less likely to be referred to or joinprogrammes. A wide range of local interventionssuggested awareness of the problem of uptake. In anNHS-funded randomised controlled trial, possiblyrepresenting more optimal protocol-led care, medicaland nursing staff identified 73–81% of patients withacute myocardial infarction as eligible for cardiacrehabilitation. Excluded patients tended to be olderwith more severe presentation of cardiac disease.Experiences of patients suggested that uptake may beimproved by addressing issues of motivation andrelevance of rehabilitation to future well-being, comorbidities,site and time of programme, transport andcare for dependants. Systematic review of studiessupported the use of letters, pamphlets or home visitsto motivate patients and the use of trained lay visitors.Self-management techniques showed some value inpromoting adherence to lifestyle changes. Studiesexamining professional compliance found thatprofessional support for practice nurses may have valuein the coordination of postdischarge care. Averagecosts in 2001 of cardiac rehabilitation to the healthiii© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Abstractservice per patient completing a cardiac rehabilitationprogramme were about £350 (staff only) and £490(total). If services were modelled on an intermediatemultidisciplinary configuration with three to five keystaff, approximately 13% more patients could betreated with the same budget. Depending on staffingconfiguration an approximate 200–790% budgetincrease would be required to provide cardiacrehabilitation to all potentially eligible patients.Conclusions: Provision of outpatient cardiacrehabilitation in the UK is low and little is known aboutthe capacity of cardiac rehabilitation centres to increasethis provision. There is an uncoordinated approach toaudit data collection and few interventions aimed atimproving the situation have been formally evaluated.Motivational communications and trained lay volunteersmay improve uptake of cardiac rehabilitation, as mayself-management techniques. Experience of low-costinterventions and good practice exists withinrehabilitation centres, although cost informationfrequently is not reported. Increased provision ofoutpatient cardiac rehabilitation will require extraresources. Further trials are required to compare thecost-effectiveness of comprehensive multidisciplinaryrehabilitation with simpler outpatient programmes, alsoresearch is needed into economic and patientpreference studies of the effects of different methodsof using increased funding for cardiac rehabilitation. Anevaluation of a range of interventions to promoteattendance in all patients and under-representedgroups would also be useful. The development ofstandards is suggested for audit methods and foreligibility criteria, as well as regular and comprehensivedata collection to estimate the need for and provisionof cardiac rehabilitation. Further areas for interventioncould be identified through qualitative studies, and theextension of low-cost interventions and good practicewithin rehabilitation centres. Regularly updatedsystematic reviews of relevant literature would also beuseful.iv

Health Technology Assessment 2004; Vol. 8: No. 41ContentsList of abbreviations ..................................Executive summary ....................................1 Background ................................................ 1Cardiac rehabilitation ................................ 1Effectiveness in patients with coronaryheart disease ............................................... 1Effectiveness in specific patient groups ..... 2Uptake of cardiac rehabilitation ................ 2Under-representation in cardiacrehabilitation .............................................. 2Barriers to uptake and adherence ............. 4Economic aspects of cardiacrehabilitation .............................................. 5Conclusions ................................................ 62 Objectives .................................................. 73 Population need for cardiac rehabilitationin the UK .................................................... 9Objectives ................................................... 9Background ................................................ 9Methods ...................................................... 9Results ........................................................ 10Discussion ................................................... 11Conclusions ................................................ 124 Provision and uptake of cardiacrehabilitation in the UK: national surveyof UK cardiac rehabilitation services ........ 13Objective ..................................................... 13Methods ...................................................... 13Results ........................................................ 13Conclusions ................................................ 155 Audit of cardiac rehabilitation in England:National Service Framework for CoronaryHeart Disease recommendations .............. 17Objective ..................................................... 17Background ................................................ 17Methods ...................................................... 17Results ........................................................ 17Discussion ................................................... 19Conclusions ................................................ 206 Uptake and adherence in a randomisedcontrolled trial of cardiac rehabilitationafter myocardial infarction ........................ 23Objective ..................................................... 23viiixIntroduction ............................................... 23Methods ...................................................... 23Results ........................................................ 23Discussion ................................................... 26Conclusions ................................................ 277 Systematic review of interventions toimprove uptake, adherence andprofessional compliance with cardiacrehabilitation .............................................. 29Definitions .................................................. 29Objective ..................................................... 29Methods ...................................................... 298 Systematic review of interventions toimprove uptake of cardiacrehabilitation .............................................. 31Background ................................................ 31Results ........................................................ 31Discussion ................................................... 34Conclusions ................................................ 389 Systematic review of interventions to improveadherence to cardiac rehabilitation .......... 39Background ................................................ 39Results ........................................................ 39Discussion ................................................... 45Conclusions ................................................ 4810 Systematic review of interventions toimprove professional compliance withcardiac rehabilitation ................................. 49Background ................................................ 49Results ........................................................ 49Discussion ................................................... 52Conclusions ................................................ 5311 Health service costs of cardiacrehabilitation in the UK ............................. 55Objectives ................................................... 55Health service costs associated withcardiac rehabilitation ................................. 55The national budget attributable tocardiac rehabilitation ................................. 58Discussion ................................................... 60Conclusions ................................................ 6312 Conclusions ................................................ 65What is the population need for cardiacrehabilitation? ............................................ 65v

ContentsWho is not receiving cardiacrehabilitation? ............................................ 65What is the effectiveness of differentmethods of improving uptake and ofdifferential targeting of cardiacrehabilitation? ............................................ 66What is the potential budgetimpact of increasing uptake of cardiacrehabilitation using different uptakeinterventions? ............................................. 6813 Key findings ................................................ 71Implications for healthcare ........................ 71Acknowledgements .................................... 73References .................................................. 75Appendix 1 Need for cardiacrehabilitation in the UK ............................. 83Appendix 2 Need for and estimatedlevel of cardiac rehabilitation provisionin the UK .................................................... 87Appendix 3 British Association for CardiacRehabilitation additional postalquestionnaire .............................................. 91Appendix 4 Literature searchstrategies ..................................................... 95Appendix 5 Inclusion/exclusion form ....... 97Appendix 6 Data extraction form ............. 99Appendix 7 Flow diagram of the systematicreview of interventions to improve uptake ofcardiac rehabilitation (QUOROM statementflow diagram) ............................................. 103Appendix 8 Studies evaluating interventionsto improve the uptake of cardiacrehabilitation .............................................. 105Appendix 9 Studies excluded fromthe review of interventions toimprove uptake of cardiacrehabilitation .............................................. 113Appendix 10 Flow diagram of the systematicreview of interventions to improve adherenceto cardiac rehabilitation (QUOROMstatement flow diagram) ............................ 115Appendix 11 Studies evaluating interventionsto improve adherence to cardiacrehabilitation .............................................. 117Appendix 12 Studies excluded from thereview of methods to improve adherence tocardiac rehabilitation ................................. 129Appendix 13 Flow diagram of the systematicreview of interventions to improveprofessional compliance with cardiacrehabilitation (QUOROM statement flowdiagram) ..................................................... 131Appendix 14 Studies evaluatinginterventions to improve professionalcompliance with cardiac rehabilitation ...... 133Appendix 15 Studies excluded from thereview of interventions to improveprofessional compliance with cardiacrehabilitation .............................................. 139Appendix 16 Estimates for unit costs fordifferent staff categories and grades .......... 141Appendix 17 List of equipment ................ 143Appendix 18 Staff input: average hours perweek ............................................................ 145Appendix 19 Referral, uptake andcompletion rates for 30 randomly selectedUK cardiac rehabilitation programmesin 2000 ....................................................... 149Appendix 20 Average cost estimates forcardiac rehabilitation (detailed table) ........ 151Health Technology Assessment reportspublished to date ....................................... 153Health Technology AssessmentProgramme ................................................ 163vi

Health Technology Assessment 2004; Vol. 8: No. 41List of abbreviationsAMIacute myocardial infarctionIQRinterquartile rangeBACRBHFBritish Association for CardiacRehabilitationBritish Heart FoundationMINSF-CHDmyocardial infarctionNational Service Framework forCoronary Heart DiseaseCABGcoronary artery bypass graftORodds ratioCHDCICRcoronary heart diseaseconfidence intervalcardiac rehabilitationPEDWPTCAPatient Episode Database forWalespercutaneous transluminalcoronary angioplastyHESHospital Episode Statistics(England)QUOROMquality of reporting ofmeta-analysesHFheart failureRCTrandomised controlled trialHISICD-10Hospital Inpatient Systems(Northern Ireland)International Classification ofDiseases-10RRSDUArelative riskstandard deviationunstable anginaIHDischaemic heart diseaseVATvalue added taxAll abbreviations that have been used in this report are listed here unless the abbreviation is well known (e.g. NHS), orit has been used only once, or it is a non-standard abbreviation used only in figures/tables/appendices in which casethe abbreviation is defined in the figure legend or at the end of the table.vii© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Executive summaryBackgroundThe National Service Framework for CoronaryHeart Disease (NSF-CHD) identifies patients withacute myocardial infarction and following coronaryrevascularisation as eligible for outpatient cardiacrehabilitation. However, rehabilitation uptakeremains low, particularly in some specific patientgroups. While many barriers to patientparticipation have been described, theeffectiveness of interventions to improve uptakeand adherence has not been assessed by systematicreview. Furthermore, the cost implications ofinterventions to improve uptake and adherenceand of increasing overall provision to meet totalpopulation need have not been estimated.Objectives●●●●To estimate UK population need and updateestimates of cardiac rehabilitation provision.To identify patient groups not receiving cardiacrehabilitation.To review effectiveness of methods to improveuptake and adherence to cardiac rehabilitation.To estimate cost implications of increasinguptake of cardiac rehabilitation.Methods●●●●Analysis of hospital discharge statistics toascertain the population need for outpatientcardiac rehabilitation in the UK.Surveys of cardiac rehabilitation programmes todetermine UK provision, uptake and audit activity,and to identify local interventions to improveuptake. Estimation of eligibility for cardiacrehabilitation and non-attendance in a recent trial.Systematic review of interventions to improvepatient uptake, adherence and professionalcompliance in cardiac rehabilitation.Assessment of costs of improving uptakeidentified from national survey, systematic reviewand sampled cardiac rehabilitation programmes.ResultsPopulation need and provisionIn England, Wales and Northern Ireland nearly146,000 patients discharged from hospital with aprimary diagnosis of acute myocardial infarction,unstable angina or following revascularisation werepotentially eligible for cardiac rehabilitation. InEngland in 2000, 45–67% of these patients werereferred, with 27–41% attending outpatient cardiacrehabilitation. If all discharge diagnoses of ischaemicheart disease (including angina pectoris and heartfailure) were considered, nearly 299,000 patientswould be potentially eligible, with rates of referraland attendance of 22–33% and 13–20%, respectively.Rates of referral and attendance were similar inWales, but somewhat lower in Northern Ireland.Patient uptakeReferral and attendance of older people andwomen at cardiac rehabilitation tended to be low.There was a suggestion that patients from ethnicminorities and those with angina or heart failurewere less likely to be referred to or joinprogrammes. A wide range of local interventionssuggested awareness of the problem of uptake.The survey of cardiac rehabilitation centres inEngland identified an uncoordinated approach toaudit, with variations in methods and contentdespite guidelines and the NSF requirements.In an NHS-funded, multicentre, randomisedcontrolled trial, possibly representing moreoptimal protocol-led care, medical and nursingstaff identified 73–81% of patients with acutemyocardial infarction as eligible for cardiacrehabilitation. Excluded patients tended to beolder with more severe presentation of cardiacdisease. Experiences of patients suggested thatuptake may be improved by addressing issues ofmotivation and relevance of rehabilitation tofuture well-being, co-morbidities, site and time ofprogramme, transport and care for dependants.Systematic reviewA comprehensive search strategy identified studiesrelating to uptake, adherence or professionalcompliance with cardiac rehabilitation. Of 3261references identified, 957 were acquired aspotentially relevant. Reports were frequently notpublished in easily accessible form. The majorityof studies were small, of short duration and not ofhigh quality. Consequently, none of the findingsix© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Executive summaryxcan be considered definitive. Few studies reportedcost implications.Eight studies (three randomised) evaluatedmethods to improve patient uptake of cardiacrehabilitation. These supported the use of letters,pamphlets or home visits to motivate patients.Some encouragement was found for the use oftrained lay visitors. Fourteen studies (sevenrandomised) evaluated methods to improveoverall patient attendance or maintenance oflifestyle changes associated with cardiacrehabilitation. Self-management techniquesshowed some value in promoting adherence tolifestyle changes. Six studies (two randomised)evaluated methods to improve patient uptake andadherence to cardiac rehabilitation by improvingprofessional compliance with guidelines and goodpractice. Although no effective interventionsspecifically aimed at improving professionalcompliance were found, professional support forpractice nurses may have value in the coordinationof postdischarge care.Healthcare costsAverage costs in 2001 of cardiac rehabilitation tothe health service per patient completing a cardiacrehabilitation programme were about £350 (staffonly) and £490 (total). It is estimated thatoutpatient cardiac rehabilitation represented anNHS cost of £15–24 million in the UK. Variationin cost per patient across centres was partlyexplained by the duration of rehabilitation andstaff-to-patient ratio. If services were modelled onan intermediate multidisciplinary configurationwith three to five key staff, approximately 13%more patients could be treated with the samebudget. If the most modest services were provided,40% more patients could be treated. Dependingon staffing configuration an approximate200–790% budget increase would be required toprovide cardiac rehabilitation to all potentiallyeligible patients.ConclusionsImplications for healthcare●●Provision of outpatient cardiac rehabilitation inthe UK is low, well below the NSF-CHD goal of85% of patients with acute myocardial infarctionand following revascularisation being offeredoutpatient cardiac rehabilitation.Information on referral to and uptake of cardiacrehabilitation is incomplete, with widely varyingestimates of provision, particularly in underrepresentedgroups. Little is known about the●●●●●●●●capacity of cardiac rehabilitation centres toincrease provision.There is an uncoordinated approach to auditdata collection.Reasons reported by patients for nonattendanceare amenable to intervention, butfew interventions have been formally evaluated.Many interventions aimed at improving patientuptake, adherence and professional compliancewith guidelines and good practice have beenproposed, but few have been formallyevaluated.Motivational communications and trained layvolunteers may improve uptake of cardiacrehabilitation.Self-management techniques may help topromote lifestyle change associated with cardiacrehabilitation.Information on costs of interventions isfrequently not reported.Experience of low-cost interventions and goodpractice exists within rehabilitation centres.Increased provision of outpatient cardiacrehabilitation will require extra resources.Recommendations for researchand development●●●●●●●Trials comparing the cost-effectiveness ofcomprehensive multidisciplinary rehabilitationwith simpler outpatient programmes.Economic and patient preference studies of theeffects of different methods of using increasedfunding for cardiac rehabilitation, and evaluationsof the impact of any increased funding.Evaluation of a range of interventions(including self-management techniques,motivational communication and the use oftrained lay volunteers) to promote attendancein all patients and under-represented groups.Development of standardised audit methods inthe context of modern records systems,appropriate training for dedicated staff anddialogue between service contributors.Standardisation of criteria for patient eligibility,regular and comprehensive data collection toestimate the need for and provision of cardiacrehabilitation.Identification of further areas for interventionthrough qualitative studies.Extension of low-cost interventions and goodpractice within rehabilitation centres.Regular updated systematic review of literaturerelating to uptake and adherence to cardiacrehabilitation to include ‘grey’ literature andnon-UK studies.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 1BackgroundCardiac rehabilitationComprehensive cardiac rehabilitation offerspatients with coronary heart disease a long-termprogramme involving medical evaluation,‘prescribed’ exercise, cardiac risk factormodification, education and counselling. 1 Inpartnership with a multidisciplinary team ofhealth professionals, patients with cardiac diseaseare encouraged and supported to achieve andmaintain optimal physical and psychosocialhealth. 2In the UK cardiac rehabilitation usually comprisesfour phases 3 in which the themes of exercise,education, psychological support and counsellingare addressed to a level appropriate to the stage ofrecovery. Throughout, consideration is given tothe processes of explanation and understanding, 4and the overall aim of long-term maintenance of ahealthy lifestyle.The first phase takes the form of counselling witha simple programme of education andpsychological support while in hospital. 5 Physical,psychological and social needs for cardiacrehabilitation are assessed and advice is given oneveryday activities with encouragement to takelight exercise in the first few weeks at home, thesecond phase of rehabilitation. Home visiting andtelephone contact, and the use of educationalmaterials or a supervised self-help programme,provide support during this period. The thirdphase of rehabilitation is delivered in anoutpatient setting by appropriate healthprofessionals and lasts typically for 6–8 weeks. Keyprogramme elements are supervised exercise,education on secondary prevention and risk factormodification, and psychological approaches torecovery. Maintenance of healthy behaviours aftercompletion of the outpatient programme is thefourth phase of cardiac rehabilitation. Continuedexercise and adherence with lifestyle changes maybe mediated through a cardiac support group.Effectiveness in patients withcoronary heart diseaseThe effectiveness of cardiac rehabilitation hasbeen the subject of several randomised trials andreviews. 6–9 Most recently, a Cochrane systematicreview concluded that exercise-based cardiacrehabilitation is effective in reducing cardiacdeaths, cardiovascular morbidity and primary riskfactors in patients who have had myocardialinfarction. 10 An earlier overview of the evidenceconducted by the NHS Centre for Reviews andDissemination stated that a combination ofexercise, psychological and educationalinterventions is the most effective form of cardiacrehabilitation, 11 but the efficacy of combinationsand durations of different components of therehabilitation package remains uncertain. InEngland, the National Service Framework forCoronary Heart Disease (NSF-CHD) concludedthat there is scope for improving services so thatall those in need are offered rehabilitation. 12Evidence for the effectiveness of cardiacrehabilitation mainly derives from studies ofpatients with myocardial infarction and there areinsufficient data to stratify systematic reviews byindication. 10 However, the inclusion in reviewedtrials of patients who have undergonerevascularisation, that is, coronary artery bypassgraft (CABG) or percutaneous transluminalcoronary angioplasty (PTCA), or who have hadangina pectoris or coronary artery disease definedby angiography suggests the possibility of benefitfor these groups. Furthermore, while there is noconclusive evidence that cardiac rehabilitationreduces mortality in patients with heart failure, arecent systematic review looking specifically atexercise interventions found physiological benefitsand positive effects on quality of life in selectedsubgroups. 13Guidelines recommend that outpatient cardiacrehabilitation should be available for patientsfollowing myocardial infarction, PTCA and CABG,and for patients with angina, heart failure 14 andarrhythmia. 15,16 The Fifth Report on the Provisionof Services for Patients with Coronary HeartDisease states that in the UK patients must haveaccess to rehabilitation when required, forexample after a heart attack, cardiac surgery andintervention. 17 In England the NSF-CHDidentifies patients who have survived acutemyocardial infarction and those who have1© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Background2undergone CABG or PTCA as priorities forcardiac rehabilitation. 12 When high-quality-cardiacrehabilitation is available to these patients theNSF-CHD recommends that services should beextended to patients with angina and heart failure.In Wales, cardiac rehabilitation should be providedfor all those who have had an episode of acutecoronary syndrome, some of whom will haveundergone a revascularisation procedure. 18 TheCanadian Association for Cardiac Rehabilitationstates that cardiac rehabilitation should beconsidered standard, usual care for virtually allpatients with documented cardiovasculardisease. 19Effectiveness in specific patientgroupsEvidence for the effectiveness of cardiacrehabilitation in older and female patients islimited, as participants in trials tend to be youngerand predominantly male. In the most recentsystematic review the mean age of patients inexercise only studies was 53 years (range of means50–70 years) and in trials of comprehensivecardiac rehabilitation 56 years (range of means47–63 years). 10 Women comprised 4% of patientsin exercise-only and 11% of patients incomprehensive cardiac rehabilitation trials. Trialsshow a bias towards the inclusion of men and mostexcluded older people. However, systematicreviews provide no evidence to suggest that elderlyor female patients benefit less than younger ormale patients. 10 Indeed, it is possible that thepatients who would benefit most from cardiacrehabilitation are those excluded from trials on thegrounds of age, gender or co-morbidity. 20In elderly patients the goals of cardiacrehabilitation may differ from those of youngerpatients, and include the preservation of mobility,self-sufficiency and mental function. 21 Cardiacrehabilitation may represent an opportunity toprovide effective healthcare and achieve a highquality of life for older patients. 22 Similarly, thefrequently lower level of fitness observed inwomen at the time of hospitalisation suggests agreater potential for health improvement withcardiac rehabilitation. 23,24In trials of cardiac rehabilitation the ethnicbackground of patients is seldom reported, 10 but itis likely that trial participants are mainly whiteCaucasian. There is neither evidence nor amechanism to suggest lack of benefit in ethnicminority groups. 11Thus, evidence from randomised controlled trials(RCTs), as demonstrated in Figure 1, supports theeffectiveness of cardiac rehabilitation in a range ofcardiac diagnoses including post-myocardialinfarction, post-PTCA, post-CABG, anginapectoris and heart failure. To date, althoughpatients with different cardiac conditions, andfemale, elderly and non-white Caucasian ethnicgroups, have been poorly represented in trials ofcardiac rehabilitation there is no evidence tosuggest that outcomes are less favourable.Uptake of cardiac rehabilitationAlthough it is considered effective in quickeningrecovery and improving prognosis, not all patientsparticipate in a cardiac rehabilitation programme.Several recent UK surveys have reported theuptake of cardiac rehabilitation by patients with adischarge diagnosis of coronary heart disease. 25–29These are summarised in Table 1.Surveys in the UK show low levels of patientparticipation (14–43% after myocardial infarction)with similarly low attendance reported inAustralia, 30 France, 31 New Zealand 32 and theUSA. 33–35 Low patient participation is aconsequence of low levels of provision, referraland invitation, and of poor uptake by patients.Under-representation in cardiacrehabilitationPatients participating in cardiac rehabilitationprogrammes have tended to be male, middle-agedand diagnosed with uncomplicated myocardialinfarction. 36 Those who do not participate in aprogramme often have greater degrees offunctional impairment and are the patients mostin need of and most likely to benefit fromrehabilitation. 20Variation in referral rates for patients withdifferent cardiac diagnoses reflects the traditionalindication for cardiac rehabilitation services ofmyocardial infarction and CABG. 37 CABG patientstend to be younger than those with myocardialinfarction and this may explain some of theincreased rehabilitation uptake seen after CABG(see Table 1). 26 Patients admitted for PTCA are lesslikely to be invited or participate, probably as aconsequence of the short hospital stay and thelimited opportunities for recruitment. Also, theprocedure is less invasive and painful than CABG,with a quicker recovery and return to work and

Health Technology Assessment 2004; Vol. 8: No. 41Review:Comparison:Outcome:Studyor subcatogoryExercise-based rehabilitation for coronary heart disease02 Exercise plus other rehabilitation versus usual care04 Pooled mortality, non-fatal MI, CABG, PTCATreatment Control Peto OR Weight Peto ORn/N n/N 95% CI % 95% CIEngblom 29/119 35/109 6.37 0.68 (0.38 to 1.22)WHO Balatonfured 16/80 9/80 2.93 1.93 (0.83 to 4.53)WHO Brussels 25/85 24/81 4.79 0.99 (0.51 to 1.92)WHO Bucharest 16/65 23/64 3.78 0.59 (0.28 to 1.24)WHO Budapest 38/101 29/99 6.18 1.45 (0.81 to 2.61)WHO Dessau 4/29 7/25 1.23 0.42 (0.11 to 1.58)WHO Erfut 13/63 15/56 2.97 0.71 (0.31 to 1.66)WHO Ghent 19/84 12/84 3.51 1.73 (0.80 to 3.77)WHO Helsinki 41/188 56/187 9.95 0.65 (0.41 to 1.04)WHO Kaunas 19/66 17/49 3.37 0.76 (0.34 to 1.68)WHO Prague 15/59 20/53 3.34 0.57 (0.26 to 1.26)WHO Rome 8/34 6/29 1.52 1.18 (0.36 to 3.83)WHO Tel Aviv 14/63 8/51 2.44 1.52 (0.60 to 3.85)WHO Warsaw 6/39 8/40 1.61 0.73 (0.23 to 2.31)Sivarajan 82 10/86 10/84 2.45 0.97 (0.38 to 2.47)Bengtsson 83 12/81 10/90 2.65 1.39 (0.57 to 3.40)Fridlund 91 26/87 39/91 5.73 0.57 (0.31 to 1.05)Oldridge 91 3/99 4/102 0.94 0.77 (0.17 to 3.46)PRECOR 6/60 11/61 2.03 0.52 (0.19 to 1.44)Bertie 92 1/57 4/53 0.66 0.27 (0.04 to 1.59)Schuler/Niebauer 20/56 25/57 3.77 0.71 (0.34 to 1.51)Heller 93 46/213 54/237 10.73 0.93 (0.60 to 1.46)Fletcher 94 3/41 4/47 0.90 0.85 (0.18 to 3.97)SCRIP 26/145 44/155 7.42 0.56 (0.33 to 0.95)Taylor 97 13/293 10/292 3.06 1.31 (0.57 to 3.01)Carlsson 98 CABG 0/33 0/34 Not estimableCarlsson 98 AMI 2/85 2/83 0.54 0.98 (0.14 to 7.05)Lifestyle Heart 14/53 24/40 3.07 0.25 (0.11 to 0.58)Bell 99 8/102 8/102 2.05 1.00 (0.36 to 2.77)Total (95% CI) 2566 2535 100.00 0.81 (0.70 to 0.93)Total events: 453 (treatment), 518 (control)Test for heterogeneity: χ 2 = 34.08, df = 27 (p = 0.16), I 2 = 20.8%Test for overall effect: z = 2.88 (p = 0.004)0.1 0.2 0.5 1 2 5 10Favours treatment Favours controlFIGURE 1 RCTs of the effects of cardiac rehabilitation. Source: Jollife et al., 2002 10 with permission of The Cochrane Library.CI, confidence interval; MI, myocardial infarction; OR, odds ratios; WHO, World Health Organization.normal activities. 38,39 Heart failure patients areless likely to be referred for cardiac rehabilitationthan other cardiac patients 40 and the complexityof the medical condition is identified as a barrierto physician referral. 41 In the UK few programmesrecruit heart failure patients, possibly reflectingthe perceived need for further evaluation ofeffectiveness and safety in this patient group. 42 InEngland, provision for both heart failure andangina may be limited by the priorities identifiedin the NSF-CHD: “once Trusts have an effectivesystem recruiting people who have survived amyocardial infarction or who have undergonecoronary revascularisation to high quality cardiacrehabilitation, they should extend theirrehabilitation services to people admitted tohospital with other manifestations of coronaryheart disease, e.g. angina and heart failure.” 12Patients with chronic non-cardiac medicaldisorders may be excluded from cardiacrehabilitation. 43 Medical reasons for non-invitationinclude impaired mobility, more severe angina andperipheral arterial disease, 29 chronic obstructivepulmonary disease and asthma, 39 arthritis andback problems, 44 and alcohol addiction. 453© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

BackgroundTABLE 1 Uptake of cardiac rehabilitation in recent UK surveysAuthor Year of Region Study design Total no. of Participation assurvey eligible patients percentage ofeligible patientsEvans et al., 2002 25 2000 UK CR programme 208,080 total 17% MIsurvey compared (calculated from 44% CABGwith BHF statistics percentages) 6% PTCABethell et al., 2001 26 1997 UK CR programme 150,000 total 14–23% MIsurvey compared33–56% CABGwith BHF statistics6–10% PTCAMelville et al., 1999 27 1996 Nottingham CR enrolment lists 261 43% MIcompared withhospital dischargeCampbell et al., 1996 28 1994 Scotland CR programme 29,294 (calculated 17% CHDsurvey compared from 4980 =with CHD survival 17% of total)Pell et al., 1996 29 1994 Glasgow CR department 887 21% MIlists compared(12% completed)with hospital dischargeBHF, British Heart Foundation; CHD, coronary heart disease; CR, cardiac rehabilitation.4Rehospitalisation, health deterioration andplacement in a nursing home are also associatedwith reduced participation in cardiacrehabilitation. 46 Patients with communicationdifficulties including short-term memory loss andconfusion, poor cognitive functioning orneurological impairment may be less likely toparticipate in cardiac rehabilitation, 39,46–48 and agreater dropout rate has been observed in patientswith symptoms of depression. 49Older patients may not receive the same amountof advice from physicians on cardiac risk reductionas younger patients. 46 Invitation to cardiacrehabilitation is often lower in olderpatients. 3,29,32,33,39,41,43,46,49–52 In a US survey olderpatients expressed a preference for home-basedprogrammes, whereas younger patients preferredcomprehensive clinic-based programmes. 53Women tend to be under-represented in cardiacrehabilitation. 20,32,33,43,54 Referral rates may belower, 33,43 possibly reflecting the increased age ofwomen presenting with cardiovascular disease 55and the presence of co-morbid conditions. 20Women may be reluctant to participate in formalcardiac rehabilitation 56 and perceived as lessmotivated to attend structured programmes withstrenuous exercise. 23 However, rehabilitationprofessionals may seem less helpful and lessencouraging in promoting cardiac rehabilitationfor women. 57 Invitation to a predominantly maleexercise group may also serve to discourageparticipation by women. 57Participation rates of patients living in areas of highsocial deprivation are low, probably reflectingreduced uptake rather than referral. 27,29,58,59Patients with no paid employment may also be lesslikely to attend a cardiac rehabilitation programme. 59In a survey of Canadian cardiac rehabilitationprogrammes participation by non-Englishspeakingpatients was seen to be considerablylower than by English-speaking patients. 39 Nosimilar surveys have been published in the UK,but a retrospective hospital audit found lowattendance at cardiac rehabilitation amongpatients of South Asian origin. 60 This wasattributed to poor access and inadequate use ofinterpreting services by patients and staff, and lackof translated written information.Barriers to uptake and adherenceCardiac rehabilitation should be accessible andacceptable to patients. A balance must be achievedbetween a programme of sufficient intensity andduration to be effective, and the tendency of along programme to encourage dependence insome and dropout in others. 61 Many patients

Health Technology Assessment 2004; Vol. 8: No. 41make recommended lifestyle changes, but othersmake no change or find it difficult to maintainnew behaviours. 62 The initial improvements inexercise tolerance and psychosocial well-beingobserved in some trials are not evident over thelonger term and this has been attributed toreductions in compliance. 63 Patients and providershave identified numerous possible reasons for lowlevels of uptake, adherence and professionalcompliance with cardiac rehabilitation. 64Some patients show a lack of interest and arereluctant to change their lifestyle. 14,41,46,59,65,66Affective reactions to disease can lead tomaladaptive responses and fear. 44,66,67 The patientmay not perceive that they will benefit fromparticipating in a programme or may receivecontradictory advice from other sources. 41,51,68,69Conversely, after a short period of rehabilitationpatients may be satisfied and choose to continueindependently. 70 Patients may dislike classes or thehospital setting. 34,51Patterns of personal or family living can influenceparticipation in cardiac rehabilitation. 71Conflicting work or domestic commitments andtime conflicts are associated with reducedattendance at cardiac rehabilitation. 24,51,66,70 Lackof family support may be a barrier to uptake ofcardiac rehabilitation services. 44,66In the USA, reimbursement issues and cost ofrehabilitation services limit attendance at cardiacrehabilitation. 14,34 Patients with insurancecoverage for cardiac rehabilitation are more likelyto be referred and programme directors identifyfinancial issues as the major barrier forrehabilitation uptake. 37 Fee-for-service patients aremore likely to receive cardiac rehabilitation thanhealth maintenance organisation patients. 72 Arequirement for continuous ECG monitoringduring exercise sessions, physician evaluation oftraces and exercise prescription also limitsrehabilitation provision. 73Cardiologists may be more likely to refer patientsto cardiac rehabilitation than primary carephysicians. 14,29,50,66 Differing rates of referral mayreflect professional scepticism or a poorknowledge base about the effectiveness of cardiacrehabilitation 39,46,62 and it is possible thatphysicians recommend rehabilitation to youngerpatients or those expected to comply. 46,74 Thesource of referral may also influence patientattendance at cardiac rehabilitation, with physicianreferral and in particular that of a cardiologistshown to improve uptake. 29,39,40,75The location, convenience and accessibility of acardiac rehabilitation programme influenceattendance. 14,34,41,43,66,76 Patients living closer tothe programme are more likely to receive areferral and attend. 37,52,65 Patients living in citiesor urban areas are more likely to attend cardiacrehabilitation. 39,77 Inconvenient transportation,lack of and cost of transport, and parkingproblems are frequently cited as barriers toattendance at cardiac rehabilitation. 34,51,70Barriers to uptake and adherence may besummarised as follows.Patient factors:● lack of interest● reluctance to change lifestyle● depression● dislike of classes/hospitals● work or domestic commitments● lack of family support● rural residence.Service factors:● cost and reimbursement● ECG monitoring requirement● location and accessibility● car parkingProfessional factors:● knowledge and attitudes● referral● prejudice (age, race, gender).Economic aspects of cardiacrehabilitationCosts of cardiac rehabilitation services vary byformat of delivery. The German approach tocardiac rehabilitation with 4–6 weeks of inpatientcare is estimated to cost about seven times that ofan outpatient service. 78 Information on the directcosts of outpatient cardiac rehabilitation asprovided in the UK is limited. The results ofrecent UK costs studies 25,79–82 are shown inTable 2.Comparison of studies is difficult as the authorsused different methodologies and sources of costestimates. The most recent BACR/BHF surveysuggests that cost varies widely, with a range of£50–712 per patient treated depending on thelevel of staffing, the equipment used and theintensity of the programme. 25 Staffing representsthe most important share, with estimates of64–80% of total direct costs. 79,815© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

BackgroundTABLE 2 Studies reporting costs of cardiac rehabilitation in the UKAuthor Year Type of programme Costs considered ResultsEvans et al., 2002 25 2000 Annual BACR/BHF survey: Staff costs, possibly some £50–712 per patientbudget statements from allowances for stationery (median £256)37 centres (2000 prices)Osika, 2001 82 1997/98 Based on four cardiac Staff costs, non-staff costs £292 per patient, rangerehabilitation centres in (not specified) £250–375 (1997–98 prices)GwentTaylor and Kirby, 1999 81 1995 One UK centre with Staff costs, equipment £140 per patient12-week programme costs, capital costs, £6 per patient per sessionwith two outpatient visits transport (1995 prices)Gray et al., 1997 80 1994 Survey of 16 UK centres Staff costs £371 per patient (medianwith an average of 9.2£223), £47 per patient persessions per patient session (median £26)(10.2 hours per patient) (1994 prices)Turner, 1993 79 1992 Based on ten cardiac Staff costs, overhead costs, £200 per patientrehabilitation programmes equipment costs, capital (1992 prices)in the Wessex region costsBACR, British Association for Cardiac Rehabilitation.Little information is available on the costs ofestablishing or expanding a rehabilitation service.Previously, it was considered that the resourcesneeded to establish a cardiac rehabilitationprogramme were present in most districthospitals. 3,83 With changes in healthcaremanagement and increasing demands on facilitiesand space this may not now be the case.ConclusionsOutpatient cardiac rehabilitation should beavailable to patients with a range of cardiovasculardiagnoses and after revascularisation procedures.Previous surveys have suggested that uptake ofoutpatient cardiac rehabilitation services is low,with specific patient groups under-represented. Toquantify the shortfall in cardiac rehabilitationservice availability and uptake, estimates of currentUK need and provision are required.Barriers to participation in outpatient cardiacrehabilitation have been identified, but theeffectiveness of interventions to improve uptakeand adherence has not been assessed by systematicreview. Such a review is needed to identifyappropriate methods for increasing service useand to suggest areas meriting furtherresearch.Previous economic evaluations of cardiacrehabilitation services have given a wide range ofcost estimates and little information on costs otherthan those attributable to staffing. A thoroughassessment of current UK costs of services isneeded to include staff, overhead, equipment andcapital costs. Furthermore, if greater numbers ofpatients are to receive outpatient cardiacrehabilitation an estimate of the cost implicationsof increasing provision by the establishment ofnew or expansion of existing services isrequired.6

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 2ObjectivesThe questions posed in this project are asfollows.●●●●What is the population need for cardiacrehabilitation?Who is not receiving cardiac rehabilitation?What is the effectiveness of different methods ofimproving uptake and of differential targetingof cardiac rehabilitation?What is the potential budget impact ofincreasing uptake of cardiac rehabilitation usingdifferent uptake interventions?The questions will be tackled using the followingsources of information:●●population need for cardiac rehabilitation inthe UK from analyses of the English HospitalEpisode Statistics (HES) and equivalent nationaldatabasesprovision and uptake of cardiac rehabilitation inthe UK by means of a national survey of cardiacrehabilitation services, ad hoc surveys andaudits●●●uptake and adherence to cardiac rehabilitationfrom a recent multicentre RCTa systematic literature review of interventions toincrease patient uptake, adherence andprofessional compliance with cardiacrehabilitationthe costs associated with improving uptake anddifferential targeting of cardiac rehabilitationfrom the national survey, systematic review andcosting data from sampled cardiac rehabilitationprogrammes.Improving uptake of cardiac rehabilitationwas conceived in a series of related stages:need for rehabilitation (in terms of ability tobenefit from rehabilitation), coverage ofexisting services; pattern (i.e. by age, gender,ethnicity) of referral to services, and adherencein terms of both acceptance of invitation toattend services and completion of treatment.Interventions to improve uptake could beenvisaged for each of these stages. Thisprocess is shown schematically inFigure 2.Need for rehabilitationHESCoverageNational survey +economicReferralNational surveyUptake and adherenceAcceptance of invitation+Completion of treatmentAdherence in RCTSystematic reviewof literatureEconomic appraisalFIGURE 2 Improving the uptake of cardiac rehabilitation7© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 3Population need for cardiac rehabilitation in the UKObjectives●●Determination of the population need forcardiac rehabilitation in the UK by analysis ofthe English HES and similar national databases.Estimation of the level of uptake of cardiacrehabilitation by patients with a dischargediagnosis of coronary heart disease.BackgroundThe NSF-CHD states that every hospital shouldensure that more than 85% of people dischargedfrom hospital with a primary diagnosis of acutemyocardial infarction or after coronaryrevascularisation are offered cardiacrehabilitation. 12 When cardiac rehabilitation isavailable to these patients, the NSF-CHDrecommends that this service should beextended to patients with angina and heartfailure. However, there is only limitedinformation available on population need,that is, the total number of patients who maybenefit from cardiac rehabilitation and the currentnationwide level of service provision and patientuptake.MethodsData from the HES for England and similarsources for Wales [Patient Episode Database(PEDW)] and Northern Ireland [HospitalInpatient Systems (HIS)] were used to estimatethe need for cardiac rehabilitation, that is, thenumber of patients discharged from hospital whohave the capacity to benefit from this therapy.Scottish data were not available. Data fromEngland, Wales and Northern Ireland werecollected from 1 April 1999 to 31 March 2000and provide a comprehensive picture of thenumber of patients discharged from hospitalswith particular conditions. Information wascollected for all patients discharged alive fromhospital with a primary diagnosis of ischaemicheart disease [International Classification ofDiseases-10 (ICD-10) codes I20–I25].Furthermore, data for subcategories of thesepatients were collected:1. acute myocardial infarction (ICD-10 code I21)2. heart failure (ICD-10 code I50)3. unstable angina (ICD-10 code I20.0)4. CABG (OPCS-4 codes K40–K46)5. PTCA (OPCS-4 codes K49–K50)6. CABG patients with one or more of thefollowing discharge diagnoses or procedurecodes: acute myocardial infarction, unstableangina, heart failure or PTCA7. all other ischaemic heart disease cases.Categories 1–7 are mutually exclusive, so thatpatients are only recorded once using eitherdiagnosis or procedure codes. In case of multipleevents with the same code each patient was onlycounted once. When a person was admitted morethan once in a year, each extra admission wasincluded.The total number of patients eligible to receivecardiac rehabilitation was derived by adding thenumbers in categories 1–7. These data werestratified by gender and age groups. Populationstatistics 84 were used to derive rates per 100,000individuals.The uptake of cardiac rehabilitation by eligiblepatients was estimated. Data from the 2000BACR/BHF survey of cardiac rehabilitationservices and an additional short postalquestionnaire (as described in Chapter 4) wereused to obtain the number of services and toestimate the total number of patients referred andjoining outpatient cardiac rehabilitation inEngland, Wales, Scotland and Northern Ireland.In total, 284 centres were identified for the wholeof the UK in 2000 (220 centres in England, 36centres in Scotland, 18 centres in Wales and tencentres in Northern Ireland).Of these, 191 (67%) responded to the additionalquestionnaire. Where a centre had not responded,a value relating to the upper interquartile range(IQR) derived from the responding centres wasimputed and added to the aggregated figures ofthe responding centres to estimate the upperrange of service provision for England. The lowerrange of service provision was similarly estimatedby imputing, where data were missing for centres,the lower IQR derived from the responding9© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Population need for cardiac rehabilitation in the UKcentres. Uptake was estimated from the number ofeligible patients (using different need criteriagiven above) and the estimated number ofpatients referred, joining and completing cardiacrehabilitation. These estimates were then linkedwith the population need data.Analyses were undertaken to estimate the level ofuptake with different criteria of eligibility forcardiac rehabilitation:●●●All patients with the above-mentioned dischargediagnoses and procedure codes were consideredeligible.Only patients with acute myocardial infarction,unstable angina, CABG and PTCA wereconsidered eligible.Only patients younger than 75 years wereconsidered eligible.The last two analyses were conducted bytruncating the population data using thesespecified criteria.ResultsBased on hospital discharge statistics it wasestimated that the total numbers of hospitaldischarged patients potentially eligible to receivecardiac rehabilitation in 2000 were as follows:England 266,833; Wales 17,560 and NorthernIreland 13,988. Total counts of discharged caseswith acute myocardial infarction, heart failure,unstable angina, CABG and PTCA stratified bycountry, gender, age group and rates of dischargediagnoses and procedure code per 100,000 personsare presented in Appendix 1 (Tables 35–37).Table 3 shows the estimated number of patientsreferred to and joining outpatient cardiacrehabilitation programmes by country. Moredetailed information is shown in Appendix 2(Tables 38–41). Age, gender and diagnosis-specificestimates of need could not be provided as only aminority of cardiac rehabilitation centres were ableto supply relevant information broken down by thevariables required to link estimates of need toservice supply.Similar proportions of all eligible patients inEngland and Wales were referred to cardiacrehabilitation (between 22 and 36%). However, theproportion of referred patients in NorthernIreland was significantly less (12–17%). Theproportions of all eligible patients joining cardiacrehabilitation programmes in England and Waleswere also similar (13–21%), but joining was lesscommon in Northern Ireland (9–12%).Using more limited criteria of need for cardiacrehabilitation considering only patients dischargedwith a diagnosis of acute myocardial infarction,unstable angina or a procedure code of CABG orPTCA as eligible, under-provision was lessmarked, with about 45–67% referred to and27–41% joining cardiac programmes in England(see Table 3 for other countries).TABLE 3 Estimated uptake of cardiac rehabilitation by patients with different manifestations of coronary heart diseaseEngland Wales Scotland NorthernIrelandEstimated number of patients referred to CR a 59,400–87,200 4,600–6,400 5,800–9,100 1,700–2,400Estimated number of patients joining CR a 35,700–53,100 3,000–3,600 3,500–6,000 1,200–1,700Eligibility criteriaAll patients 266,800 17,700 NA 14,000% referred to CR 22–33% 26–36% NA 12–17%% joining CR 13–20% 17–21% NA 9–12%Patients with AMI, unstable angina, CABG 131,100 7,900 NA 6,800and PTCA% referred to CR 45–67% 59–81% NA 25–36%% joining CR 27–41% 38–46% NA 18–25%Patients

Health Technology Assessment 2004; Vol. 8: No. 41Provision was also estimated considering patientsunder age 75 years as eligible. This analysissuggests that 30–43% of patients were referredand 18–26% joined cardiac rehabilitation inEngland (see Table 3 for other countries). Adetailed summary of this analysis is displayed inAppendix 2 (Tables 38–41).DiscussionThe objective of this analysis was to estimate thepopulation need for cardiac rehabilitation and toprovide up-to-date information about the level ofuptake of cardiac rehabilitation in the UK. Thisinformation should assist healthcare policy makersto improve the provision of cardiac rehabilitationservices to all patients who have the capacity tobenefit.The analysis suggests that provision of cardiacrehabilitation at the inception of the NSF-CHDwas low. This was still apparent when consideringonly patients with acute myocardial infarction,unstable angina, PTCA and CABG as eligible or,in a second analysis, only patients younger than75 years.There appears to be variation in service provisionacross the UK, with a higher proportion ofeligible patients referred to and joining cardiacrehabilitation programmes in England andWales than in Northern Ireland. Since the needfor rehabilitation is substantially greater inNorthern Ireland (and Scotland), this representsa considerable mismatch between uptake andneed.Although a different approach was used toestimate the level of service provision, the presentfindings are in concordance with previous researchexamining the relationship between need andsupply. Bethell and colleagues estimated thatbetween 14 and 23% of myocardial infarctionpatients, between 33 and 56% of CABG patients,and between 6 and 10% of PTCA patientsattended cardiac rehabilitation in 1997. 26 Themost recent update provided by the same groupsuggests that 17% of all myocardial infarction, 44%of all CAGB and 6% of all PTCA patients receivedcardiac rehabilitation in 2000. 25 It should beemphasised, however, that this estimate was basedon only 69% of all UK centres. The true level ofprovision may be higher if non-participatingcentres were providing a service with betterreferral and joining rates, but this seemsimprobable.The analysis presented here illustrates the lack ofcomprehensive and reliable data to estimate thelevel of service provision and should beinterpreted with some caution. By using data fromthe HES a number of assumptions had to be madeto estimate need. Although patients managed athome or in the private sector will be missed, itmay be assumed that the HES are complete and aprimary diagnosis of ischaemic heart diseaseindicates a need for cardiac rehabilitation.Furthermore, it is assumed that each finishedconsultant episode for these diagnoses equates toone person; the ratio of spells to finishedconsultant episodes is generally around one. 85However, the number of discharge diagnoses maybe slightly higher than the number of patientsbecause in some instances myocardial infarctionpatients receive revascularisation procedures suchas CABG or PTCA within a few weeks. The timebetween these two distinctive admissions may notbe sufficient for enrolment in a rehabilitationprogramme after the first event. The estimatesassume that a patient suffering two or more eventsin a year represents a need for two (or more)courses of rehabilitation. This seems legitimate assuch patients may be considered to berehabilitation ‘failures’, and may have slippedthrough the net on earlier occasions.Another potential limitation is the approachadopted to estimate the current level of serviceprovision. These estimates are based on a postalsurvey with a response rate of 67% of the samplingframe of all cardiac rehabilitation centres existingin 2000 in the whole of the UK. Approximately80% of these centres could provide data for thenumber of patients referred to and joining cardiacrehabilitation programmes and therefore theestimates are based on a sample of about 55% ofall UK centres. However, by imputing the IQR totake account of missing data, the resultingestimates should provide a fair estimate of thecurrent situation.The apparent inability of centres to providecomprehensive activity data is possibly due to thelack of automated systems to extract these data,lack of audit facilities or centres being in theprocess of installing systems to collect audit datato satisfy the requirements of the NSF-CHD.Therefore, the current level of service provisioncould only be estimated indirectly by assumingthat all patients with a primary diagnosis ofischaemic heart disease are eligible. Some limitingcriteria of need were also used, namely restrictingthe eligibility for cardiac rehabilitation to certain11© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Population need for cardiac rehabilitation in the UKgroups of patients (acute myocardial infarction,unstable angina, CABG and PTCA patients, andall ischaemic heart disease patients younger than75 years). A more appropriate approach would beto obtain information on the number of patientsreferred to, joining and completing programmesstratified by gender, age and discharge diagnosisdirect from cardiac rehabilitation centres andrelate these to data that represent need, such asthe HES, or to information obtained fromhospitals in the catchment area of therehabilitation service by means of comprehensivecoronary heart disease registers.It was not possible to assess the level of uptake ofcardiac rehabilitation by patients of ethnicminority groups for two reasons. First, nationalhospital data stratified by ethnicity were onlyavailable for England. These were not completelycoded for ethnicity, with about 30% missing data.Second, as reported in Chapter 4, the majority ofcentres in the BACR/BHF survey were not able toprovide data on the referral and uptake of cardiacrehabilitation by ethnic minority groups.ConclusionsThe analysis suggests that the level of serviceprovision of cardiac rehabilitation during 2000 waslow. Therefore, the achievement of the NSF-CHDgoal of 85% of acute myocardial infarction andrevascularisation patients receiving cardiacrehabilitation is far from fulfilled. In addition, theshortcomings of this analysis clearly emphasise theneed for a more comprehensive data collection toestimate reliably the provision of cardiacrehabilitation services and its relationship to need.12

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 4Provision and uptake of cardiac rehabilitationin the UK: national survey of UK cardiacrehabilitation servicesObjective●Assessment of the provision and uptake ofcardiac rehabilitation in the UK by means of anational survey of cardiac rehabilitationservices.MethodsBACR, with financial backing from the BHF, hasconducted several surveys of cardiac rehabilitationservices in the UK. 25,26,86 The authors of thesereports have kindly provided this group with thedata that they collected, and contact details of allservices identified for the latest survey conductedin 2001, which included data from 1 January to31 December 2000. This latest survey includedquestions concerning the total number of patientsreferred, joining and completing outpatient(phase 3) cardiac rehabilitation, the numbersbroken down by diagnosis of myocardial infarctionor cardiac surgery and by age groups and gender,time spent per week for each programme byvarious staff members, current funding andquestions relating to outcome measures.For the purposes of the current project there wasalso a need to know the numbers of patients fromtraditionally under-represented groups (women,the elderly, people from ethnic minority groupsand people with heart failure or angina) referred,joining and completing cardiac rehabilitationprogrammes to be able to determine currentservice provision in these groups. There was alsoan interest in knowing how many services activelypromoted adherence to programmes in theseunder-represented groups, and details of whatinterventions were used to achieve this. In termsof the extent of coverage and level of serviceprovision, questions were asked for all patients andthe study also sought to determine whetherservices had spare capacity for additional patients.A short postal questionnaire was devisedspecifically addressing these issues and sent to allthose respondents of the 2000 BACR/BHF survey.ResultsBy contacting the cardiac rehabilitation liaisonperson for each local health authority in the UK,284 cardiac rehabilitation services were identifiedin 2000. Of these, 242 services responded to theBACR/BHF questionnaire, giving a response rateof 85%. The additional short postal questionnairedevised for the purposes of the current project(see letter of request and questionnaire inAppendix 3) was then sent to those respondents ofthe original survey, asking for information duringthe same period (1 January to 31 December 2000)so that data from the two sets of questionnairescould be linked. The response rate to thisadditional questionnaire following telephoneprompting was 79% (191 questionnaires returned).Data returned were entered into a MicrosoftAccess database and transferred to STATA (Version7) for data cleaning and analysis. Data arepresented as proportions, medians, IQR andrange, or means and standard deviations (SD).Numbers of patients referred to,joining and completing cardiacrehabilitation programmes in 2000Most services were able to provide this information,as shown by the relatively high number ofresponders in Table 4 (maximum n = 191). Of thetotal number of patients referred, two-thirds ofpatients actually joined cardiac rehabilitationprogrammes and only half of those referredcompleted the programme. The number ofpatients attending individual programmes variedwidely across the UK, as shown by the large ranges.Capacity to increase provisionThirty-one of 191 centres (16.2%) stated that theyhad spare capacity within their service, and couldaccommodate a median of four (two to 20) extrapatients each week.Level of service provision across the UKin 2000The content of outpatient cardiac rehabilitationprogrammes was determined by the duration and13© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Provision and uptake of cardiac rehabilitation in the UK: national survey of UK cardiac rehabilitation servicesTABLE 4 Overall referral, uptake and completion rates for UK cardiac rehabilitation programmes in 2000Median IQR Range No. of programmes % of referralsNo. referred per centre 271 164–424 2–1564 156No. joined per centre 172 101–254 2–1066 153 63%No. completed per centre 130 75–186 3–450 133 48%TABLE 5 Level of service provision for cardiac rehabilitation programmes in 2000Weighted a mean (SD) Range No. b of programmes providing dataExerciseNo. of weeks 7.4 (2.1) 1–12 144No. of sessions per week 1.7 (1.5) 1–14 146Average length of sessions (h) 1.2 (0.4) 0.5–3 145Total time spent by patient (h) c 12.9 (1.7) 3–98 143Average no. of patients per session 15.7 (6.2) 1–50 139Health educationNo. of weeks 6.2 (2.1) 1–12 141No. of sessions per week 1.3 (1.6) 0.25–14 141Average length of sessions (h) 1.0 (0.4) 0.25–3 139Total time spent by patient (h) c 7.0 (2.0) 0.75–98 139Average no. of patients per session 16.1 (6.8) 1–40 132Psychological interventionNo. of weeks 5.0 (2.7) 1–12 126No. of sessions per week 1.3 (1.5) 1–7 123Average length of sessions (h) 0.8 (0.5) 0.17–2 122Total time spent by patient (h) c 3.2 (2.0) 0.5–16 119Average no. of patients per session 14.6 (6.1) 1–40 116a Weighted by the size of the service (number of patients who joined). Data not normally distributed were transformedbefore weighting.b Data were not provided for all questions by all services, so the numbers of respondents to each question are provided.c Calculated as the number of weeks multiplied by the number of sessions per week multiplied by the duration of thesession in hours.14the number and length of sessions for each of thecomponent parts: exercise, health education andpsychological interventions (stress managementand relaxation).The mean values across services have beenweighted by the number of patients joining eachprogramme. Again, there was a reasonableresponse rate to these questions, as shown by therelatively high numbers who provided data. Table 5highlights just how variable the programmecontent and intensity of each intervention is acrossthe UK. Overall, exercise is the dominantcomponent, with the total time spent by a patientalmost twice that of health education and fourtimes that of psychological interventions. Thisreflects the origins of cardiac rehabilitation, theweight of evidence for benefit from exercise-basedprogrammes and the expertise of the principalmembers of most rehabilitation programmes.Under-represented groups: number ofreferrals, joiners and completersbroken down by age, gender, diagnosisand ethnicityResponse rates to questions on numbers ofpatients referred to, joining and completingprogrammes from under-represented groups weremuch poorer. Reported reasons included lack ofautomated systems and audit facilities, or thatcentres were in the process of installing systems tocollect audit data to satisfy the requirements of theNSF-CHD. 12 The representativeness of Table 6should therefore be interpreted with some caution.The numbers of patients with heart failure orangina, or from ethnic minority groups, were sosmall that it was not possible to look at theproportions of those referred, joining andcompleting rehabilitation. Similar proportions ofjoiners and completers relative to those referredwere seen for postmyocardial infarction patients

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 6 Under-represented groups: referral, uptake and completion rates for UK cardiac rehabilitation programmes in 2000Median per IQR Range n a % of referralscentre per yearNo. of male patients referred 213 111–334 2–1066 83joined 118 66–185 2–747 84 55%completed 84 43–154 2–329 65 39%No. of female patients referred 85 36–130 1–498 83joined 36 17–60 1–319 84 42%completed 27 12–45 0–140 65 32%No. of patients aged >65 referred 142 61–228 0–887 66joined 72 37–152 0–596 71 51%completed 54 30–110 4–212 51 38%No. of black/Asian patients referred 5 1–19 0–196 59joined 2 0–7 0–127 63completedNo. of post-MI patients referred 160 78–286 0–881 97joined 91 49–149 0–446 88 57%completed 66 31–103 0–425 69 41%No. of CABG patients referred 86 47–142 0–563 91joined 50 22–99 0–407 83 58%completed 45 13–82 0–367 65 52%No. of HF patients referred 0 0–2 0–28 61joined 0 0–1 0–12 59completed 0 0–1 0–9 46No. of angina patients referred 6 0–27 0–200 71joined 1 0–8 0–134 70completed 0 0–5 0–73a Data were not provided for all questions by all services, so the numbers of respondents to each question are provided.HF, heart failure.(not an under-represented group, here only forcomparison), the over 65-year-olds and malepatients, with slightly fewer women joiningrehabilitation programmes relative to thosereferred. The number of patients post-CABGreferred for cardiac rehabilitation shows relativelyhigher rates of completion than other groups.Efforts to promote attendance inunder-represented groupsFinally each service was asked whether they madeany special efforts to promote adherence tocardiac rehabilitation programmes in each of theunder-represented groups, and to detail anyinterventions that they used to achieve this. Ofthose services (126/191, 66%) that indicated thatthey promoted attendance in at least one of theunder-represented groups, 46% stated that theypromoted attendance in women, 48% in theelderly, 55% in revascularisation patients, 34% inethnic minority groups, and 17% and 18% inpatients with heart failure and angina, respectively.Of the 126 services that stated that they promotedattendance in under-represented groups, 97provided details of the interventions that theyused to achieve this. A member of the report teamwith extensive clinical experience of cardiacrehabilitation delivery examined these free textresponses. Among the under-represented groupsthere emerged themes of a variety of interventionsthat were being, or could be used generically,across the different patient groups, and some thatwere definitely more specific to each of theparticular groups. These are presented in Table 7.The numbers in parentheses refer to the numbersof services which described each particularintervention. The majority of services that statedthat they promoted adherence did so in a way thatwould benefit most patient groups; for example,follow-up phone calls, free transport, home visitsand personalised invitations. Of those interventionsthat were specific to under-represented groups,individualised classes, appropriate ‘buddy’ systems,attendance of relative or spouse were among thosemost commonly stated. Direct referrals fromsurgery and specialist clinics were also used asmethods to ensure uptake and adherence.ConclusionsAlthough it is feasible to obtain useful informationabout means of improving uptake and adherenceusing ad hoc postal questionnaires, routineelectronic audit data are likely to provide a more15© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Provision and uptake of cardiac rehabilitation in the UK: national survey of UK cardiac rehabilitation servicesTABLE 7 Interventions used by cardiac rehabilitation programmes to improve uptake and adherence (number of programmes reportingindicated intervention)Any intervention (97)Generic interventionsFollow-up telephone call post-discharge (69)Preassessment clinic appointment and individualisedcoronary heart disease advice (58)Free organised transport (51)Home visit by specialist cardiac/BHF liaison nurse (43)Personalised invitation by letter or telephone to attend (42)Inpatient follow-up and verbal explanation (28)Non-attenders followed up and offered furtherappointments (26)Range and choice of menu options for classes (13)Community GP and practice nurse encourageattendance (5)Choice of sessions offered (venue/day/time) (5)Anxious patients met at the entrance of the venue (1)Travel grants and transport-sharing scheme (1)Invitation letter marketed and evaluated to encourageuptake of classes (1)Specific interventions for womenWomen patients ‘buddy’ system (15)Individualised exercise plans (14)Separate classes for women (6)Encouragement of husband or friend to attend (6)Choice of community or hospital-site sessions (3)Female volunteer befriending service and help-line (3)Focus groups to assess women’s needs (2)Smaller exercise groups for women (1)Health benefits for women explained (1)Women’s changing facilities (1)Female-only staff to facilitate rehabilitation sessions (1)Specific interventions for age > 65 yearsSeparate and smaller classes for the elderly/frail (6)Flexible start date if patient slow to recover (4)Elderly patients’ buddy system (3)Relative /spouse encouraged to attend (3)Lower impact exercise class (3)Choice of sessions offered (day and times) (3)Focus groups to assess elderly needs (2)Elderly volunteer befriending service and help-line (2)One-to-one exercise supervision (1)Elderly education sessions (1)Audiotapes of education sessions (1)Specific interventions for ethnic minority groupsAsian relative/friend encouraged to attend (8)Coronary heart disease leaflets in Asian languages (5)Audiotapes of education sessions (3)Asian-speaking nurses for home visits (5), education andexercise (3)Involvement of Asian support groups (3)Community elders from voluntary sector supportingrehabilitation (3)Asian education programme (3)Asian patient buddy system (2)Provision of culturally sensitive classes (2)Regular Asian focus groups to assess need (1)Separate exercise class for Asian women (1)Encouragement to wear traditional dress (1)Specific interventions for CABG/PTCASurgical tertiary centre referral system (13)Specific revascularisation programme led by arevascularisation rehabilitation nurse (12)Strong recommendation by surgeon/consultant (2)Theatre list referral system (1)Buddy system (1)Preangiogram talk about rehabilitation (1)Video about cardiac rehabilitation (1)Specific interventions for anginaSpecific angina education sessions (5)No exclusion to attend (5)Direct referral from rapid-access chest pain clinic (2)Referral while awaiting CABG (1)Buddy system (2)Referral followed up by specialist angina nurse (1)Specific interventions for heart failureSpecific heart failure programme (9)Community specialist heart failure nurse encouragesattendance (5)No exclusion to classes (4)Low-impact exercise classes (2)Buddy system (1)Community-based programme (1)Audiotapes of health education provided (1)One-to-one exercise supervision (1)16comprehensive picture, and more accurate data onreferral and uptake.Relative to post-myocardial infarction patients,older people and women tended to be less oftenreferred and were less likely to join a programme.Data on ethnic minorities and those with diagnosesof angina and heart failure were too sparse toevaluate formally. However, the low numbersreported indicate that these groups are veryunlikely to be referred or to join programmes.Many different interventions are reported byservices, suggesting high levels of awarenessof the general problem of uptake. Theseinterventions vary in complexity and cost; forthose that are either complex or costly, moreformal evaluation of their effects on uptakeand adherence would be valuable. Examplesof low-cost, sensible good practice (e.g.telephone call follow-ups) should be widelydisseminated and would not require formalevaluation.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 5Audit of cardiac rehabilitation in England:National Service Framework for Coronary HeartDisease recommendationsObjective●Assessment of cardiac rehabilitation auditactivity by survey of rehabilitation centres.submit a copy were contacted a second time bytelephone and subsequently by letter. The medicaldirector of the NHS trust was then contacted,asking the hospital trust to follow-up the request.BackgroundPatient uptake of outpatient cardiac rehabilitationservices is poor, particularly by under-representedgroups including women, the elderly and ethnicminorities. 25,33,36 Although guidelines on provisionof services exist, 2,4,87 audit of cardiac rehabilitationservices has previously been piecemeal and notroutinely undertaken and, where data exist,adherence to guidelines is poor. 88 In England, theNSF-CHD has recognised the benefits ofcomprehensive cardiac rehabilitation and the needfor services to be extended. 12 It states that everyhospital should ensure that more than 85% ofpatients discharged with a primary diagnosis ofacute myocardial infarction or after coronaryrevascularisation are offered access to cardiacrehabilitation. This has implications for clinicalgovernance and the need to audit cardiacrehabilitation services.In view of the NSF stated objectives, the aim wasto ascertain the level of cardiac rehabilitation auditactivity in the south-west of England and areaswith high ethnic minority populations in Londonand the midlands.MethodsCardiac rehabilitation centres in the south-west ofEngland, London and the Midlands werecontacted by telephone and asked to supply areport on their most recent audit. Information onany special efforts to improve attendance byspecific patient groups (e.g. women, the elderlyand ethnic minorities) was also requested. Centresreporting no available data were asked to providereasons for not undertaking audit. Centres withaudit data or a report available but that did notResultsResponse rateFrom January to July 2002, 51/57 (89%) of cardiacrehabilitation centres that were approached weresuccessfully contacted. Twenty-six centres (46%)responded to the first telephone request. Furthercontact by telephone and letter led to replies fromone (2%) and eight centres (14%), respectively.Finally, after written communication with medicaldirectors, replies were received from a further 16centres (28%). Audit data were received from 24(42%) centres, nine (16%) reported that an audithad been undertaken but did not send it, and 18(32%) stated that no audit had ever beenundertaken. Two centres supplied their audit asanonymous individual patient data and one centrewas only able to provide an audit report limited toa single ethnic group.Audit methodsThe means of data collection varied betweencentres. It was not possible to determine themethod of data collection by 12 centres (50%). Ofthose where this was clear, six (50%) relied on a‘paper system’ with retrospective data extractionfrom patient notes and attendance registers, whilesix (50%) used regularly updated computeriseddatabases. Commenting on the collection of data,respondents regarded paper systems as timeconsuming,tedious and unreliable, while centresusing computerised methods reported thatfrequently there was a lack of trained staff for datamanagement.The mean length of audit was 10.4 months(SD 2.8, range 4–12 months). Times for datacollection also varied between centres. Mid-pointdates were in 2002 (two hospitals), 2001 (ten17© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Audit of cardiac rehabilitation in England: National Service Framework for Coronary Heart Disease recommendationsTABLE 8 Audit activity specified in NSF-CHDStated NSF-CHD recommendationNo. of centres collectinginformation (%) (n = 23)No. (%) of patients discharged from hospital after coronary revascularisation or with a 10 (43%)primary diagnosis of AMIDocumentation of arrangements for cardiac rehabilitation in discharge communication to GP 0 (0%)Information on gender of patients 20 (87%)Information on age of patients 18 (78%)Information on ethnic group of patients 16 (70%)No. recruited to cardiac rehabilitation 21 (91%)Outcome information: 1 year after discharge, regular physical activity of at least 30 min 2 (9%)duration on average five times a week, not smoking, body mass index < 30 kg/m 218hospitals), 2000 (eight hospitals), 1999 (twohospitals) and 1998 (one hospital). One audit didnot provide dates. The main reasons cited for notcollecting audit data were: time constraints, lack ofadequate resources and computing facilities, lackof appropriate personnel to input data, limited orno audit training, or lack of informationtechnology support for the audit process.National Service FrameworkThe number of centres collecting information asstated in the NSF-CHD for annual collection ispresented in Table 8. The audit with data on asingle ethnic group is not included.Of importance is that data received were often notcomparable. Regarding age, only six (26%) centresprovided age information adequate for theassessment of attendance by age group. This wassimilarly the case with ethnicity, where only five(22%) provided information that permittedcomparisons of cardiac rehabilitation uptake byethnic groups. This information was more likely tobe collected by centres in areas with high numbersof patients from ethnic minorities. Eleven out of13 centres (85%) from areas with high ethnicminorities collected information on provision forspecific ethnic minority groups, but in only four(31%) could this be used to assess some feature ofuptake. In areas with relatively low numbers ofpatients from ethnic minorities limitedinformation was reported by five out of 10 centres(50%), with only one centre collecting adequateinformation to assess differences in attendancerates (10%).Of the 19 centres supplying relevant information,12 (63%) provided rehabilitation for patients withmyocardial infarction, coronary bypass surgery,TABLE 9 Additional information included in auditsAdditional information collectedNo. of centrescollectinginformation (%)(n = 23)Patient reasons for non-attendance 12 (52%)Patient clinical history and risk factors 6 (26%)Secondary prevention outcomes 4 (17%)Patient opinions and satisfaction 2 (9%)Patient’s home postcode 2 (9%)Psychological morbidity 1 (4%)Exercise outcomes 1 (4%)Reasons for non-referral 1 (4%)Referrals by consultant 1 (4%)angioplasty and heart failure. Six (32%) wereexclusively for myocardial infarction patients andone (5%) for surgical patients.Audits also contained information not directlyrelevant to the objectives of the NSF. This issummarised in Table 9.Numbers of patients per yearThe annual baseline mean number of patientsdischarged alive from hospital and eligible forcardiac rehabilitation and the numbers of patientsreferred to, attending and completing cardiacrehabilitation are presented in Table 10 for allcentres and for those providing services to a highproportion of ethnic minorities.The proportion of discharged patients attendingrehabilitation was 35% (weighted by number ofpatients discharged, SD 12, range 14–54%) and ofthose referred or invited to cardiac rehabilitationattendance was 55% (weighted by number ofpatients invited, SD 12, range 35–80%). Seventy-

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 10 Numbers and proportions of patients discharged alive, referred/invited, attending and completing cardiac rehabilitationper yearMean no. (SD) Proportion of Proportion of Proportion ofof patients (range) discharged a referred/invited a attended a(no. of centres) (no. of centres) (no. of centres) (no. of centres)All centresDischarged patients 390 (182)(167–684) (n = 10)Referred/invited for cardiac 308 (223) 59% (n = 7)rehabilitation (62–1066) (n = 18)Attending cardiac rehabilitation 176 (110) 35% (n = 8) 55% (n = 16)(23–533) (n = 19)Completing cardiac rehabilitation 148 (53) 32% (n = 3) 48% (n = 8) 77% (n = 7)(66–233) (n = 8)Centres providing service to a high proportion of ethnic minoritiesDischarged patients 398 (184)(167–620) (n = 6)Referred/invited for cardiac 334 (262) 60% (n = 5)rehabilitation (62–1066) (n = 11)Attending cardiac rehabilitation 189 (135) 29% (n = 5) 57% (n = 10)(23–533) (n = 11)Completing cardiac rehabilitation 137 (58)(66–233) (n = 6) 37% (n = 2) 48% (n = 6) 79% (n = 5)a For centres providing complete information.seven per cent of patients (weighted by number ofpatients attending, SD 13, range 57–91%) attendinga programme subsequently completed it.The proportion of patients discharged whocompleted a programme was 32% (weighted bynumber of patients discharged, SD 6, range28–42%). However, this was based on informationfrom only three centres.In five centres providing a service to a highproportion of ethnic minorities the percentage ofdischarged patients referred was significantlylower than in three centres from other areassurveyed and which provided appropriate data(29% compared with 45%). Otherwise, theproportions of patients referred, attending andcompleting programmes were similar.Measures reported to improve patientattendance at outpatient cardiacrehabilitationEight centres (35%) reported a variety of measuresto improve attendance and these are summarisedin Table 11. Three of these measures concentratedon the uptake of ethnic minorities, one on womenpatients, but none on the elderly. Regrettably,information evaluating the success of thesemeasures was not available.TABLE 11 Measures taken to improve attendance at cardiacrehabilitationDiscussionNo. of centresreportingmeasures toimproveattendance (%)Community, non-hospital-based 5 (22%)programmeTranslator or interpreter 3 (13%)Evening programme 2 (9%)Community liaison or link worker 2 (9%)Women-only programme 1 (4%)Programme for specific ethnic group 1 (4%)Programme on days appropriate to 1 (4%)religious beliefsSatellite services in local hospitals 1 (4%)Audio information for visuallyimpaired 1 (4%)Clinical governance incorporates audit to ensurethat clinical care is up to date and effective. 89However, a commitment to the accuracy,appropriateness, completeness and analysis of19© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Audit of cardiac rehabilitation in England: National Service Framework for Coronary Heart Disease recommendations20healthcare information is required if judgementsabout clinical quality are to be made and theimpact of clinical governance is to be assessed. 90Major barriers to clinical audit are lack of resources,lack of expertise or support, and organisationaldifficulties. 91 This survey highlighted that aminority of centres was able to provide informationon outpatient cardiac rehabilitation audit, with onethirdof centres reporting that no audit wasavailable. Some centres reported that audit hadbeen conducted, but were not eager to disseminatethe information outside the hospital. This mayreflect the perceived disadvantages associated withclinical audit of diminished clinical ownership andhierarchical and territorial suspicions. 91 Althoughthe medical directors of NHS trusts were contactedfuture studies should consider methods to improvesharing of audit information.Nearly half of the audits provided some relevantinformation on clinical audit as specified in theNSF-CHD. However, information on potentiallyunder-represented groups was limited. To someextent the style and content of audit reportsprobably reflect local interests and concernsrelating to cardiac rehabilitation provision. Basicinformation on the initiating event in particular,and on referral, invitation, attendance andcompletion was collected in sporadic and nonstandardways. A few audit reports werecomprehensive, with comparison of total numbersof discharged patients and patient attendance andcompletion of outpatient cardiac rehabilitation. Toallow comparison of provision between centres andover time, a baseline figure of total initiatingevents is required, as well as information oninvitation to, and completion of, the programme.Examples of clinical audit tools have beenincluded in cardiac rehabilitation guidelines. 2,4However, with the exception of initiating event,these have been limited in their inclusion ofinformation on potential sources of underrepresentation.A more recent resource considersage, gender and ethnicity, 92 and is currently underevaluation. 93 Development and acceptance of acomprehensive, standard audit tool with flexibilityregarding local issues would be helpful for use infuture audits. It may be possible to merge this intoa hospital critical care pathway and routinelycollected Myocardial Infarction National AuditProject (MINAP) 94 data. If the targets laid down inthe NSF-CHD are to be met and the healthoutcomes are to be successful then the challengelies in the development of an effective anduncomplicated audit tool that can be appliednationally to serve all cardiac populations.A difficulty identified in several audits and fromcentres unable to provide information was thatpatients may be referred to a programme fromone or more hospitals or from one hospital toseveral different programmes. This complicatesthe audit of programmes in both urban and ruralsettings. In one city unable to provide auditinformation a group of hospitals reported theimminent introduction of a joint database.Where available, audits of outpatient cardiacrehabilitation varied considerably in style andcontent. Some were thorough documents coveringmany aspects of audit, whereas other centres hadbeen unable to prepare a formal document butwere able to provide raw data. Some audits wereprepared by staff trained in clinical audit, whereasothers were by less experienced staff or werestudent projects. Other facilitating factors foraudit include modern medical records systems,effective training, dedicated staff, protected time,structured programmes, and a shared dialoguebetween purchasers and providers. 91Only a minority of centres was able to providecomplete information on numbers of patientsreferred (seven centres) and who attended cardiacrehabilitation (eight centres) in relation tonumbers discharged. There was a suggestion thatreferral and invitation of patients were similar incentres providing services in areas with highproportions of ethnic minorities compared withcentres in other areas. However, the proportion ofpatients attending a programme was lower inareas with high ethnic minority populations.Changes to services and interventions to improveuptake of cardiac rehabilitation by ethnic minoritygroups may be indicated.A series of measures had been undertaken bycentres to help patients to participate inoutpatient cardiac rehabilitation. These rangedfrom holding classes in community settings and attimes to suit patients, to the establishment ofclasses dedicated to women or ethnic groups. Asinterventions may be of interest to other centres,evaluation by controlled trials or within areproducible audit framework would be valuablein determining their overall effectiveness inimproving attendance at outpatient cardiacrehabilitation.ConclusionsThe findings from this more detailed survey ofaudit activity complement those obtained from the

Health Technology Assessment 2004; Vol. 8: No. 41national survey of cardiac rehabilitation servicespresented in Chapter 4. The authors had hopedto be able to find interventions specific to underrepresentedgroups by focusing on services locatedin areas with relatively high proportions of blackand ethnic minorities and with rather more agedpopulations (the south-west of England). However,the quality of audit, the reports and the datacollected were insufficient to support robustinterpretation of the performance of theseservices.The findings highlight a national uncoordinatedapproach to audit data collection in England withlarge variations in methods and content despitethe standards set out in the NSF and cardiacrehabilitation guidelines. The use of modernmedical records systems, appropriate training fordedicated staff and dialogue between allcontributors to services is suggested. Developmentof a national and policy-driven standardisedaudit tool would facilitate the identification ofpatients by cardiac event and the following of allpatients through the cardiac rehabilitationprocess.Limited analysis of audit data suggests that uptakeof cardiac rehabilitation is particularly low in areaswith high proportions of ethnic minorities.Information on under-represented groups andlocal interests should be incorporated into auditdata collection in a standardised way so that futurecare can be targeted to the needs of the localcardiac population.Some cardiac rehabilitation programmes haveattempted to improve attendance with measuresappropriate for all patients or for specific groups.Evaluation and dissemination of information oneffective and ineffective interventions may helpother programmes to improve services and useresources appropriately.21© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Objective●To evaluate uptake and adherence using datafrom a recent multicentre RCT. 95IntroductionIn the UK provision of rehabilitation for patientsfollowing acute myocardial infarction is arequirement of the NHS-CHD 12 and comparableguidelines in Scotland and Wales. 2,18 Beforedischarge all patients should be invited toparticipate in a multidisciplinary cardiacrehabilitation and secondary preventionprogramme. Some patient groups are not thoughtto benefit from the exercise component of cardiacrehabilitation after acute myocardial infarction.Patients with more severe cardiac illness and thosewith psychiatric conditions that may compromisesafety are considered ineligible. 96 Pragmatically,patients with poor functional capacity, significantco-morbidity, frailty or confusion are not suitablefor outpatient-delivered rehabilitation.Consequently, these factors may influence referraland uptake in clinical practice.The uptake and adherence achieved in a clinicaltrial setting was examined because this shouldreflect the best that can be achieved in optimalroutine clinical practice. 95 It would certainly beunlikely that NSF targets representing a higherlevel of uptake and adherence than that seen in acontemporary trial would be feasible. Whereappropriate, data for all patients discharged aftermyocardial infarction were analysed. For issuesrelating to attendance only those patientsallocated to cardiac rehabilitation were considered.MethodsPatients were recruited in 18 typical acute generalhospitals in England and Wales. The trial protocolplanned for all potentially eligible myocardialinfarction patients to be identified onconfirmation of diagnosis. At discharge ineligibleChapter 6Uptake and adherence in a randomised controlledtrial of cardiac rehabilitation aftermyocardial infarctionpatients (significant co-morbidity, etc.) wereexcluded under protocol guidelines of minimalexclusions and reasons were recorded, usually by anominated coronary care unit nurse. Patientseligible for rehabilitation were advised of the trialin an introductory letter.Each patient was visited by a research interviewerapproximately 1 week after discharge. Patientswere given full details of the trial and, after beingasked for informed consent, answered the baselinestructured interview. Following entry into the trialand central blind randomisation, the names ofpatients allocated to cardiac rehabilitation weregiven to rehabilitation teams for invitation,treatment and follow-up as normal practice for theprogramme.There were two opportunities for patient selection:by hospital medical or nursing staff (before trialentry) according to criteria in protocol; or bycardiac rehabilitation staff (after randomisation).There were also three opportunities for refusal ofthe trial or rehabilitation by patients: whenadvised of the trial by hospital staff; after a fulldescription of the trial and informed consent bythe research interviewer; or when given the date,time and venue of their first rehabilitationappointment. Patients were interviewed after1 year and asked about their experiences ofcardiac rehabilitation.ResultsThe collection of names of potentially eligiblepatients and recording of clinical summaries werenot complete in all hospitals. Some progressivelyreduced the flow of forms of excluded patients.Consequently, analyses were undertaken bothfor all hospitals and for those hospitals in whichrecord-keeping was thought to be nearlycomplete.In total, 3264 potentially eligible patients wereidentified in the 18 hospitals. Of these, 1400 werein five hospitals with complete registration.23© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Uptake and adherence in a randomised controlled trial of cardiac rehabilitation after myocardial infarction1400 patients identified withacute myocardial infarction722 (52%) patients excludedfrom trial678 (48%) patients admitted to trialof cardiac rehabilitation95 (7%) patients did not participatein the trial for practical reasons(could not speak English, livedoutside area, living in nursing home)151 (11%) patients chose not toparticipate in trial (trial andrehabilitation refusals and electiverehabilitation requests)103 (7%) patients excludedfor reasons concerning the trial(previous attendance at cardiacrehabilitation, interviewer unable tocontact)373 (27%) patients excludedfor medical and relatedreasons1027 (73%) of patients eligible forcardiac rehabilitationFIGURE 3 Exclusions from trial hospitals reporting complete myocardial infarction registration24Figure 3 shows reasons for exclusion from the trialand overall eligibility for cardiac rehabilitation inthe hospitals with complete registration. Seventythreeper cent of patients had no medical reason(identified during hospital stay by medical ornursing staff) for not attending a programme ofcardiac rehabilitation. Thus, nearly three-quartersof patients discharged within 28 days followingmyocardial infarction were deemed eligible forcardiac rehabilitation.Follow-up interviews were completed for 959patients randomised to rehabilitation atapproximately 1 year, by when 75 patients haddied. Attendance information for a further91 patients was provided by rehabilitationcoordinators.Medical reasons for exclusion, identified duringthe hospital stay, are shown in Table 12. Some ofthese patients may have been eligible forrehabilitation at a later date or in a differenthospital (patients awaiting surgery, transferred toanother hospital, having extended hospital stay orbeing readmitted). Including these patients raiseseligibility from 73 to 81%.Patients excluded from rehabilitation for medicalreasons tended to be older (mean age 71.9 yearscompared with 64.6 years for eligible patients,p < 0.0001) and were more likely to be female(36.5% versus 21.9% males excluded, p < 0.0001).Table 13 shows patient exclusion in males andfemales in different age groups. There was a trendfor increasing exclusion in both men and women

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 12 Reasons for exclusion in hospitals with completepatient registrationNo. ofpatients (%)Admitted to trial 678 (48.4%)Practical, personal and trial exclusion 349 (24.9%)Medical exclusion:Significant co-morbidity 75 (5.4%)Frail or confused 154 (11.0%)Transferred to other hospital 75 (5.4%)>28 days in hospital 13 (0.9%)Readmitted within 28 days 16 (1.1%)Awaiting surgery 5 (0.4%)Uncooperative 7 (0.5%)Other 28 (2.0%)Total 1400with age. In multivariate analysis the associationbetween gender and exclusion was not significantafter adjustment for age.Excluded patients tended to have pre-existingcardiovascular disease (previous myocardialinfarction or angina) or more severe presentationof the index myocardial infarction (Table 14).However, previous hypertension was not associatedwith eligibility.In total, 2144 patients were entered into the trial. Ofthese, 1100 were allocated to cardiac rehabilitation.Attendance figures are shown in Table 15.Not all patients allocated by the trial torehabilitation were offered cardiac rehabilitation.At least 22% and possibly as many as 33% ofpatients considered eligible by medical or nursingstaff at time of discharge were not offeredrehabilitation by cardiac rehabilitation staff.Patients invited tended to be younger than thosenot invited (mean age 62.8 years compared with68.1 years) with a clear trend for non-invitation inolder age groups (Table 16).There was a tendency for women to be overlookedmore often than men (31% compared with 22%).In less elderly patients (under 70 years), 18% ofwomen were overlooked compared with 13% ofmen. However, the trend was not significant inmultivariate analysis.At interview, patients who had been invited tocardiac rehabilitation (n = 721) estimated thenumber of classes that they had attended and, ifappropriate, gave reasons for non-attendance ordropout. Overall, 78% of patients invited torehabilitation attended at least one session. Ofpatients aged 65 years or more 72% attended atleast one session compared with 82% of thoseyounger than 65 (p = 0.001). These data providesupport for the observation that older patients areless likely to attend cardiac rehabilitation thanyounger patients. There were no statisticallysignificant differences in initial uptake betweenmen and women (79% versus 74%).TABLE 13 Medical exclusions by age and gender (1349 patients with age and gender known)Age at MI (years) Male exclusions Female exclusions All exclusions(% potentially eligible) (% potentially eligible) (% potentially eligible)

Health Technology Assessment 2004; Vol. 8: No. 41sample of hospitals than the 18 trial hospitals, asimilar pattern is observed in the wholesample.It is not possible to differentiate trial refusals fromcardiac rehabilitation refusals. By refusing toparticipate in the trial, patients may be seekingcardiac rehabilitation or may be turning it down.However, in an audit of rehabilitation servicesboth these patient groups must be considered aseligible. Patients who were excluded from the trialfor methodological (trial) reasons may have beeneligible and contactable if given a definiteinvitation while in hospital. Patients excluded forthe practical reason of living outside the area mayhave been eligible for a cardiac rehabilitationprogramme local to their home. Indeed, it ispossible that these patients received rehabilitationelsewhere. At the time there were no specificinterventions to facilitate uptake by patients whospoke no English, in the study hospitals (specificprogrammes for non-English-speaking patientsmay have been introduced more recently). It isunlikely that they would have receivedrehabilitation elsewhere.The reasons reported by patients for nonattendanceat and for early dropout from a cardiacrehabilitation programme suggest that uptake maybe improved by addressing issues of motivationand the perceived relevance of rehabilitation tofuture well-being, minor co-morbidities orperceived illness, site and timing of sessions,transport and arrangement of care fordependants.ConclusionsMedical and nursing staff identified 73–81% ofpatients discharged from hospital after acutemyocardial infarction as being eligible for cardiacrehabilitation. Excluded patients tended to beolder, were more likely to have suffered fromangina or had a previous myocardial infarctionand showed more severe presentation ofcardiovascular disease. Reduced invitation andattendance of women was largely explained bytheir greater age at myocardial infarction. Theexperiences of patients invited to cardiacrehabilitation suggest that uptake may beimproved by addressing issues of motivation andthe perceived relevance of rehabilitation to futurewell-being, co-morbidities, site and time ofsessions, transport and arrangement of care fordependants.27© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 7Systematic review of interventions to improveuptake, adherence and professional compliance withcardiac rehabilitationChapters 7–10 present the methods andfindings for a series of related systematicreviews. The issue of improving uptake was splitinto three major questions: how can recruitment tocardiac rehabilitation be improved; how canpatients’ adherence to cardiac rehabilitation andmaintenance of lifestyle changes be improved; and,how can professionals be encouraged to complywith guidelines and good practice? The sources ofdata to answer these questions may overlap, asresearchers will not necessarily have conceivedtheir questions in the same form as the presentgroup has. With awareness of this, it was ensuredthat each pair of reviewers dealing with a specificquestion read source material with a view toidentifying potential relevance to other questions.Definitions●●●Uptake: patients attending any outpatientcardiac rehabilitation (i.e. successfulrecruitment).Adherence: patients attending all or majority ofoutpatient programme, or maintaining lifestylechanges associated with cardiac rehabilitation.Professional compliance: healthcareprofessionals complying with guidelines or goodpractice regarding invitation and support ofpatients’ cardiac rehabilitation.Objective●How effective are different methods forimproving uptake, adherence or professionalcompliance with cardiac rehabilitation?MethodsA systematic review of interventions to increaseuptake, patient adherence and professionalcompliance with cardiac rehabilitation isdescribed. This was supported by members of theCochrane Heart Group (KR, MB) who assistedwith designing search strategies and identifyingreports.Data sourcesA general search strategy was designed to identifyall studies relating to the uptake, adherence orcompliance with cardiac rehabilitation services.The choice of sources was intended to find bothpublished and unpublished studies (greyliterature). Details of terms used in the search aregiven in Appendix 4. The terms used were thosefor ‘heart disease’ together with terms for ‘cardiacrehabilitation’. A broad approach to rehabilitationterms was chosen to identify not only formalcardiac rehabilitation programmes but also nontraditionalprogrammes that could contribute tocardiac rehabilitation. Studies identified werefurther searched for terms relating to uptake,adherence, compliance and costs. Studymethodology terms were not included, as theintention was to find all studies irrespective ofmethodology used. No language restrictions wereapplied.The following databases were searched frominception (as appropriate) to June 2001:●●●●●●●●●●MEDLINE on OvidEMBASE on Ovidthe Cochrane Library (2001 Issue 2). Thisincludes the Cochrane Controlled TrialsRegister, Cochrane Database of SystematicReviews, Database of Reviews of Effectiveness(DARE), HTA Database and NHS EconomicEvaluation DatabaseCINAHL on OvidPsycINFO on BIDS Silverplatter WebSPIRSISI Web of Science and ISI ProceedingsECONLIT on Silverplatter WebSPIRSBritish Library InsideSIGLE (System for Information on GreyLiterature in Europe)HMIC (Health Management InformationConsortium database)29© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve uptake, adherence and professional compliance with cardiac rehabilitation●●COPAC (joint catalogue of CURL – theConsortium of University Research Libraries)National Research Register.compared with authors of published papers, whichwould tend to bias the information towards studieswith more favourable outcomes.30Additional searching of literature:●●●●●The Journal of Cardiopulmonary Rehabilitation,1990–2001, was handsearched.Coronary Health Care, 1997–2001, washandsearched.Abstracts from conference proceedings werehandsearched:– American Association of Cardiovascular andPulmonary Rehabilitation (AACVPR)– American College of Cardiology (ACC)– British Cardiac Society (BCS)– British Association for Cardiac Rehabilitation(BACR)– European Society of Cardiology (ESC)– International Network of Agencies for HealthTechnology Assessment (INAHTA)– Society for Social Medicine (SSM)– World Congress of Cardiology (WCC).The reference lists of relevant studies andreviews were scanned.Expert opinion was sought.Study selectionPreliminary literature searches suggested that onlya small number of RCTs would be found and thatnon-randomised studies would form an importantpart of the review. Consequently, all studiesreporting evaluations of interventions wereconsidered. A total of 3261 references wasidentified in the searches, and the title andabstract of each article were examined by at leastone reviewer. Articles were only rejected if thereviewer could determine from the title andabstract that the article was not a report of anintervention. When a paper could not be rejectedwith certainty, the full text of the article wasobtained for further evaluation. A total of 957references was identified as potentially relevantand acquired for more detailed consideration.There was concern about publication bias andtherefore special efforts were made to identifystudies that might report negative findings bysearching the grey literature, and handsearchingabstracts of scientific meetings. Many of theidentified interventions were found in the greyliterature, which tends to include studies reportinglower effectiveness than those published injournals. 97 No attempt was made to contactauthors of studies as a lower response rate wasanticipated for supplementary information fromauthors of conference abstracts and thesesAfter the discarding of purely descriptive reportstwo reviewers (from AB, KR, SE, MB, IG, FT andRW) assessed articles using a three-questioninclusion/exclusion form (Appendix 5). A thirdreviewer (SE or KR) resolved disagreements overinclusion/exclusion.Reports were included for data extraction if thefollowing criteria were met:●●●evaluation of intervention to increase uptake,patient adherence or professional compliance tocardiac rehabilitationpatients with myocardial infarction, CABG orPTCA, with heart failure or angina, or coronaryheart diseaseoutcomes relevant to the reviews, specificallynumbers attending and patient adherence tocardiac rehabilitation and its exercise, educationand lifestyle components.Data extractionOnce the decision had been made to includestudies in the review two reviewers independentlyabstracted the relevant data (data extraction formin Appendix 6). Data extracted included details ofpatients, intervention, study type, quality andresults. No attempt was made to contact authorsfor additional information.The quality of non-randomised studies wasrecorded in accordance with recent reviews,specifically information relating to selection bias,power and analysis. 98 No formal scale was used tocategorise quality, but features of individual studiesare presented in the results sections. The methodof group allocation, sample size, comparison ofgroup characteristics at baseline and concomitantservice changes independent of the interventionare used as the basis for quality assessment.AnalysisThe systematic review takes the form of threequalitative overviews. No attempt was made topool study results as the number of trials was smalland the study designs and interventions varied.Studies are grouped by quality of evidence. Thebest evidence comes from RCTs while nonrandomisedand before-and-after study designsprovide less reliable evidence. 99,100 Studies werecharacterised by type and size, participants,intervention, comparison group, principal andother outcomes, and authors’ conclusions.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 8Systematic review of interventions to improveuptake of cardiac rehabilitationBackgroundBarriers to attendance at outpatient cardiacrehabilitation have been identified, but theeffectiveness of interventions to improve uptakehas not been assessed by systematic review. Such areview is needed to identify appropriate methodsfor increasing patient use of services and tosuggest areas meriting further research. For thereview uptake was defined as any patientattendance at outpatient cardiac rehabilitation(i.e. successful recruitment).ResultsStudies included in review ofinterventions to improve uptake ofcardiac rehabilitationThe flow of articles through the review process isshown in Appendix 7 in accordance with Qualityof Reporting of Meta-analyses (QUOROM). 101Twenty-seven articles reporting 22 studies wereidentified as relevant to the review of methods toimprove uptake of cardiac rehabilitation and wereformally included in the review. Reading by tworeviewers (AB and RW) found eight studiesreporting evaluation of an intervention relating touptake by an appropriate patient group and with arelevant outcome.A brief summary of each study is shown inTable 18, with detailed descriptions presented inAppendix 8. More than one report was identifiedfrom the trials of Jolly and colleagues 102–104 andWyer and colleagues. 105,106 The referenceproviding the main source of information for thesystematic review is cited in the tables and text.Studies excluded from review ofinterventions to improve uptake ofcardiac rehabilitationSixteen papers describing 14 studies selected fordata extraction but not included in the review aresummarised in Appendix 9. One paper reportinga before-and-after study of an intervention toimprove uptake of cardiac rehabilitation by CABGpatients was published after June 2001. 112 Eightstudies had either no outcome data 113–119 or nocomparison group, 120 or the study wasretrospective in design. 121 In five studies theoutcomes were referral 122,123 or commitment toparticipate, 124 or the study was related tosecondary prevention. 125–127Methodological qualities of studiesincluded in review of interventions toimprove uptake of cardiacrehabilitationSix studies reported interventions with a specificobjective of increasing uptake of outpatient cardiacrehabilitation. 41,82,104,106,110,111 Two papersdescribed interventions to improve uptake ofcommunity or voluntary services (cardiac or heartclubs) after discharge from inpatient cardiacrehabilitation. 108,109 All studies were of patientswith myocardial infarction, and in two studiespatients with angina 104 and following cardiacsurgery 110 were included.Three of the eight studies were RCTs, withrandomisation on an individual basis in two 106,108and by general practice in one. 104 Methods ofrandomisation and blind outcome assessment wereclearly described for two of the three RCTs 104,106and intervention groups were similar at baseline inall three trials.Five articles reported non-randomisedcomparisons. 41,82,109,110,111 In one study a districtproviding an intervention was compared with adistrict with no intervention. 82 The districts hadpopulations with similar demographics that wereserved by the same general hospital. The otherfour papers reported uptake of cardiacrehabilitation in periods before and afterimplementation of an intervention. 41,109,110,111Baseline characteristics of groups were notreported in these studies. One before-and-afterstudy reported percentage uptake but did notprovide patient numbers or tests of statisticalsignificance. 11031© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve uptake of cardiac rehabilitationTABLE 18 Studies evaluating interventions to improve uptake of cardiac rehabilitationAuthors, year Study type and Intervention Findings relevant to Commentsand country patients uptakeRCTsWyer et al.,2001 106UKRCT, 87 MIpatientsLetters based on the theoryof planned behaviour (Ajzen& Madden) 107 designed toincrease attendance at CRUptake of outpatient CR was86% in the interventiongroup and 57% in thecontrol group (p < 0.0025)Intervention open toalternativeinterpretation; lettersconveying a ‘fear’messageHillebrandet al., 1995 108GermanyRCT, 94 MIpatientsFollowing inpatient CRpatients had four telephoneand at home conversationswith social worker over a 6-month period57% of patients whoreceived the interventionattended a cardiac groupcompared with 27% ofcontrols (p < 0.005)Outcome is cardiacgroup attendance afterinpatient CRJolly et al.,1999 104UKCluster RCT, 67general practices,597 MI andangina patientsLiaison nurse encouragespatients to see practice nurseafter discharge and supportspractice nurses. Patient-heldrecord card to prompt andguide follow-up42% of patients in theintervention group attendedat least one outpatient CRsession compared with 24%of controls (p < 0.001)Multifacetedintervention.Management of patientsin control practices notexplicitNon-randomised studiesOsika, 2001 82UKComparison oftwo districts withdifferentprovision, 175 MIpatientsWeekly home visits bytrained lay volunteers andaccompaniment to first CRsessionIn the district with layvolunteer visiting 71% ofpatients attended a firstappointment at outpatientCR compared with 47% inthe control district(p = 0.02)Intensity of lay volunteervisiting comparable withtypical CRKrasemann &Busch, 1988 109Germany200 MI patientsattending indifferent periodsAfter completion of inpatientCR pamphlet given withinformation designed tomotivate patients to join anoutpatient heart group66% of patients whoreceived the interventionattended a heart groupcompared with 31% in thecontrol group (p < 0.001)No baseline groupcomparisonsMosca et al.,1998 41USABefore-and-afterstudy, 199 MIpatientsPrompt for outpatient CR indischarge critical carepathwayCritical care pathway No baseline groupassociated with a nonsignificantincrease incomparisonsoutpatient CR participation(OR 1.9, 95% CI 0.6 to5.5). aImich, 1997 110UKBefore-and-afterstudy, MI andcardiac surgery,patient numbernot reportedNurse support, educationand counselling in thepostdischarge, preoutpatientCR periodAttendance at outpatient CRby invited patients increasedfrom 55% before to 75%after instigation of theprogrammePatient numbers notreported, so notpossible to assessstatistical significanceScott et al.,2000 111AustraliaBefore and afterstudy, 649 MIpatientsDissemination of clinicalguidelines to hospital staffand GPs. Feedback on clinicalindicatorsAfter intervention outpatientCR utilisation increased from24% to 54% (p = 0.003)Baseline periodcorresponds to CRprogramme start-upa Lower 95% CI estimated.32

Health Technology Assessment 2004; Vol. 8: No. 41Themes identified from the review ofinterventions to improve uptake ofcardiac rehabilitationThe interventions identified in the systematicreview can be grouped into four themes:healthcare professional-led interventions at thepatient level, trained lay volunteers,coordination of postdischarge care at theservice level, and written motivationalcommunications.Healthcare professional-led interventions at thepatient levelThree studies were identified but nonesatisfactorily assessed the value of patient contactwith healthcare professionals in improving cardiacrehabilitation uptake. 104,108,110 In the RCT ofHillebrand and colleagues 108 attendance at acardiac group after inpatient rehabilitation wassignificantly increased in myocardial infarctionpatients who received regular contact with a socialworker. The social worker–patient contacts attimes relevant to improvements in uptake were avisit in hospital and a telephone call 4 weeks afterdischarge. The authors considered this to be amotivational intervention. In the before-and-afterstudy of Imich 110 postdischarge at-home nursingsupport for myocardial infarction and cardiacsurgery patients was associated withimprovements in attendance at outpatient cardiacrehabilitation. However, little information relatingto study quality and conduct was reported. In thenursing intervention reported by Jolly andcolleagues, 104 although patients saw a liaisonnurse in hospital before discharge theintervention was aimed mainly at professionalorganisation of care and is discussed in thatsection.Trained lay volunteersOne study looked at an intervention with trainedlay volunteers. 82 In the thesis, Osika 82 describesincreased cardiac rehabilitation uptake associatedwith an intervention by trained lay volunteers.The study compared myocardial infarctionpatients in two districts with similar populationsserved by the same general hospital. In onedistrict patients were offered the assistance of apatient who had previously attended cardiacrehabilitation. Patients in the district with the layvolunteer intervention were significantly morelikely to attend the first session of the outpatientcardiac rehabilitation programme. In theabsence of randomisation the author attemptedto validate the method by reporting similaritiesin demographics and service access betweengroups.Coordination of referral and postdischarge careat the service levelInterventions aimed at increasing uptake ofoutpatient cardiac rehabilitation by improving thecoordination of postdischarge care were reportedin three studies. 41,104,111 Jolly and colleagues 104reported a cluster RCT of coordination of carebetween hospital and general practice by specialistcardiac liaison nurses for myocardial infarctionand angina patients. Attendance at one or morecardiac rehabilitation sessions was significantlyincreased in the intervention group. Theintervention consisted of three main elements:liaison nurse encouragement for patient to seepractice nurse, liaison nurse support for practicenurses, and prompts and guidance for patients bymeans of a personal record card. The study designdoes not allow the effect of components to beassessed individually. Mosca and colleagues 41compared patient participation before and afterthe introduction of a prompt for cardiacrehabilitation in a discharge critical care pathway.An improvement in participation in outpatientcardiac rehabilitation was observed, but this wasnot statistically significant. Group characteristicswere not reported and other factors may haveinfluenced levels of participation. In the study byScott and colleagues, 111 patients admitted tohospital in three periods were compared. Thesewere before, during and after the dissemination ofclinical guidelines and feedback of clinicalindicators to health professionals. The cardiacrehabilitation programme was operational duringthe implementation period and this was used asthe baseline period for evaluation. A steadyincrease in utilisation of the outpatient cardiacrehabilitation service was observed during theimplementation period and the authors attributethis to the intervention. However, no comparisonsof patient characteristics were available for therelevant periods and, although the authors reportthat the new cardiac rehabilitation service was fullyoperational, an increase in uptake might beexpected with a new service.Motivational communicationsOne study showed significantly increasedoutpatient cardiac rehabilitation uptake aftermotivational letters 106 and two showed improvedattendance at an outpatient heart group aftermotivational pamphlets 109 or conversations. 108 Inthe RCT of Wyer and colleagues, 106 motivationalletters were sent to patients at 3 days and 3 weekspostmyocardial infarction. The letters were basedon Ajzen and Madden’s theory of plannedbehaviour 107 and designed to influence acceptanceand attendance, although the authors noted that33© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve uptake of cardiac rehabilitation34the intervention many have been interpreted as afear message. Krasemann and Busch 109 describeda before-and-after study in which the interventiongroup received a pamphlet with motivationalinformation about outpatient heart groups as acontinuation of inpatient cardiac rehabilitation.Patients in both intervention and comparisongroups received the addresses of local outpatientheart groups. The patients receiving themotivational pamphlet were more likely to attendthe heart group, but no comparison of baselinecharacteristics of the patient groups was reported.The RCT of Hillebrand and colleagues 108evaluated regular contact between a social workerand patients starting at the end of an inpatientcardiac rehabilitation programme. A motivatingconversation predischarge and a telephone callafter 4 weeks were associated with improvedattendance at an outpatient heart group.Another before-and-after study of an interventionto improve uptake of cardiac rehabilitation waspublished outside the review time-frame. 112 Theintervention comprised a telephonecommunication about the benefits of cardiacrehabilitation plus assistance in the referralprocess. The authors reported an increase inenrolment, but the significance of this interventionis not clear, as other between-group comparisonswere not described.Resource implications of interventionsto improve uptake of cardiacrehabilitationInformation provided on resource use associatedwith effective interventions from studies ofreasonable quality is summarised in Table 19. Theinterventions can be summarised into threecategories: home visits by trained lay volunteers,coordination of referral and postdischarge care bypaid liaison nursing staff; and motivationalcommunication letters or pamphlets distributed bypaid staff.Studies provided limited information on theresource inputs required. None providedinformation on the costs associated with theseresource inputs. It was unclear from theinformation available whether interventions can beimplemented by existing staff or requireemployment of extra staff and, if so, how many.Osika 82 does not specify the number of trained layvolunteers per patient population required tocarry out home visits in order to encouragepatients’ attendance for cardiac rehabilitation. Themain cost incurred by the health service is thatassociated with the one-off training programmeprovided by the hospital for the lay volunteers. Acardiac rehabilitation coordinator, a counsellor, aresuscitation officer and a safety officer conductedthe training, which comprised seven 5-hoursessions. The specific time input of each staff type,however, was not clearly specified, nor was it clearhow many lay volunteers were trained over thisperiod. Lay volunteers were reimbursed mileagecosts to attend for training and home visits, butthese were not quantified.Similarly, the staff implications of liaison nursecoordination of referral and postdischarge carewere unclear. The intervention evaluated by Jollyand colleagues 104 comprised three cardiac liaisonnurses who coordinated the referral andpostdischarge care of 277 patients over 18months. This suggests that one nurse could beresponsible for coordinating the referral andpostdischarge care of 62 patients per annum. It isunderstood that these liaison nurses were newappointments. Although mentioned in the study,transport costs incurred by the liaison nursevisiting practices and by the practice nursesattending training and support groups were notquantified, nor was the resource input or cost oftraining the liaison and practice nurses.The use of motivational letters and pamphletsmay require some initial preparation and printing,but at little additional resource input to thestandard programme invitation, as these are likelyto replace existing letters. Motivational telephoneconversations and home visits by social workerswill require staff time and transport costs, butthese were not quantified in the study byHillebrand and colleagues. 108Further interventions that may improveuptake of cardiac rehabilitationsuggested in the literatureThe literature review identified a number ofsuggested interventions for improving uptake ofcardiac rehabilitation. Although these potentialinterventions were not evaluated, the studiesprovided some evidence to suggest methodsmeriting further investigation. Examples ofinterventions excluded from the review at bothformal extraction and the earlier inclusion stage,but with possible value in improving uptake, aresummarised thematically in Table 20.DiscussionFew studies aimed at improving uptake ofoutpatient cardiac rehabilitation were found. The

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 19 Resource implications of interventions to improve uptake of cardiac rehabilitation (only considering period of interventionrelating to improving uptake)Staff Equipment Consumables NotesLay volunteersHome visits by trainedlay volunteers andaccompaniment to firstCR session(Osika, 2001 82 )Multidisciplinary teamproviding 35 hourstraining per group ofvolunteersCar mileage for sevenlocal visits per patient,training at hospitalConsiderable timedemands on volunteersCoordination of referral and postdischarge careLiaison nurseencouragement to seepractice nurse. Supportfor practice nurses.Patient-held record card.(Jolly et al., 1999 104 )Motivational communicationTelephone and at-homeconversations with socialworker (Hillebrand et al.,1995 108 )Three cardiac liaisonnurses visit bothpatients and practicenursesSocial worker visitspatientsTelephone calls; trainingfor cardiac liaison nursesand practice nurses; carmileage allowance fortraining and supportmeetingsTelephone calls; carmileage for one localvisit per patientThe study employedthree nurses managingthe care of 277 patientsin 18 monthsMotivational lettersdesigned to increaseattendance (Wyer et al.,2001 106 )(Support staff; minimalrevision of normalpractice)Letters and postageLetters substitute forexisting invitationsPamphlet withinformation designed tomotivate patients to joinoutpatient heart groups(Krasemann & Busch,1988 109 )(Support staff; minimalrevision of normalpractice)Pamphletssource of studies was diverse with five paperspublished in peer-reviewed journals, two as theses(the paper by Wyer and colleagues 106 waspublished after June 2001 but had previously beenwritten as a thesis) and one as a conferenceabstract. The systematic review identified studiesof four types of intervention aimed towardsimproving uptake of outpatient cardiacrehabilitation and heart groups: written or auralmotivational communications, healthcareprofessional-led interventions at the patient level,coordination of referral and postdischarge care atthe service level, and lay volunteers.The evidence for benefits from motivationalcommunications was reasonably good, withimprovements in uptake of outpatient cardiacrehabilitation and heart groups shown in tworandomised trials 106,108 and one before-and-afterstudy. 109 Methods of communication used werewritten letters 106 or pamphlets, 109 or conversationwith a health professional. 108No conclusions can be drawn on the effectivenessof an intensive home-based nurse-led approach inpromoting outpatient cardiac rehabilitationuptake, owing to the limited information in theone report looking at this type of intervention. 110A multifaceted approach to the coordination oftransfer of care from hospital to general practiceincluding patient self-management was effective inimproving cardiac rehabilitation uptake in arandomised trial. 104 Particular aspects of theintervention were not evaluated separately, and itis not possible to compare the relative importanceof inpatient nurse contact, professional support ofpractice nurses and self-empowerment of patientswith record cards. Issues relating to study qualitylimit further support from two non-randomisedtrials. 41,111Regular support and practical assistance from layvolunteers was effective in improving uptake ofoutpatient cardiac rehabilitation in a non-35© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve uptake of cardiac rehabilitationTABLE 20 Further interventions that may improve uptake of cardiac rehabilitation suggested in the literatureAuthors, date Intervention Description of reportNon-specificSuskin et al., 2000 124 Physician endorsement RCT. Outcome is intent toparticipate in CRCaulin-Glaser & Education of health professionals Before-and-after study. OutcomeSchmeizel, 2000 123is referral to CRKalayi et al., 1999 122 Computerised referral pathway Before-and-after study. Outcomeis referral to CRCannistra et al., 1995 128 Early social services involvement could improve social Prospective study comparingsupport and therefore uptake and adherence, by reducing black and white womenhome stressBeach et al., 1996 71 Self-care limitations assessment may help to assess, plan and Longitudinal interviewsfacilitate healthy perceptions and behaviour post-MI andpromote CRTack & Gilliss, 1990 129 Early information and follow-up can improve recovery Prospective, longitudinalexpectations, give support and promote healthy coping and interviewsCR uptakeHershberger et al., Assessment of patient personality type could give a better Retrospective study1999 130 indicator of compliance, uptake and adherence, and allowprofessionals to target those most in needAlternative methods of provisionDeBusk et al., 1985 131 Home-based rehabilitation with ECG monitoring RCT. Comparison of methods ofdeliveryLewin et al., 1992 132 Home-based rehabilitation RCT. EffectivenessShaw, 1999 133 Physiologically monitored exercise and health education Reviewover the InternetAdes et al., 2000 134 Home-based telephone-monitored CR Trial: group allocation by distancefrom CR. Comparison ofmethods of deliveryRoitman et al., 1998 135 Case-management and risk stratification ReviewBethell & Mullee, 1990 136 Community-based CR: achieves high patient uptake RCT. EffectivenessPell & Morrison, 1998 51 Community-based CR: more patient friendly, improving Audituptake, particularly if run in socially deprived areasContractor et al., 2000 137 Community-based CR: may increase accessibility of services RCT. EffectivenessInterventions for womenRadley et al., 1998 57 Implementing a one-off women-only education session in a Retrospective studyCR programme may help to address gender-sensitive issuese.g. returning to sexual relations and houseworkMoore, 1996 138 Women-specific social support. Strategies to improve social Focus-group interviewssupport: better exercise variety and choice, and socialopportunities during the programmeBrezinka et al., 1998 139 Women-specific counselling and smaller exercise sessions Comparative semistructuredinterviews and questionnairesCannistra et al., 1992 24 Provision of childcare/home-help for women attending Prospective study comparing menoutpatient CRand womenToobert et al., 1998 140 Women’s retreat could increase uptake by improving RCT. Effectivenessemotional social support and relationships with CR staff36randomised trial of demographically similardistricts with different service provision. 82All authors reported benefit for interventions toimprove uptake of outpatient cardiacrehabilitation. This observation should be treatedwith caution, as it is suggestive of positivepublication bias. 141 However, the wide-rangingsearch of grey literature including conferenceabstracts and theses should have identified studiesconsidered of limited value for dissemination byauthors and publishers. Although it is reasonable

Health Technology Assessment 2004; Vol. 8: No. 41to anticipate some improvement, it is not knownhow many similar or different interventions havebeen tried without success and, becauseunsuccessful, not reported. Similarly, equivocalresults relating to cardiac rehabilitation uptakemay not have been included in publications ofstudies with multiple findings. Three of the eightstudies included in the review reported substantialmaterial on other outcomes or observations, whichwould have merited full publication.It would be inappropriate to draw firmconclusions relating costs to effectiveness of theinterventions described in the above studies. Theirresource use implications are not clearly described.However, order of magnitude costs may beinferred and these suggest a wide range in impliedcosts. Motivational interventions need not becostly, as they may replace the existing method ofinvitation. Individual home visits clearly add tothe costs of a service otherwise provided in anoutpatient setting. However, more visiting isbecoming part of postdischarge care and cardiacrehabilitation. The study by Jolly and colleagues 104was an evaluation of the introduction of liaisonnurses (which would certainly be more costly) andmay serve to define their role in supportingpatients and other healthcare professionals andcoordinating postdischarge care. Incorporation ofmotivational elements into home visiting may beappropriate, with little further implications forresources.The literature contained many suggestedinterventions as facilitators of improved uptake ofcardiac rehabilitation, but with no relevantevaluation. At the service level, appropriateeducation of health professionals and use ofdischarge care pathways and case managementmay improve referral and subsequent attendanceat rehabilitation. 122–124,135 In a similar vein to themotivational approach of Wyer and colleagues 106the form of the recommendation to attend may beimportant, with endorsement by a physician ofpossible value. 124Early support postdischarge by healthcareprofessionals may be appropriate in promotingcardiac rehabilitation and improving uptake. 128,129In addition, at an early stage the assessment ofpatients with regard to self-care limitations andpersonality type may be helpful in the targeting,planning and optimisation of postdischarge care,including rehabilitation. 71,130Home-based programmes are frequently used inthe period between hospital discharge andattendance at outpatient cardiac rehabilitation. 142The home-based programme usually takes theform of a written booklet with an exerciseschedule, psychosocial interventions andeducation relating to risk factor managementappropriate for the early stages of recovery. Thismay serve to maintain patient motivation tolifestyle change in a period with limited contactwith health professionals and hence promote lateruptake of outpatient rehabilitation services. Thismerits further evaluation.Home-based cardiac rehabilitation has also beenpromoted as a substitute for attendance at anoutpatient programme. 131,132 Trials have shownsimilar effectiveness in risk factor managementand patient quality of life after home-based andoutpatient methods. Appropriately delivered andassessed home-based cardiac rehabilitation may bea safe and effective form of provision for low- tomoderate-risk patients. However, application ofthe home-based approach as a means to improvethe reach of cardiac rehabilitation services shouldreplicate the methods used in the trials ofeffectiveness and include frequent nurse visits,multidisciplinary input, psychological evaluationand thorough assessment. Patient acceptance of ahome-based package does not equate to uptake ofcardiac rehabilitation and the demonstration ofcomparability with an existing service may merelybe observation of similar natural recovery in thepatient groups. Consequently, home-basedrehabilitation is not an appropriate substitute foroutpatient services in patients with more severedisease or those with low motivation or lack ofinterest. It may have value in motivated low- tomoderate-risk patients, particularly those livingdistant from current services. If there is arequirement for monitoring and assessment thiscould be undertaken using telemedicineapproaches, including ECG monitoring andtelephone contact during exercise sessions. 131,133,134An alternative approach to cardiac rehabilitationprovision outside the hospital setting is the use offacilities in the community. 51,136,137 The serviceprovided can be identical to the outpatientprogramme in content and multidisciplinarynature but avoid features associated with reducedattendance at hospital, including accessdifficulties. Similarly, factors limiting the uptake ofhome-based cardiac rehabilitation may be avoided,including reduced reliance on patient selfmotivation.Cardiac rehabilitation in a communitysetting merits evaluation as a method forimproving patient uptake and may be particularlyvaluable in socially deprived areas. 5137© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve uptake of cardiac rehabilitationThe need to adapt cardiac rehabilitation servicesto suit female patients has been acknowledged byseveral authors. 57,128,138–140 Many female patientshave a higher level of psychosocial impairmentand lower level of physical function than men, andtherefore need gender-specific approaches torehabilitation. 139 Suggested interventions toimprove uptake include a women-only educationsession, 57 appropriate exercise choices, 138 specificcounselling, 139 strategies to improve socialsupport, 138 provision of childcare and homehelp,24 and a women’s retreat. 140There are numerous reports of reasons for nonattendanceand under-representation in outpatientcardiac rehabilitation, so it is surprising that thesystematic review of published literature identifiedso few evaluations of interventions to improveuptake. All those found were generic interventionsapplicable to all patients. The effectiveness ofsimple targeted interventions to facilitateattendance is not reported in the literature. Noevaluations of interventions were reported toaddress the frequently cited patient reasons fornon-attendance of perceived illness, transportdifficulties, inconvenient timing or dependentrelatives. Transport schemes, non-hospital settings,programmes for specific patient groups (singlegender, elderly, ethnic minority groups) andprovision of respite care for dependants have beensuggested, but not evaluated as possible measuresfor improving service uptake. It is possible thatsome programme coordinators have recogniseddeficits and the need for improvement inservices and implemented changes: provision ofservices for a patient group previously nottargeted for rehabilitation is likely to showinitial improvement in uptake; but this is aHawthorne effect which may not be sustained.However, the lack of evidence for benefit foundfor the use of critical care pathways and thelimited evidence for other interventionsdemonstrate the requirement for good qualityRCTs of new methods.ConclusionsThe systematic review of the literature suggeststhat approaches aimed at motivating patients maybe of value in improving the uptake of cardiacrehabilitation. The content of invitation letters,pamphlets and home visits may be used as avehicle for motivational messages. Someencouragement was also found for use of trainedlay visitors in facilitating patient attendance atcardiac rehabilitation. The implied costs ofinterventions varied widely.Overall, few trials aimed at improving uptake ofcardiac rehabilitation were identified. The needfor trials of interventions applicable to all patientsand targeting specific under-represented groups issuggested by observational studies.38

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 9Systematic review of interventions to improveadherence to cardiac rehabilitationBackgroundFollowing successful recruitment of patients to acardiac rehabilitation programme it is importantto promote patient adherence to the programmeand to maintain associated lifestyle changes. Thissystematic review aims to assess the effectiveness ofmethods for increasing patient adherence tocardiac rehabilitation and to suggest areas forfurther research. For the review adherence wasdefined as patient attendance at all or the majorityof a programme, or maintenance of lifestylechanges associated with cardiac rehabilitation.ResultsStudies included in the review ofinterventions to improve adherencewith cardiac rehabilitationIn Appendix 10 the flow of articles through thereview process is shown in accordance withQUOROM. 101 Thirty-eight articles reporting 37studies identified as relevant to the review ofmethods to improve adherence to outpatientcardiac rehabilitation and its components wereformally included in the review. A broad definitionof adherence was applied, with included studiesreporting attempts to improve overall programmeattendance or compliance with aspects of cardiacrehabilitation. In-depth reading by two reviewers(KR and AB) found 14 studies reportingevaluation of an intervention relating toadherence in an appropriate patient group andwith a relevant outcome. Only studies with anexplicit statement in their objectives that theintervention under evaluation was designed topromote adherence or those studies withobjectives that were explicitly to examine theeffects of an intervention on adherence wereincluded in the review.Studies were characterised by study design andsize, the study participants, nature of theintervention, comparison group, principal andother outcomes, and authors’ conclusions.A brief summary of studies is presented inTable 21, with further details in Appendix 11. Tworeports were identified describing the study ofMiller and colleagues. 143,144Studies excluded from the review ofinterventions to improve adherence tocardiac rehabilitationPapers not included in the review are summarisedin Appendix 12. Nine out of 23 studies excludedfrom the review looked at the effectiveness ofdifferent rehabilitation formats: home-basedcardiac rehabilitation, 73,131,134,158 differentintensities or duration of exercise training, 159–161group counselling 162 or structured teaching. 163These were not included in the review as theyreported effectiveness of interventions with nospecific aim at improving patient adherence tocardiac rehabilitation. Thirteen studies had eitherno relevant outcome 126,164–172 or no comparisongroup. 173–175 One study presented retrospectivedata with no indication of how patients came toreceive an intervention. 121Methodological qualities of studiesincluded in the review of interventionsto improve adherence to cardiacrehabilitationFourteen studies were identified, of which halfwere RCTs 145–151 and half were non-randomisedstudies. 143,144,152–157 One randomised 147 and onenon-randomised study 156 reported two distinctinterventions. In the non-randomised studiespatients were designated to groups by alternateallocation, 143,144,152 before and afterimplementation of an intervention 154,156,157 and byrandom allocation with some non-randomallocation aimed at increasing numbers in theintervention group. 153 In two studies the allocationto groups was not clearly described. 155,156In six studies patients with one specific diagnosiswere included and in eight studies less specificselection was applied. Patients represented weremyocardial infarction (nine studies), CABG (eightstudies), angina (three studies), PTCA (three39© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve adherence to cardiac rehabilitationTABLE 21 Studies evaluating interventions to improve adherence to cardiac rehabilitationAuthors, year Study type and Intervention Outcome relevant to Commentsand country patients adherenceRCTsOldridge &Jones, 1983 145CanadaDaltroy, 1985 146USARCT, 120 MI,CABG and anginapatientsRCT, 174 MI,CABG, PTCA andangina patientsSelf-management: agreementto participate in exerciserehabilitation programmesigned by patient andcoordinator; self-report diarywith monitoring of heartrates; questionnaires of dailyactivities; weight loss andsmoking diaries. Progressdiscussed with coordinator atregular intervalsPersuasive telephoneeducation intervention toimprove patient adherenceto exercise regimens. Oralcommitment to attend.Spouse telephone counsellingAttendance at >60% ofexercise sessions was 54% inthe intervention group and42% in the control group(not statistically significant)Attendance at exercisesessions by patients was63.8% in the interventiongroup and 62.2% in thecomparison group (notstatistically significant)Mahler et al.,1999 147USARCT with twointerventiongroups and onecontrol group,215 CABGpatientsPost-CABG surgeryvideotape. (1) Mastery:depicts patients as calm andconfident, making steadyprogress with relative ease.(2) Coping: recoveryportrayed as steady forwardprogression of ups anddownsExercise complianceimproved with bothinterventions compared withcontrols (p < 0.02 top < 0.05). Reduction indietary cholesterol andsaturated fat at 1 month inboth intervention groupscompared with controls(p < 0.05) but not at3 monthsAish & Isenberg,1996 148CanadaRCT, 104 MIpatientsNursing intervention ofnutritional self-care. Foodhabits assessed andsuggestions for changes givenwith patient commitment.Follow-up telephone callsTotal dietary and saturatedfat significantly reduced inthe intervention group(p < 0.01). Also significantimprovements in food habits(p < 0.05)Ashe, 1993 149USAAllocation byform in sealedenvelope,41 MI, CABG,angina, valveproblem patientsMotivational relapseprevention during the CRprogramme: identification offactors interfering withadherence; goals forprogramme; coping withslips; stressors affectinglifestyle. Also stressmanagement, exercise andrelaxation procedureTotal adherence to themaximum number ofexercise sessions was 90% inthe intervention group and89% in the control group(not significant)40continued

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 21 Studies evaluating interventions to improve adherence to cardiac rehabilitation (cont’d)Authors, year Study type and Intervention Outcome relevant to Commentsand country patients adherenceHopper,1995 150USARCT, 80 MI,CABG,angioplastypatientsMonthly telephone contactby a cardiac nurse orexercise physiologist topromote exercise, healthydiet, medication usage, riskfactor knowledge andidentification of symptomsNo difference betweengroups in exercise habit orintention to exercise.Conditions that facilitated theperformance of exercisewere improved in theintervention group comparedwith control (p < 0.05)Losses to follow-up of45% in the interventiongroup and 47% incontrolsDuncan et al.,2001 151USARCT, eight heartfailure patientsBehavioural feedbackintervention on dietarysodium intake. Discussion ofproblem-solving strategies toreduce sodium intakeSodium intake wassignificantly less in theintervention group (1569 vs2836 mg, p < 0.05)Small feasibility study,four patients in eachgroupNon-randomised studiesLeslie &Schuster,1991 152USAAlternateallocation, 30 MI,CABG,angioplasty,coronary diseasepatientsWritten exercise contractnegotiated with the patient.On completion of thecontract patients received arewardNo significant difference inattendance in intervention(90%) and control (89%).Significant increase inexercise knowledge in theintervention groupMiller et al.,1988, 1431989 144USAAlternateallocation,115 MI patientsNurse intervention toimprove medical regimen.(1) Assessment: attitudes andregimen compliance.(2) Problem identification.(3) Goal settingNo significant differences inhealth behaviour and attitudescalesRepeated self-evaluationquestionnaires and visitsmay have acted asintervention in controlgroupLack, 1985 153USAPart random, partnon-randomised,48 CHD, MI,CABG patientsInsight-orientated grouppsychotherapy. Supportive,cooperative and goaldirected. Highlight andpromote change in noncompliancewith physicianrecommendationsSelf-report measures ofcompliance 2.57 forintervention and 2.37 controlgroups (not significant).Intervention group attended88.4% of the prescribedexercise sessions comparedwith 75.7% in the controlgroup (p < 0.05)Marshall et al.,1986 154USAPatients seen indifferent periods,60 CABGpatientsNurse-led structuredteaching programme toincrease patients’ knowledgeand compliance tomedication, diet, smokingcessation and exerciseOverall compliance scoreassessed by self-report was86.8 in the interventiongroup, and 79.5 in thecontrol group (p < 0.05).Compliance better inintervention than controlgroup for activity (15.6 vs 7blocks walked, p < 0.005)Huerin et al.,1998 155ArgentinaNon-randomisedstudy, 509 CHDpatientsAdherence strategy withsigned commitment torehabilitation, familyinvolvement, sports,recreational activities andtalksAttendance at ≥ 66%sessions. RR 2.3 (95% CI 1.8to 2.9) at 12 weeks, 2.9 (2.3to 3.7) at 24 weeks, 4.25(3.2 to 5.6) at 52 weeks (logranktest between strategies,p < 0.001)No information ongroup allocationcontinued41© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve adherence to cardiac rehabilitationTABLE 21 Studies evaluating interventions to improve adherence to cardiac rehabilitation (cont’d)Authors, year Study type and Intervention Outcome relevant to Commentsand country patients adherenceMcKenna et al.,1998 156UK(1) Nonrandomisedstudy. (2) MIpatientscompared withhistorical controls(1) Low-intensity exercise forpatients unable to take partin standard exercise owing toco-morbidity. (2) WomenonlygroupsAttendance was 82% in thelow-intensity exercise groupand 34% in the standardrehabilitation comparisongroup. Significance notassessed as patient numbersnot reported. Attendance inthe women-only group was75%, compared with 6%historicallyPatient numbers notreported. Themagnitude of changescannot be assessedErling &Oldridge,1985 157CanadaBefore-and-afterstudy, 90 CHDpatientsSpouse support in outpatientCR. Compares baselinebefore spouse participation,patients with spouseparticipation and patientswith no spouse participationAttendance increased from44% to 90% for programmewith spouse participation(p < 0.001), and 67% forprogramme with no spouseparticipation (p < 0.05)RR, relative risk.42studies), heart failure (one study), valvereplacement (one study) and non-specific coronaryheart disease (four studies).In eight studies (two randomised) the outcome wasattendance at exercise sessions. 145,146,149,152,153,155–157In six studies (four randomised) the outcome wasquestionnaire assessment of diet or exercisebehaviours to determine compliance with lifestylechanges. 143,144,147,148,150,151,154The method of randomisation was described intwo of the seven RCTs 145,146 and blind outcomeassessment in one randomised study. 146 None ofthe seven non-randomised studies reported blindoutcome assessment. Baseline characteristics ofintervention and comparison groups weredescribed in three randomised trials, 146,148,150 andin one trial patients were stratified by factorspredictive of dropout from cardiacrehabilitation. 145 In five non-randomised studiesbaseline characteristics of patients werereported. 143,144,152–155 Eight studies providedinformation on losses to followup.143–147,149,150,152,153 In one randomised study lossto follow-up was particularly high at 45–47%. 150Themes identified from the review ofinterventions to improve adherence tocardiac rehabilitationInterventions to improve adherence to cardiacrehabilitation or elements of the rehabilitationprocess were varied and frequently multifaceted.However, five general themes are apparent: formalpatient commitment, spouse or familyinvolvement, strategies to aid self-management,education, and psychological intervention. Studieswith more than one component are included ineach appropriate theme.Formal patient commitmentIn four studies an agreement between the patientand the programme staff was a key element of theintervention. 145,146,152,155 In the trial of Leslie andSchuster 152 the intervention was solely a writtencontingency contract with rewards for successfulcompletion of attainable exercise behaviours.Patients were allocated alternatively to interventionand comparison groups, which were reasonablywell matched. Attendance rates at exercise sessionswere similar in the two groups, although patientsin the contract group showed a significant increasein exercise knowledge compared with controls.Overall participation at exercise sessions wasnotably high (90%). In the RCT of Oldridge andJones, 145 a self-managed adherence promotingstrategy incorporating signed commitment wasassociated with a non-significant improvement inattendance at an exercise rehabilitationprogramme. Huerin and colleagues 155 reported asignificant increase in cardiac rehabilitationattendance in patients receiving an adherencepromotingstrategy with signed agreement, butlittle information was presented on the allocationof patients to groups. Daltroy 146 reported an RCTin which oral commitment was included in a

Health Technology Assessment 2004; Vol. 8: No. 41persuasive telephone intervention to improvepatient adherence to an exercise programme. Noimprovement in attendance was seen in patientsreceiving the intervention.Spouse or family involvementThree studies included an intervention directedat the patient’s spouse or family. 146,155,157 Erlingand Oldridge 157 reported a before-and-after studyin which a spouse support programme wasassociated with significantly increased patientattendance at a cardiac rehabilitation programme.The authors showed no baseline comparisons ofthe two groups. In the RTC of persuasivetelephone education, Daltroy 146 providedtelephone counselling to patient spouses. Noimprovement in attendance was associated withthe intervention. Family involvement was also acomponent of the adherence strategy of Huerinand colleagues. 155 Improved outpatient cardiacrehabilitation attendance was observed, but lack ofinformation on group allocation limits the value ofthe study.Strategies to aid self-managementFive studies reported interventions based on selfmanagementtechniques. 143–145,148,149,151 In theRCT of Oldridge and Jones, 145 as well as signedagreement, patients completed and receivedfeedback on self-report diaries of heart rate, dailyactivities, weight loss and smoking habit. Theintervention was associated with a non-significantincrease in attendance at the exerciserehabilitation programme. Aish and Isenberg 148reported an RCT of nutritional self-care based onthe model of Orem, 176 in which patients had foodhabits assessed and individualised nutritionalgoals set. Significant improvements in dietaryvariables were achieved in the self-care patients. Asimilar programme of assessment, problemidentification and goal setting was assessed in atrial by Miller and colleagues, 143,144 in whichpatients were allocated alternatively tointervention and control after completion of acourse of inpatient rehabilitation. Regimencompliance measured by health behaviour andattitude scales did not differ between groups. Theauthors noted that the frequent completion of selfevaluationquestionnaires and data collection visitsby nurses may have served as an effectiveintervention in the control group. In the thesis byAshe 149 an apparently randomised approach wasused to evaluate a motivational relapse preventionprogramme, based on Marlatt and Gordon’smodel, 177 for patients after an outpatient cardiacrehabilitation programme. Patients were allocatedto groups according to the forms contained insealed envelopes. As with other self-managementinterventions this included assessment, problemidentification and goal setting. Adherence toexercise was similar in the intervention andcontrol groups. However, it should be noted thatthe control patients received an interventionwhich, although not designed as a motivationalprogramme, did provide patients with anequivalent number of extra sessions of exerciseeducation. A small RCT described by Duncan andcolleagues 151 applied self-management andbehavioural feedback methods to the control ofsodium intake. Heart failure patients attending acardiopulmonary rehabilitation programme wererandomised to receive an intervention withassessment of sodium intake, discussion ofproblem-solving strategies and follow-up. Patientsreceiving the intervention had a significantlyreduced sodium intake.Educational interventionFour studies of educational interventions aimed atimproving adherence to components of cardiacrehabilitation were identified. 146,147,150,154 TheRCT reported by Daltroy included an educationalintervention in the form of telephonecounselling. 146 This was designed as a persuasivecommunication with emphasis on the benefits ofexercise, realistic expectations of recovery andcoping methods. Attendance at exerciseprogrammes was not improved in the group withthe educational intervention. In the RCT ofMahler and colleagues 147 patients were showneducational videotapes before discharge fromhospital. The tapes provided informationregarding recovery delivered by a healthcareexpert. Compliance with exercise and dietaryadvice measured by questionnaire was increased inpatients receiving the intervention. The authorssuggest that presenting the information in aformat describing a realistic coping approach torecovery may be beneficial. Hopper 150 describedan RCT of regular educational and supportivetelephone calls. Although no difference was shownin exercise behaviour between groups, thosepatients receiving the supportive educationalintervention did report improvement in conditionsfacilitating the performance of exercise. Marshalland colleagues 154 compared the effect of nurse-ledstructured and non-structured postoperativeteaching in two consecutive groups of patients.Patient characteristics and risk factors were similarin the two groups. Measures of compliance basedon self-report of activity, smoking, and acomposite of activity, smoking, diet andmedications were improved in the structuredteaching group.43© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve adherence to cardiac rehabilitationTABLE 22 Resource implications of interventions to improve adherence to cardiac rehabilitationStrategy to aid self-managementNursing intervention ofnutritional self-care(Aish & Isenberg, 1996 148 )Staff Equipment Consumables NotesNurse Telephone One visit in hospital andlocal visit; threetelephone calls perpatient; dietary recordsBehavioural feedback ondietary sodium (Duncanet al., 2001 151 )NurseTwo interviews duringoutpatient rehabilitation;dietary recordsEducational interventionPost-CABG surgeryvideotape (Mahler et al.,1999 147 )Video recordingand playerVideo shown in hospitalbefore discharge44Psychological interventionOne intervention describing a specificallypsychological intervention was identified. 153Lack 153 describes an insight-orientated grouppsychotherapy intervention in a partiallyrandomised study. The randomised group wasaugmented with non-randomised patients if therewere insufficient numbers to form an interventiongroup. For the intervention patients wereencouraged to communicate thoughts and feelingsand to promote changes in behaviours likely toaffect recovery. Self-report and physiologicalmarkers of compliance to exercise were littlechanged by the intervention. However, there was asignificantly higher attendance at exercise sessionsin the patients receiving the psychotherapyintervention.Other interventionsTwo reports described interventions that did notfit into the above themes. 155,156 McKenna andcolleagues 156 reported that attendance wasincreased after implementation of women-onlyand low-intensity exercise programmes. Numbersin comparison groups and patient characteristicswere not reported. In the study by Huerin andcolleagues 155 recreational activities and sports wereincluded in the adherence strategy. Again, thereporting precludes any assessment of value.Resource implications of interventionsto improve adherence to cardiacrehabilitationInformation provided on resource use associatedwith effective interventions from studies ofreasonable quality is summarised in Table 22. Theinterventions can be summarised into twocategories: strategies to aid self-management, andeducational interventions. Studies providedlimited information on the resource inputs and noinformation on the costs associated with theseresource inputs.It is unclear whether the strategies to aid selfmanagementcan be implemented by existing staffor require the employment of extra staff and, if so,how many. The intervention described by Aish andIsenberg 148 consisted of two interviews with anurse for dietary assessment and three follow-uptelephone calls by a nurse for each patient. Thefirst interview was conducted in hospital and thesecond at a home visit. Duncan and colleagues 151evaluated a similar intervention, but in their studyboth interviews were conducted during outpatientcardiac rehabilitation sessions. In both studiespatients were also required to complete a 3-daydietary intake log. The videotape intervention ofMahler and colleagues 147 was provided in hospitalbefore discharge. After initial preparation ofeducational material the main resource input ofthe intervention would be the appropriateaudiovisual equipment.Further interventions that may improveadherence to cardiac rehabilitationsuggested in the literatureThe literature review identified a number ofsuggested interventions for improving adherenceto cardiac rehabilitation. Although these potentialinterventions were not evaluated, the studiesprovided some evidence to suggest methodsmeriting further investigation. Examples ofinterventions excluded from the review at bothformal extraction and the earlier inclusion stage,but with possible value in improving uptake, aresummarised thematically in Table 23.

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 23 Further interventions that may improve adherence to cardiac rehabilitation suggested in the literatureAuthors, year Intervention Purpose of studyNon-specificMcGee & Horgan, 1992 54 Former patients as models may help to promote adherence AuditTooth & McKenna, Strategy to improve self-efficacy. Patient modelling on video Review1996 178 and audiotape and in leaflets. Patients view other patients(e.g. healthy meal preparation)Koikkalainen et al., Social skill and taste-training may remove barriers to a Structured interviews1996 179 healthy lifestyleKnapp & Blackwell, Offering specific and practical assistance for spouses Review1985 180 (e.g. menus and recipes) to help improve, lifestyle changeEdgren, 1998 181 Hydrotherapy as part of the exercise component. To Generic case study and interviewsincrease self-training patients attend a gym and/or hydrosession weeklyLee et al., 1996 161 Lower intensity exercise programme RCT of effectivenessOldridge, 1984 182 Vary programme and include swimming and different Reviewexercise equipmentDeBusk et al., 1994 183 Case management and risk stratification. A more RCT of effectivenessRoitman, et al., 1998 135 individualised package of care may lead to improved ReviewadherenceHoepfel-Harris, 1980 184 Provide classes at convenient times, including before work ReviewComoss, 1988 185 and evenings ReviewEmery, 1995 186ReviewInterventions for womenRadley et al., 1998 57 A one-off women-only education session may help to Retrospective studyaddress gender-sensitive rehabilitation issuesMoore & Kramer, 1996 187 Women-specific social support, exercise variety and choice, Focus-group interviewsand social opportunitiesBrezinka et al., 1998 139 Women-specific counselling and smaller exercise sessions Comparative semistructuredinterviews and questionnairesToobert et al., 1998 140 Women’s retreat to improve emotional social support and RCT of effectivenessrelationships with cardiac rehabilitation staffCannistra et al., 1992 24 Provision of childcare or home-help for women attending Prospective comparison study ofcardiac rehabilitationmen and womenInterventions for the elderlyAllen & Redman, 1996 188 Awareness of elderly-specific hindrances. Shorter education Reviewsessions with less information run at a slower paceInterventions for ethnic groupsCaulin-Glaser & Take account of cultural and racial differences when Prospective observational studySchmeizel, 2000 123 attempting to improve diet and exercise habits.African–American males showed fewer improvements indiet than CaucasiansEftekhari et al., 2000 189 Translation and presentation of educational material for Programme descriptionAsian patientsDiscussionInformation came from journals (seven studies),theses (three studies) and conference abstracts(four studies). It was disappointing to find thatsome evaluations of potentially valuable methodsto improve adherence to outpatient cardiacrehabilitation and its components were only foundin the grey literature, with little chance of beingread by health professionals.Systematic review of the literature identified fivemain intervention themes: formal patientcommitment, spouse or family involvement,strategies to aid self-management, education, andpsychological therapy.The review of the literature gave little support tothe use of written and oral commitments topromote exercise adherence. The one studylooking exclusively at written contracting showed45© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve adherence to cardiac rehabilitation46no effect using a non-randomised design. 152Attendance in both intervention and controlgroups was high at about 90%. A randomised trialof a self-management programme incorporatingsigned agreement to participate as an adjunct toan exercise rehabilitation programme showed anon-significant improvement in attendance. 145The small benefit cannot be attributed entirely towritten communication as the interventionincluded several other self-managementapproaches. Similarly, in a study of persuasiveintervention by telephone with additional spousecounselling, oral commitment constituted one partof the intervention. 146 No improvement inattendance was attributed to this package ofmeasures. One further study provided littlemethodological information to substantiate anobserved improvement in cardiac rehabilitationafter an adherence strategy incorporating signedcommitment. 155 Overall, the value of formalcommitment in promoting adherence to cardiacrehabilitation is not supported by evidence fromthe literature. The identification of only one studylooking specifically at written agreement toparticipate, but in which attendance was uniformlyhigh in intervention and comparison groups maysuggest the need for more trials. However, it isprobable that the use of written and oralcommitment has better application in thepromotion of outpatient cardiac rehabilitationuptake rather than adherence.Evidence for the benefit of spouse or familyinvolvement in increasing rehabilitation adherencewas limited by the designs of studies. One studylooking specifically at a spouse supportprogramme provided no information on baselinecharacteristics or group allocation. 157 In anotherrandomised study telephone counselling forspouses was provided in addition to a moreintensive patient counselling intervention, but noimprovement in attendance was observed. 146Another study incorporated family involvementinto an adherence-promoting strategy, but littleinformation on the design or conduct of the studywas reported. 155 None of these studies addressedspecifically the issue of spouse or familyinvolvement in promoting rehabilitationattendance in an adequately designed trial.Evidence for the effectiveness of counselling andsupport in helping spouses and families to copewith patient illness suggests that interventions mayhave value other than in promoting adherence tocardiac rehabilitation. 144,190Studies reporting strategies to aid selfmanagementaimed at improving adherence torehabilitation goals give some suggestion ofbenefit. In a randomised trial of self-evaluationand information feedback on exercise and riskfactors a non-significant improvement inattendance at rehabilitation was observed. 145 Inthis trial patients were also asked for writtencommitment. Another randomised trial reportedimprovements in dietary habits, 148 and a small,randomised trial showed reduced sodium intakeafter individualised assessment and goal setting. 151However, two trials, one randomised 149 and theother with non-random allocation to groups, 143,144suggested no benefit for assessment and goalsetting in improving health behaviours or exerciseadherence. The authors noted that controlpatients in these studies received regular selfevaluationquestionnaires and nurse visits for datacollection 143,144 or an educational interventionunrelated to self-management, 149 which may haveaffected outcomes. In trials a repeatedlyadministered evaluation tool may act as anintervention. In conclusion, the uses of appropriatetechniques promoting self-management in specificareas of rehabilitation are at least worthy offurther study.Studies of educational interventions to improveadherence to components of cardiac rehabilitationgave little encouragement. No benefits ofeducation and counselling on attendance at anexercise programme were seen in two RCTs usingtelephone interventions. 146,150 A predischargevideotaped educational intervention was effectivein improving exercise and dietary compliance. 147Although this may be of benefit in the early phasesof rehabilitation it is likely to have limited value inthe promotion of adherence to outpatient cardiacrehabilitation. However, the study did suggest thatpresentation of information in a format describingrecovery based on a coping approach may be mosteffective. The importance of the method ofdissemination of educational information was alsosuggested by a before-and-after study showingbenefit for a structured teaching approach. 154One partially randomised study reported apsychological intervention aimed at improvingexercise adherence. 153 Although no significantimprovement in self-reported exercise wasobserved, the patients receiving a 12-weekpsychotherapy intervention attended more cardiacrehabilitation exercise sessions. This improvedattendance may be a consequence of thepsychological features of the intervention or of theextra requirement to attend the rehabilitationcentre. The lack of an effect on self-reportedexercise tends to support the latter.

Health Technology Assessment 2004; Vol. 8: No. 41Two studies reported other approaches toimproving adherence. These were the inclusion ofrecreational activities and sports in theprogramme 155 and the introduction of outpatientcardiac rehabilitation designed specifically forwomen. 156 Little can be learned from either studyas insufficient information on patients and studymethods was reported.Little information on the costs associated witheffective interventions can be inferred from thepublished reports. One can do little more thanguess the time commitment of nursing staff toundertake these extra tasks. Strategies to improvepatient self-management, such as dietaryassessments, could be incorporated into outpatientcardiac rehabilitation session and may thereforenot require home visits. Nurses would still requiretraining in the evaluation of diet and the analysisof questionnaire data, and this may serve toformalise assessment and procedures already inplace in cardiac rehabilitation.With respect to the educational intervention,videos are frequently used to provide informationto patients before discharge, and presentation ofinformation in an alternative delivery formatwould not have resource implications. However,initial preparation or purchase of appropriateeducational videos would be required.Outside those trials aimed at improving adherenceto cardiac rehabilitation several interventions havebeen suggested but not evaluated. These includemore approaches based around improvements inpatient self-efficacy. The inclusion in programmesof previous patients or representation on video ofbehaviour of model patients showing appropriatelifestyle change (e.g. relating to food preparation)may be a useful format for delivery ofinformation. 54,178 Similarly, practical demonstrationthat a healthy diet can be palatable and enjoyablemay be a method to promote adherence to dietarychange. 179 In these areas of intervention theinvolvement of spouses may be appropriate. 180Alternative forms of exercise, includingswimming, 182 hydrotherapy 181 and lower intensitytraining, 161 may be worthy of evaluation inimproving adherence to rehabilitation.Programmes with lower exercise intensities may bemore likely to achieve maximum attendance, 161but an extended length programme may berequired to maintain benefits. 144 Slower paced andless detailed sessions may be appropriate in theprovision of educational information to elderlypatients. 188Women patients may prefer different kinds ofexercise to men and be more likely to adhere torehabilitation other than treadmill and cycle. 187Other interventions that may improve adherenceby women patients suggested in the literatureinclude education, counselling and social supportaddressing issues specific to women’srecovery. 57,139,187 Provision of childcare or homehelpfor women attending cardiac rehabilitationmay improve adherence. 24Taking into account cultural and racial differencesin the promotion of exercise and diet may help toimprove adherence by ethnic groups torehabilitation. 123 Translation of educationalmaterials and presentation in an appropriate waymay improve adherence in ethnic minoritygroups. 189Other forms of rehabilitation based around riskstratification and case management are suggestedas methods to improve patient adherence, 135,183but the effectiveness of this approach comparedwith outpatient rehabilitation with appropriateoutcome measures is not known. Similarly,provision of support at a women’s retreat mayserve to promote lifestyle change, although itseffectiveness as an adjunct to outpatient care hasnot been evaluated. 140Providing classes at times to suit patients mayimprove adherence to cardiac rehabilitation.Patients may find it easier to attend classes timedbefore work and in the evening. 184–186The systematic review of the literature found fewstudies of sufficient quality to make specificrecommendations of methods to improveadherence to outpatient cardiac rehabilitation andits components. The most promising approach wasthe use of self-management techniques basedaround individualised assessment, problemsolving, goal setting and follow-up. This is mostlikely to be effective in improving specific aspectsof rehabilitation, including exercise and diet.Further investigation of this approach may be bestcarried out by a systematic review of selfmanagementinterventions in less specific patientgroups than considered here. Patient commitmentto attend did not suggest benefit in the promotionof adherence to aspects of outpatient cardiacrehabilitation, but may be useful in improvinguptake of rehabilitation. Other interventionsidentified in the literature may already bestandard practice: use of educational video andclasses, and psychological support are features ofthe modern rehabilitation programme. Similarly,47© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve adherence to cardiac rehabilitationspouse and family support may be provided as anadjunct to a rehabilitation programme for reasonsunrelated to patient adherence.Perhaps the most disappointing outcome of thereview is the dearth of literature reporting theevaluation of simple interventions aimed atimproving adherence to cardiac rehabilitation forall patients or specific groups of patients. Nointerventions were reported to address thefrequently cited patient reasons for non-adherenceof perceived recovery, illness, transport difficulties,inconvenient timing or care of dependentrelatives. Similarly, no evaluations were identifiedof programmes designed to improve adherencefor specific patient groups frequently underrepresentedin outpatient cardiac rehabilitation(including women, the elderly and ethnic minoritygroups). The lack of published studies may reflectan under-appreciation by both rehabilitation staffand journal editors of the value of trials inevaluating new interventions to improveadherence to cardiac rehabilitation. Surveys andaudits suggest that programme coordinators mayrecognise deficits and the need for improvementsin services and implement changes to provisionwithout formal evaluation. The ineffectiveness ofseveral types of intervention to improve adherenceto cardiac rehabilitation identified in thissystematic review demonstrates that innovations inservices should be tested in well-designed studies.ConclusionsThe systematic review identified few studies ofsufficient quality to assess the effectiveness ofinterventions to improve adherence to cardiacrehabilitation. Half of the studies found were insources outside the mainstream of medicalliterature.Self-management techniques suggested somevalue in the promotion of specific aspects oflifestyle change and a further review in a broadercontext of health and disease may be appropriate.Educational interventions aimed at improvingadherence gave equivocal results and suggest thatthe format of the intervention merits furtherstudy.Observational studies identify many areas whereinterventions may serve to improve patientadherence to cardiac rehabilitation, and surveysand audits show that interventions have alreadybeen implemented. The systematic review of theliterature suggests that, before implementation,interventions should be evaluated in wellconductedstudies with economic assessment, andthe results disseminated widely and reviewedregularly.48

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 10Systematic review of interventions to improveprofessional compliance with cardiac rehabilitationBackgroundBarriers to attendance at and adherence withoutpatient cardiac rehabilitation associated withservice factors have been identified. However, theeffectiveness of interventions to improveprofessional compliance with the provision ofcardiac rehabilitation has not been assessed bysystematic review. This systematic review aims toassess the effectiveness of methods for increasingprofessional compliance with cardiac rehabilitationand to identify areas meriting further research.The review includes interventions to encouragehealthcare professionals to comply with guidelinesor good practice regarding invitation and supportof patients’ cardiac rehabilitation.ResultsStudies included in the review ofinterventions to improveprofessional compliance with cardiacrehabilitationEighteen articles reporting 17 studies wereidentified as relevant to the review of methods toimprove professional compliance with cardiacrehabilitation and were formally included in thereview. Reading by three reviewers (SE, FT andAB) found six studies reporting evaluation of anintervention relating to an appropriate patientgroup and with a relevant outcome. The flow ofarticles through the review process is shown inaccordance with QUOROM in Appendix 13. 101Studies were characterised by type and size,participants, intervention, comparison group,principal and other outcomes, and authors’conclusions.A brief summary of studies is presented in Table 24,with further details in Appendix 14. More thanone report was identified from the trial of Jollyand colleagues 102–104 and the reference providingthe main source of information for the systematicreview is cited in the tables and text.Studies excluded from the review ofinterventions to improveprofessional compliance with cardiacrehabilitationEleven studies selected for data extraction but notincluded in the review are summarised inAppendix 15. More than one report was identifiedfrom the trial of Campbell and colleagues. 125,126The excluded studies either had no relevantoutcome data, 116,117,125–127,164,171,191 provided onlydescriptions of services with no outcomes 119,175,192or were retrospective in design. 116Methodological qualities of studiesincluded in the review of interventionsto improve professional compliancewith cardiac rehabilitationSix studies were identified that evaluatedinterventions to improve professional compliancewith cardiac rehabilitation. Two reported RCTs. Inone trial randomisation was on an individualbasis, 124 but no other information on the methodof randomisation, blind outcome assessment orbaseline characteristics of groups was reported. Inthe other trial patients were randomised bygeneral practice. 104 The authors of this trialdescribed methods of randomisation, blindoutcome assessment and baseline characteristics ofgroups. Loss to follow-up was low in this study.None of the other studies reported loss to followup.Four studies described outcomes in periodsbefore and after implementation of anintervention. 41,111,122,123 Baseline groupcharacteristics for appropriate periods were notreported in any of these studies.In three studies the outcome wasattendance, 41,104,111 in two referral 122,123 and inone patient commitment to attend cardiacrehabilitation. 124 Four studies included onlymyocardial infarction patients. 41,111,122,124 Onestudy included myocardial infarction and anginapatients 104 and another only post revascularisationpatients. 12349© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve professional compliance with cardiac rehabilitationTABLE 24 Studies evaluating interventions to improve professional compliance with cardiac rehabilitationAuthors, year Study type and Intervention Findings relevant to Commentsand country patients uptakeRCTsJolly et al.,1999 104UKCluster RCT, 67general practices,597 MI andangina patientsLiaison nurse supportspractice nurses andencourages patients to seepractice nurse afterdischarge. Patient-heldrecord card to prompt andguide follow-up42% of patients in theintervention group attendedat least one outpatient CRsession compared with 24%of controls (p < 0.001)Multifacetedintervention.Management of patientsin control practices notexplicitSuskin et al.,2000 124CanadaRCT, 50 patientsAttending physician provideswritten endorsement62% of patients in theintervention group gavecommitment to participate inCR compared with 38% inthe control group (p = 0.08)Abstract onlyNon-randomised studiesKalayi et al.,1999 122UKBefore-and-afterstudy, 561 MIpatientsElectronic referral pathwaywith feedback to ward staffon referral ratesAfter intervention referralincreased from 194/298(65%) to 208/263 (79%)(p = 0.0002)Disparity between longtermand short-termreferral ratesMosca et al.,1998 41USABefore-and-afterstudy, 199 MIpatientsPrompt for outpatient CR indischarge critical carepathwayCritical care pathwayassociated with a nonsignificantincrease inoutpatient CR participation(OR 1.9, 95% CI 0.6 to 5.5)No baseline groupcomparisonsCaulin-Glaser &Schmeizel,2000 123USABefore-and-afterstudy. Postrevascularisationpatients. Patientnumbers notspecifiedEducational intervention forhealthcare providers on thecomprehensive nature andbenefits of CR. Instructionsfor nurses to discuss CR withpatients and encouragediscussion of referral withphysiciansIn-hospital referral increasedby 50% (p < 0.05). Physicianoffice referral increased by61% (p < 0.05)Abstract onlyScott et al.,2000 111AustraliaBefore-and-afterstudy, 649 MIpatientsDissemination of clinicalguidelines to hospital staffand general practitioners.Feedback on clinicalindicatorsAfter intervention outpatientCR utilisation increased from24% to 54% (p = 0.003)Baseline periodcorresponds to CRprogramme start-up50Themes identified from the review ofinterventions to improve professionalcompliance with cardiac rehabilitationThree themes of interventions were identified inthe systematic review: improvement of the referralprocess, coordination of transfer of care, andphysician endorsement.Improvement of referral processFour studies were identified that evaluatedmethods to improve the referralprocess. 41,111,122,123 In a study comparing periodsbefore and after the introduction of an electronicreferral pathway Kalayi and colleagues 122 observeda significant increase in patient referral to cardiacrehabilitation. The intervention was initiated witha referral section on the electronic patient recordof patients discharged with a diagnosis ofmyocardial infarction. Subsequently, feedback onreferral was given to ward staff. No information ongroup characteristics before and after interventionwas provided and large differences betweenmonthly and longer term referral rates suggest thepresence of other sources of referral variability.Mosca and colleagues 41 compared patientparticipation before and after the introduction of a

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 25 Resource implications of intervention to improve professional compliance with cardiac rehabilitationStaff Equipment Consumables Patient costsCoordination of postdischarge careLiaison nurse support forpractice nurses (Jollyet al., 1999 104 )Three cardiac liaisonnursesTraining for cardiacliaison nurses andpractice nursesThe study employedthree nurses involving atotal of 277 patients in18 monthsprompt for cardiac rehabilitation in a dischargecritical care pathway. An improvement inparticipation in outpatient cardiac rehabilitationwas observed, but this was not statisticallysignificant. Group characteristics were notreported and other factors may have influencedlevels of participation. Caulin-Glaser andSchmeizel 123 reported the implementation of aneducational intervention for healthcare providers.Information on cardiac rehabilitation, including itscomprehensive nature and benefits, was given tomedical and nursing staff and on health outcomesand cost-effectiveness to members of the clinicalcardiology council. After the intervention both inhospitaland physician office referral weresignificantly increased. Again, no baselineinformation to assess comparability of patientgroups was provided. In the study of Scott andcolleagues 111 patients admitted to hospital in threeperiods were compared. These were before, duringand after the dissemination of clinical guidelinesand feedback of clinical indicators to healthprofessionals. The cardiac rehabilitationprogramme was operational during theimplementation period and this was used as thebaseline period for evaluation. A steady increase inutilisation of the outpatient cardiac rehabilitationservice was observed during the implementationperiod and the authors attribute this to theintervention. However, no comparison of patientcharacteristics was available for the relevantperiods and, although the authors report that thenew cardiac rehabilitation service was fullyoperational, an increase in uptake might beexpected with the new service.Coordination of postdischarge careJolly and colleagues 104 reported a cluster RCT ofcoordination of care of myocardial infarction andangina patients between hospital and generalpractice by specialist cardiac liaison nurses.Attendance at one or more cardiac rehabilitationsessions was significantly increased in theintervention patients. The intervention consistedof three main elements: liaison nurse support forpractice nurses, liaison nurse encouragement forpatients to see the practice nurse, and promptsand guidance for patients by means of a personalrecord card. The study design does not allow theeffect of components to be assessed individually.Physician endorsementSuskin and colleagues 124 conducted an RCTcomparing attending physician cardiacrehabilitation endorsement with a genericendorsement. The intervention was associatedwith a non-significant increase in patient-reportedintent to participate in cardiac rehabilitation. Nobenefit was observed for in-person delivery of theendorsement. Little information on the conduct ofthe trial or patient characteristics was reportedand the outcome of intention to attend issomewhat removed from actual attendance atcardiac rehabilitation.Resource implications of intervention toimprove professional compliance withcardiac rehabilitationInformation provided on resource use associatedwith the only effective intervention of reasonablequality is summarised in Table 25. As describedearlier, in the study by Jolly and colleagues 104three liaison nurses were employed withresponsibility for the coordination of postdischargecare of 277 patients over 18 months, a yearlyaverage of 62 patients per nurse. Transport costsincurred by liaison nurses visiting practices and bythe practice nurses attending training and supportgroups were not quantified, nor was the resourceinput or cost of training liaison and practicenurses.Further interventions that mayimprove professional compliance withcardiac rehabilitation suggested in theliteratureThe literature review identified a number ofsuggested interventions for improving professionalcompliance with cardiac rehabilitation. Althoughthese potential interventions were not evaluatedthe studies provided some evidence to suggestmethods meriting further investigation. Examples51© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve professional compliance with cardiac rehabilitationTABLE 26 Further interventions that may improve professional compliance with cardiac rehabilitation suggested in the literatureAuthors, year Intervention Purpose of studyParks et al., 2000 32 Appointment of CR programme director to lead, audit and Auditcommission appropriate resourcesParks et al., 2000 32 Programme run in accordance with national guidelines AuditYoung & Kahana, 1989 46 Physicians and insurers educated on benefits for patient Retrospective observationalBittner et al., 1999 37 groups Retrospective observationalComoss 1988 185 Referring physicians involved in programme ReviewStokes, 2000 193 Education for CR coordinators and staff ReviewParks et al., 2000 32 Explicit criteria for CR eligibility AuditKing & Teo, 1998 194 Streamlining of referral ReviewParks et al., 2000 32 Centralised CR attendance and contact records AuditLevknecht et al. 1997 116 Clinical pathway and clinical quality improvement tool Programme descriptionCannistra et al., 1995 128 Early social services involvement to improve social support Prospective comparisonand hence uptake of CREffron et al., 1986 195 CR commenced earlier Retrospective observationalRoitman et al., 1998 135 Removal of time restriction for start of programme Review52of interventions excluded from the review at bothformal extraction and the earlier inclusion stage,but with possible value in improving professionalcompliance, are summarised thematically inTable 26.DiscussionThe healthcare professional has a pivotal role inrecruitment of patients to cardiac rehabilitation,and their contribution is dependent on education,compliance with guidelines and coordination ofservices. Few studies aimed at improvingprofessional compliance with outpatient cardiacrehabilitation were found. Three studies werepublished in peer-reviewed journals and three asconference abstracts.Evaluations of three types of intervention wereidentified by systematic review: improvement ofthe referral process, coordination of postdischargecare, and physician endorsement of cardiacrehabilitation.None of the four studies reporting interventionsto improve the referral process included adequatemethodological information. The use of a beforeand-afterstudy design might have provided someevidence on the effectiveness of interventions, butthe lack of group comparisons and programmefactors influencing patient attendance precludesthis. This is disappointing, as the inclusion ofcardiac rehabilitation in a critical care pathwayeffective in promoting discharge medication isappealing. Similarly, improving referral bydissemination of clinical guidelines andsubsequent feedback of clinical indicators tohealth professionals merits further evaluation.A multifaceted approach to the coordination oftransfer of care from hospital to general practiceincluding liaison nurse support for practice nurseswas effective in improving cardiac rehabilitationuptake in a randomised trial. Particular aspects ofthe intervention were not evaluated separately,and it is not possible to compare the relativeimportance of professional support of practicenurses, in-hospital nurse–patient contact and selfempowermentof patients with record cards. Thepossibility of referral of patients from generalpractice suggests that the involvement of practicenurses may be of particular value in the referral ofangina patients who have not been admitted tohospital.The value of physician endorsement inencouraging patient participation in cardiacrehabilitation was not confirmed. However, somesupport for further evaluation is suggested by therandomised trial of Suskin and colleagues, 124 wherea non-significant tendency for increased patientcommitment was seen in patients who had receivedan endorsement from an attending physician.

Health Technology Assessment 2004; Vol. 8: No. 41Systematic review of the literature did not identifyany well-evaluated methods specifically aimed atimproving professional compliance to cardiacrehabilitation. One multifaceted approachsuggested benefit, but the importance of theintervention relating to improvement inprofessional compliance could not be distinguishedfrom other patient-directed aspects of theintervention. The resource and therefore costimplications of this intervention are also unclear.All interventions identified were aimed atimproving overall cardiac rehabilitationattendance. It was surprising that no evaluationsof interventions targeted at service improvementsfor specific patient groups were reported.Frequently under-represented patient groupsinclude women, the elderly, ethnic minorities andpatients with more severe presentation of diseaseor co-morbidity.Although few trials were found with the intentionof improving cardiac rehabilitation uptake andadherence by improving professional compliance,several areas for intervention are suggested in theliterature. Many may already be regular practice,but some may have application in the provision ofservices to under-represented groups.Uptake of rehabilitation services is influenced bythe knowledge and enthusiasm of the physicianand providers in the referral process. 46,62,194Consequently, education of physicians andproviders on the benefits of cardiac rehabilitationmay help to improve referral and uptake. 37,46 Thismay be accomplished best by the involvement ofreferring physicians in the programme. 185 Acoherent approach to the education of programmecoordinators and staff on the benefits of cardiacrehabilitation and its application in patient groupsmay lead to better understanding of patienteligibility and thus wider invitation. 193Appointment of a programme director to lead,audit and commission appropriate resources forcardiac rehabilitation may lead to improvementsin service management and provision. 32 This mayfacilitate the running of programmes inaccordance with national guidelines, which mayhelp to improve provision. 32Although none of the studies included in thesystematic review was of adequate quality tosuggest that the use of clinical pathways may be ofvalue in the management of cardiac rehabilitationreferral, this approach may merit furtherinvestigation. The use of clinical pathways withexplicit criteria for patient eligibility may be anappropriate way to manage and streamlinereferral and invitation. 32,116,194Flexibility in timing of care and support may beimportant in improving uptake of services. Visits athome by healthcare professionals may serve toprovide continuity of care and improve uptake ofcardiac rehabilitation. 128 The provision of cardiacrehabilitation early after discharge may coincidewith the time of patients’ greatest need forsupport and greatest motivation, and earlyinvitation and provision may be rewarded byincreased uptake. 195 Some patients may not find aparticular date for commencing rehabilitationsuitable, and flexibility and removal of timerestrictions may lead to an increase in uptake. 135The general scarcity of evaluated methods mayreflect an under-appreciation of the value of trialsin evaluating new interventions. Programmecoordinators may recognise deficits and the needfor improvements in services and implementchanges without formal evaluation. For example, itmay be assumed that incorporation of a promptfor referral in a discharge summary would be aneffective way of ensuring referral. However, thisdoes not necessarily mean that the crucialoutcome of increased patient uptake andattendance at cardiac rehabilitation will beachieved. The systematic review of the literaturesuggests that well-designed studies are required totest interventions aimed at improving professionalcompliance with cardiac rehabilitation.ConclusionsLittle research has been conducted aimed atimproving professional compliance with cardiacrehabilitation. The systematic review identified fewstudies that specifically looked at improvingpatient uptake and adherence by intervening atthe level of healthcare professional activities.The conduct of the healthcare professional iscentral in the recruitment of patients to cardiacrehabilitation and their contribution is dependenton education, compliance with guidelines andcoordination of services. Changes within cardiacrehabilitation services aimed at improving patientuptake and adherence should be evaluated in welldesignedstudies and the results disseminated andreviewed; otherwise, ineffective and inappropriatemethods may become routine clinicalpractice.53© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 11Health service costs of cardiac rehabilitation inthe UKObjectives●●●●To estimate the health service costs associatedwith cardiac rehabilitation programmes in theUK.To estimate the national budget attributable tooutpatient cardiac rehabilitation in the UK.To explore how coverage could be increased ifdifferent configurations of service wereprovided within the existing budget.To explore how coverage could be increasedwith additional funding.Health service costs associatedwith cardiac rehabilitationMethodsThe costs associated with the provision of cardiacrehabilitation services from the health serviceperspective were estimated by considering thestaff, overheads, building capital and equipmentcosts. The costs borne by patients such as travelcosts or expenses for special clothing were notincluded, but it was recognised that these couldhave important implications if patients perceivedthem to be large enough to deter their attendance.The BACR/BHF survey, described in Chapter 4,provided information on the typical number ofhours per week by broad staff categories spent inoutpatient (phase 3) cardiac rehabilitationprogrammes. The additional questionnaireprovided information on the number of patientsreferred, joining and completing cardiacrehabilitation, the total number of sessions, andnumber and length of sessions per week.Centres that responded to the short questionnaireand provided information on staff input (n = 186;65% of all UK centres identified by BACR/BHF)were stratified by a criterion of multidisciplinarityof staff input. This was based on the assumptionthat a greater variety of staff input is a proxy forhigher service quality. The following professionalcategories were considered to be relevant to anoutpatient cardiac rehabilitation programme(hereafter referred to as ‘key staff ’):●●●●●●●physician (GP, cardiologist, general physician)nursephysiotherapist/sport scientistoccupational therapistpsychologistdietitianpharmacist.Three groups were defined according to thenumber of different types of key staff: group 1having more than five different types of key staff,group 2 having three to five types of key staff, andgroup 3 having two or fewer. The total number ofcentres in each of these groups is shown inTable 27. Ten centres within each group werechosen at random to conduct a more detailedcosting study.Staff costsAll 30 centres in the random sample werecontacted between April and June 2002 andprovided more detailed information on the gradesof staff working in 2000 (the year of theBACR/BHF survey). Staff costs were estimated bymultiplying the average numbers of hours perweek worked for each grade of staff by the hourlypay for that grade. Hourly pay rates werecalculated by dividing the midpoint of the relevantpay scale by the numbers of hours of expectedwork per annum, excluding annual leave, bankholidays, and training/study and sickness days. AllTABLE 27 Stratification of cardiac rehabilitation centres (n = 186) by number of different types of key staffGroup No. of different types of key staff No of centres %1 >5 38 20.42 3–5 135 72.63 ≤ 2 13 7.055© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve professional compliance with cardiac rehabilitationTABLE 28 Average hours per week by staff categoryStaff grade Group 1 a Group 2 b Group 3 c(n = 10) (n = 10) (n = 10)Nurse grade:B 0.95E 7.2F 11.38 12.75 3G 21.40 16.15 9.25H 13.05 5.75 0.4Physiotherapist:Helper 2Basic 0.20Senior I 10.05 5.4 2.7Senior II 1.4 1.2Superintendent III 2.6Superintendent IV 0.6Sport scientist 1.9Exercise physiologist 3.8 0.4Occupational therapist:Basic 0.8Senior 2.7 0.05 0.3Head 0.6Dietitian 0.47 0.18Senior dietitian 0.58 0.46Pharmacist 0.46 0.35Physician 0.38 0.75Clinical psychologist 0.67 0.3Cardiac technician 0.7 2 0.5Social worker 0.1Secretary 4.2 5.04 0.3Total (SD) 74.7 (5.8) 62.6 (4.7) 18.0 (2.9)a Centres with more than five key staff.b Centres with three to five key staff.c Centres with two or fewer key staff.56pay scales were those prevailing on 1 April 2001and included employers’ on-costs (employers’contribution to national insurance plus 4% ofsalary contribution to superannuation). 196Salaries information was taken fromwww.nhscareers.nhs.uk. 197 A detailed summary ofunit cost estimates for different staff categoriesand grades is shown in Appendix 16.Non-staff-related costsNon-staff-related costs refer to the overheads,building capital and equipment costs associatedwith running cardiac rehabilitation services. Mostcardiac rehabilitation services tend to use anumber of different facilities to deliver thedifferent components of the programme and donot have these figures readily available. Hence,allowances for indirect overheads (the costs of thesupport services such as human resources, financeand estates required to carry out the servicesmain functions) and building capital (the costsassigned to treatment and non-treatment space)relative to the midpoint of the relevant pay scalewere based on Netten and colleagues 196 (seeAppendix 16).The required equipment was based on currentrecommendations from the BACR (see Appendix17). The unit costs were obtained from thecoordinator of the cardiac rehabilitation team ofthe Bristol Royal Infirmary. An equivalent annualcost was estimated by using an annuity factor of6% and assumed lifespan of 5 years. Annual costsaccounted for approximately £861 [value addedtax (VAT) included].Direct overheads, that is, the costs associated withlighting, heating and cleaning, were assumed to be11% of the sum of staff costs, indirect overheads,building capital and equipment costs. This wasbased on previous studies carried out in hospitalsettings where the direct overheads were found toaccount for 4–18% (midpoint 11%) of totalcosts. 198–200

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 29 Service provision, referral, uptake and completion rates for 30 UK cardiac rehabilitation centres in 2000 (stratified bystaff mix)Group 1 Group 2 Group 3(n = 10) (n = 10) (n = 10)Mean Median Mean Median Mean MedianHours per patient 29.0 27 24 215 20 17.5No. referred 282.4 289 352.5 255 170.7 150No. joined 157.3 148 194.3 172 97.9 104% of referrals 56 51 55 67 57 69No. completed 126.3 104 158 150 89 92% of referrals 45 36 45 59 52 62TABLE 30 Average cost estimates for cardiac rehabilitation (2000/01 prices)Costs (£) Group 1 Group 2 Group 3 Weighted costsStaff Total Staff Total Staff Total Staff Totalcosts costs costs costs costs costs costs costsPer year/centre 53,100 72,700 42,100 57,400 12,400 17,600 42,300 57,700Per patient referred 243 330 137 186 127 249 157 220Per patient joined 421 571 236 320 174 324 269 371Per patient completed 542 738 317 429 186 344 354 486Per hour 20 27 14 20 14 30 15 22The total cost of cardiac rehabilitation wasestimated for each centre, and the cost per patientreferred, joined and completed estimated. Thecost per hour was also estimated, taking the costper patient completing the programmes as thedenominator. The costs of each centre within agroup were then averaged. A weighted averagecost was also estimated, using the proportion ofcentres nationally falling within each group (seeTable 27).ResultsInformation on the weekly staff input by staffcategory and grade for each centre in the randomsample is shown in Table 28 (a detailed summaryof staff resource data for each centre is given inAppendix 18). Group 1 had higher levels ofweekly staff input (75 hours) than centres ingroups 2 (62 hours) and 3 (18 hours).The average duration of the programmes bygroup is shown in Table 29. Centres that employmore than five different key staff (group 1) providethe most intensive service per patient, with anaverage duration of 29 hours per patient. Thiscompares to 24 hours per patient for group 2centres and 20 hours for group 3 centres (detailsgiven in Appendix 19). Table 29 also gives, bygroup, the absolute number of patients referred,joining and completing cardiac rehabilitationprogrammes. This varied widely, with the highestaverage numbers being in group 2.The average staff costs and average total costs ofcardiac rehabilitation are presented for each groupin Table 30 (more details shown in Appendix 20).There was a considerable difference in the cost ofan average centre and in the average cost perpatient between each group. The total averagecost per patient completing the programme was£542 for group 1, £317 for group 2 and £186 forgroup 3. Staff costs accounted for 73% of the totalcosts for centres in groups 1 and 2, and 70% forcentres in group 3 (based on cost per centre).Figure 4 illustrates that nursing costs are the mostimportant share of total staff costs, accounting forabout 62% of total staff costs in group 1 centres,67% in group 2 and 71% in group 3.Physiotherapy costs are the second most importantshare of total costs, accounting for about 23% ingroup 1 and group 3 centres, and for about 14%in group 2 centres.Weighted average staff and total costs are shown inTable 30. The weighted average cost per patientcompleting a cardiac rehabilitation programmewas £354 (staff costs only) and £486 (total costs).57© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve professional compliance with cardiac rehabilitationProportion of total costs (%)100908070605040302010OthersPhysicianPsychologyDieteticsPharmacyOccupational therapyPhysiotherapyNursing01 2 3GroupFIGURE 4 Proportion of total staff costs attributable to staff categories58The national budget attributableto cardiac rehabilitationMethodsThe budget attributable to outpatient (phase 3)cardiac rehabilitation was estimated separately forEngland, Wales, Scotland and Northern Irelandusing data from the BACR/BHF survey and theadditional questionnaire (as described in Chapter4). The total number of patients completing acardiac rehabilitation programme was estimatedfor each country. Where centres did not providedata, the IQR derived from responding centres forthat country was used to calculate total numbers.The budget for each country was estimated bymultiplying the number of patients (reported aslower and upper bound) completing a cardiacrehabilitation programme by the weighted totalaverage costs per patient completing cardiacrehabilitation (£486; see Table 30).These estimated budget figures were then used toexplore how coverage could be increased if adifferent configuration of cardiac rehabilitationservices were provided. An estimate was made ofthe number of patients that could be treated ifrehabilitation services were entirely provided incentres with three to five key staff (as in group 2,see above) or in a different scenario, in centreswith two or fewer key staff (as in group 3). Inaddition, an estimate was made of the necessarybudget increase to provide cardiac rehabilitationto all potentially eligible patients using data fromthe analysis presented in Chapter 3. Thisestimation was undertaken by assuming that, first,cardiac rehabilitation would be uniformly providedin group 2 centres and, second, cardiacrehabilitation would be uniformly provided ingroup 3 centres.This study also explored how additional fundingcould increase coverage. If unit costs per patientfall as the number of patients completing therehabilitation programme rises (i.e. centresexperience economies of scale), additional fundingwill imply that the number of extra patientstreated is higher than proportionate. A possibleassociation between staffing costs per patient andthe number of patients treated was, therefore,examined. First, the log-transformed costs andlog-transformed numbers of patients were plotted,as both variables have a log-normal distribution.Secondly, a simple log-linear regression model wasused to estimate the relationship between costs perpatient, as the dependent variable, and the annualnumber of patients completing the rehabilitationprogramme, as the independent variable. Anadditional model was estimated, controlling fornumbers of staff employed on cardiacrehabilitation. The regression coefficient fornumber of patients in these log-linear regression

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 31 Estimated budget attributable to cardiac rehabilitation by country (2000/01 prices)England Wales Scotland NorthernIrelandEstimated no. of patients completing 25,700–39,000 2,100–3,500 2,500–4,700 1,000–1,400outpatient cardiac rehabilitation aEstimated budget attributable to outpatient £12,513,000– £1,018,000– £1,222,000– £487,000–cardiac rehabilitation 18,975,000 1,683,000 2,276,000 658,000a Numbers were estimated by using information form the BACR/BHF survey for 2000. Data for non-responding centreswere imputed by IQR for the relevant country.TABLE 32 Estimated impact of a change in service configuration given current budgetEngland Wales Scotland NorthernIrelandEstimated no. of patients completing outpatient 25,700–39,000 2,100–3,500 2,500–4,700 1,000–1,400cardiac rehabilitation in 2000 (current serviceprovision)Estimated no. of patients able to be treated 26,100–44,200 2,400–3,900 2,800–5,300 1,100–1,500with a group 2 type service (i.e. three to fivekey staff) without expanding budgetEstimated no. of additional patients if all treated 4,300 400 500 200with a group 2 type service (based on themidpoint of the ranges reported above)% increase in coverage 13 13 13 13Estimated no. of patients able to be treated 36,400–55,100 3,000–4,900 3,600–6,600 1,400–1,900with a group 3 type service (i.e. two or fewerkey staff) without expanding budgetEstimated no. of additional patients if all treated 13,400 1,100 1,500 500with a group 3 type service (based on themidpoint of the ranges reported above)% increase in coverage 41 41 41 41models measures the elasticity of the cost perpatient with respect to the number of patientscompleting the programme, that is, thepercentage change in costs for a given percentagechange in number of patients.ResultsThe estimated budgets attributable to cardiacrehabilitation by country are shown in Table 31.The current budget was estimated to beapproximately £12.5–19.0 million in England,£1.2–2.3 million in Scotland, £1.0–1.7 million inWales and £0.4–0.7 million in Northern Ireland.Overall, this would result in a budget estimate of£15.2–23.6 million for outpatient cardiacrehabilitation for the whole of the UK.Table 32 shows the estimated impact of a change inservice configuration for two different scenarios bycountry. It was estimated that approximately 5,300more patients across the UK could be treated ifthe service were provided in cardiac rehabilitationcentres with the staffing level of those in the‘group 2’ sample. This corresponds to a 13%increase in coverage compared with the currentsituation. If services were provided with cardiacrehabilitation centres with low staffing levels(group 3), approximately 16,490 more patientscould be treated, corresponding to a 41% increasein coverage compared with the current situation.As shown in Chapter 3, around 266,800 patientswere potentially eligible for cardiac rehabilitationin England in 2000. Assuming that group 2services were uniformly provided, an annualbudget of approximately £115 million would berequired for the provision of cardiac rehabilitationto all patients. This represents a 630% increase inthe estimated current budget attributable tocardiac rehabilitation.59© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve professional compliance with cardiac rehabilitationTABLE 33 Regression model for staff costs per patient completing cardiac rehabilitation (ln costs) SE 95% CI pConstant 5.97 0.643 4.647 to 7.286

Health Technology Assessment 2004; Vol. 8: No. 41800Staff costs per patient (log scale)40020050100 200 400Annual no. of patients completing CR (log scale)FIGURE 5 Relationship between size (as measured by annual throughput of patients) and unit cost (staff costs) of cardiacrehabilitation programmes (all centres)Group 1 Group 2800400200Staff costs per patient (log scale)50800400Group 3100 200 400 100 200 40020050100 200 400Annual no. of patients completing CR (log scale)FIGURE 6 Relationship between size (as measured by annual throughput of patients) and unit cost (staff costs) of cardiacrehabilitation programmes by group61© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Systematic review of interventions to improve professional compliance with cardiac rehabilitation62rehabilitation in the UK, they have been lesscomprehensive in their cost estimates.Nonetheless, the present findings are consistentwith their findings, which suggested an averagecost of cardiac rehabilitation per treated patient inthe region of £200–£400. 25,79–82For example, Gray and colleagues 80 estimated acost of £371 per patient completing a cardiacrehabilitation programme (median £223; 1994prices). This was based on a sample of 16 cardiacrehabilitation centres in England and Wales, buttheir estimate excluded non-staff costs andcontributions by non-specifically funded staff.Average staff costs of £350–425 (2001 prices) wereestimated by the Scottish Intercollegiate GuidelinesNetwork (SIGN) guideline development group,assuming 500 referred patients per year and a 90%uptake. 2 Based on the funding information givenby 37 centres in the most recent BACR/BHFsurvey, Evans and co-workers 25 reported a cost of£50–712 (median £256) per patient completingcardiac rehabilitation.By using information from the BACR/BHF surveyand an additional questionnaire, all staffcontributions could be measured and valued. Theanalysis shows that centres with higher levels of staffmix provide a more expensive service per patienttreated than centres that employ fewer types of keystaff. This is due to a longer duration of cardiacrehabilitation programmes offered by centres ingroup 1 (29 hours per patient) compared with theaverage duration of programmes offered by centresin group 2 (24 hours) and group 3 (20 hours), aswell as the higher weekly staff input intoprogrammes offered by centres with more types ofkey staff. This did not correspond with highernumbers of patients entering and completing theprogrammes. Thus, patients treated in centres witha higher level of multidisciplinarity received themost intense rehabilitation programme in terms ofthe duration of the programme and staff/patientratio. Although the heterogeneity of cardiacrehabilitation services has long been acknowledged,evidence is lacking to suggest that programmeswith a higher level of multidisciplinarity offerimproved patient outcomes. Such services may noteven represent higher service quality; for example,adherence was lower in the most multidisciplinaryand intensive services.This analysis has some limitations. Ideally,information on non-staff related costs such as directoverheads and capital costs should have beenobtained from each rehabilitation centre in thesample. From a practical point of view, this was notfeasible. The advantage of this study is the size ofthe sample, which allowed the cost differences dueto different staff configurations to be explored indetail.The costs of equipment were also included, basedon current recommendations from the BACR.However, this list did not include equipment forundertaking ECG-exercise testing. Cardiacrehabilitation centres may carry out exercise testingbefore and after cardiac rehabilitation to assesspatients and will, therefore, incur higher costs.Annual costs of equipment (treadmills, consumablesfor ECG, etc.) have been estimated to account forapproximately £25,000 (Sally Turner, Alton CardiacRehabilitation Centre: personal communication,2 December 2002).Some extrapolations had to be made to calculatethe total numbers of patients completing aprogramme, because not all centres replied to thesurvey and provided comprehensive activity data.This is possibly due to their lack of automatedsystems to extract these data, lack of audit facilitiesor being in the process of installing systems tocollect audit data to satisfy the requirements of theNSF-CHD. However, the authors believe that theirestimate of an annual budget attributable to cardiacrehabilitation of £15.2–23.6 million for the whole ofthe UK is a refined update of previous budgetestimates, for example, £8–34 million by Taylor andKirby 81 based on a converted US cost estimate.The results of the simple budget analysis show thatby providing a service as offered in group 3 centres,the overall service provision could be increased byapproximately 40% with current funding. Providingcardiac rehabilitation as offered in group 2 centres,which represents the average cardiac service in theUK, could lead to an approximately 13% increasein coverage. This could be of importance givenlimited resources and the large extent of unmetneed, as shown in Chapter 3.The resource implications for extending cardiacrehabilitation to a greater proportion of eligiblepatients and to other groups of patients asrecommended by the NSF-CHD are not clear.Only a minority of centres, as reported in Chapter4, state that they have spare capacity. It is also notobvious whether the difference between thenumber of referred and enrolled patientsrepresents spare capacity, as many centres havewaiting lists that restrict the number of patientsreceiving treatment. Therefore, the extension ofcardiac rehabilitation may require extra resources.The costs of these resources will be dependent on

Health Technology Assessment 2004; Vol. 8: No. 41local factors such as current provision of staff, theopportunity costs of extending the role of existingstaff employed in other areas, existing (spare)capacity and facilities and, if not available, thecosts of hiring facilities, for example in communitysport centres.This study confirms the finding that cardiacrehabilitation centres experience economies ofscale, as first reported by Gray and colleagues. 80However, this was only apparent whenmultidisciplinarity of staff input, as defined by thethree groups, was explicitly taken into account. Thisfinding suggests that any budget increase couldlead to a more than proportionate improvement incoverage of cardiac rehabilitation services.This analysis considers only the direct costs ofcardiac rehabilitation. Future assessments of thecost-effectiveness of cardiac rehabilitation need toconsider the future savings associated withreduced subsequent healthcare utilisation relatedto cardiac disease. The inclusion of future costsrelated to successfully rehabilitated patients livinglonger and requiring health services unrelated tocardiac disease is more controversial. Also to beconsidered are the future productivity gainsassociated, for example, with earlier return towork. The inclusion of costs incurred by patientssuch as expenses for travelling and specialclothing will depend on the perspective fromwhich the costs analysis is conducted. A fulleconomic evaluation requires the comparison ofthe resource use changes with improved healthconsequences, that is, the effectiveness of cardiacrehabilitation.ConclusionsThe average costs of cardiac rehabilitation to thehealth service per patient successfully completinga cardiac rehabilitation programme are about£350 (staff costs only) and £490 (total costs) perpatient. Outpatient cardiac rehabilitationrepresents an NHS cost of between £15.2 and 23.6million in the UK. Cost variation across centres ispartly explained by a higher dose of interventionin terms of duration and staff/patient ratio. Thereis a need to quantify the heterogeneity of servicesin terms of benefits. Trials comparing complexmultidisciplinary rehabilitation with simplerregimens require evaluation of their costs andeffectiveness.If all services were modelled on the most commonconfiguration of staffing (group 2), approximately13% more patients could be treated with the sameannual budget, but if the simpler group 3 serviceswere to be uniformly provided, 40% more patientscould be treated. The levels of need for cardiacrehabilitation, using the more modest criteria ofneed (see Chapter 3), suggest that, at best, fewerthan 30–43% of eligible patients are referred and,of these, about half join cardiac rehabilitationprogrammes. This suggests that the capacity toincrease provision by 40% within current budgetswould meet between 42 and 60% of thepopulation need for treatment.Higher funding would be needed to increaseprovision to match need and to meet NSF-CHDtargets. An approximate 260–630% increase in theannual current budget is required, to treat allpotentially eligible patients depending on thestaffing configurations of the cardiac rehabilitationprogramme. However, increased spending couldlead to a more than proportionate increase incoverage. Further work is required to examine thebest ways of using any increased funding, as it islikely that the potential of different services toincrease capacity will vary markedly, and theassociated costs will differ if, for example, newcapital schemes are required.63© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 12ConclusionsOutpatient cardiac rehabilitation should beavailable to patients with a range ofcardiovascular diagnoses and afterrevascularisation procedures, but previous studieshave shown that uptake is low, particularly in somespecific patient groups. While many barriers toparticipation have been described, theeffectiveness of interventions to improve uptakeand adherence has not been assessed by systematicreview. Furthermore, the cost implications ofinterventions to improve uptake and adherenceand of increasing overall provision to meet totalpopulation need have not been estimated.Conclusions presented here are based around theobjectives set in Chapter 2.What is the population need forcardiac rehabilitation?Population need for cardiac rehabilitation in theUK in 1999–2000 was assessed from hospitaldischarge statistics in England, Wales andNorthern Ireland. The researchers were unable touse equivalent data for Scotland.Two criteria for eligibility for cardiac rehabilitationwere considered: patients with acute myocardialinfarction, unstable angina or following arevascularisation procedure; and all patientsdischarged alive with a primary diagnosis ofischaemic heart disease or followingrevascularisation. The former, more conservativeestimate of need, identified nearly 146,000patients per year as eligible for cardiacrehabilitation in England, Wales and NorthernIreland. The latter gives a considerably largerestimate of 299,000 patients per year, but includespatients with chronic ischaemic heart disease,some of whom may be considered eligible forparticipation in some programmes and who maybenefit from rehabilitation. Although thesepatients are not currently specified as immediatepriorities for cardiac rehabilitation, for example inthe NSF-CHD, many may be deserving ofrehabilitation or appropriate lifestyle advice andmodification as services develop.Who is not receiving cardiacrehabilitation?To estimate the level of cardiac rehabilitationprovision, data from the 2000 BACR/BHF surveyof cardiac rehabilitation centres were combinedwith hospital discharge statistics. The overallresponse rate of survey centres was 67% and IQRswere imputed for non-responders. This gave arange of estimates of numbers of patients referredto and joining a cardiac rehabilitation programme.It was estimated that in England about 53% (range45–67%), in Wales about 72% (range 59–81%) andin Northern Ireland about 30% (range 25–36%) ofacute myocardial infarction, unstable angina andrevascularisation patients were referred to cardiacrehabilitation in 2000. The proportions of patientsjoining a programme were about 33% (range27–41%), 40% (range 38–46%) and 22% (range18–25%), respectively. As this considers only thelimited eligibility criteria as the denominator itreflects an overestimate if centres providedservices to other patient groups. Applying the lessinclusive eligibility criteria of any ischaemic heartdisease or revascularisation, it was estimated thatin England about 26%, in Wales about 32% and inNorthern Ireland about 14% of patients werereferred to cardiac rehabilitation in 2000. Thecorresponding figures for patients joining aprogramme were 16%, 18% and 11%, respectively.A survey of rehabilitation centres suggested thatan average of about 63% of all patients referredjoined a programme and that about 48% ofreferrals completed a course.There appeared to be variation in serviceprovision across the UK, with a higher proportionof eligible patients referred to and joining cardiacrehabilitation programmes in England and Walesthan in Northern Ireland. Since the need forrehabilitation is substantially greater in NorthernIreland (and Scotland), this represents aconsiderable disparity between uptake and need.The data demonstrate that many eligible patientswho may derive benefit are not referred or invited,65© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Conclusions66do not respond to invitation, or do not adhere tocardiac rehabilitation. Some of the shortfall inreferral and invitation may be explained by theclinical eligibility criteria used in selecting patientsas appropriate for cardiac rehabilitation. Thisselection is mainly by health status beforedischarge. From the clinical exclusion data in anRCT with minimal exclusions, about 81% ofpatients were identified as eligible forrehabilitation after myocardial infarction and,although slightly lower than the 85% stated in theNSF-CHD, this is a reasonable overall estimate.The remaining 19% of patients were consideredunsuitable for outpatient rehabilitation, mainly onthe basis of co-morbidity or frailty. This is not tosay that these patients may not gain materiallyfrom secondary prevention or individually selectedcomponents of cardiac rehabilitation. Indeed,many eligible and included patients may only besuited to, and gain benefit from, specific aspects ofrehabilitation.The definition of eligibility is important. In anRCT setting with minimal exclusions andappropriate documentation the eligibility criteriaare clearly defined. In a non-trial setting thepossibility arises that eligibility can be flexible andtake on a role in rationing services. This may, inpart, explain the extra tier of exclusion observedwithin the randomised trial context in that, afterreferral, coordinators tended to exclude olderpatients and those with more severe presentationof coronary heart disease from outpatient cardiacrehabilitation, possibly on the basis of an exercisetest. Clearly, frail, elderly people and those withco-morbidity are capable of benefiting fromrehabilitation, as shown by trials of geriatricassessment and rehabilitation units, and everydaypractice within the NHS. The nature ofrehabilitation for such patients may be lessintensive than for other patients and may involveattendance at a day hospital. Some linkagebetween cardiac rehabilitation and health servicesfor elderly people would be desirable to ensurethat appropriate rehabilitation is available to all,regardless of age.Under-represented groupsIn the national survey and in the RCT, uptake ofcardiac rehabilitation tended to be lower in olderpatients than in younger age groups. However,having attended one class there was no evidenceto suggest that older patients were more likely todrop out of rehabilitation. Women were less likelyto attend in both settings, but in the trial this waslargely explained by the increased age of womenat presentation. It is not possible to draw firmconclusions about the attendance of black or Asiangroups as national database data were incompletefor coding of ethnicity, and in the survey ofrehabilitation centres numbers referred to andattending cardiac rehabilitation tended to be low.Accessibility of informationGathering data on patient need, eligibility andrehabilitation activity was problematic. To simplifythe process and make estimates more precise,national analysis of audit data would be preferableto ad hoc surveys. Unfortunately, audit was foundto be underdeveloped in cardiac rehabilitation.The survey in England showed an uncoordinatedapproach to data collection and audit, withconsiderable variation in methods and content.With the standards set out in the NSF-CHD,reproducible and comparable methods should bein place, but little evidence was found to suggestthat this was so. The use of modern medicalrecords systems and gathering of data with anational and policy-driven standardised tool aredesirable. This would allow assessment of allstages of the rehabilitation process, starting withthe original coronary heart disease diagnosis orprocedure, and would include information onpossible causes of under-representation.Some programme coordinators reported thatdirect referral systems from surgery and clinicswere in place aimed at improving uptake ofcardiac rehabilitation. The use of methods topromote direct referral suggests that audit canbridge the gap between inpatient care andoutpatient cardiac rehabilitation.What is the effectiveness ofdifferent methods of improvinguptake and of differentialtargeting of cardiac rehabilitation?Barriers to participation in outpatientcardiac rehabilitationInterviews with patients randomised to attendrehabilitation in a trial confirmed commonlyperceived reasons for non-attendance at cardiacrehabilitation. The main reasons for nonattendanceor dropout were: lack of interest,illness, transport difficulties, scheduling and careof dependants. These responses suggest that someaspects of non-attendance are amenable tointervention by addressing issues of motivation,perceived relevance of cardiac rehabilitation tofuture well being, co-morbidities, the site and timeof sessions, transport and arrangement of care fordependents.

Health Technology Assessment 2004; Vol. 8: No. 41Interventions used in cardiacrehabilitation centresThe survey of cardiac rehabilitation coordinatorsfound a high level of awareness of the problem oflow uptake. Sixty-six percent of services thatresponded indicated that they had implementedmeasures to improve attendance. Interventionsappropriate to all patients included follow-uptelephone calls, personalised invitation, homevisits and free transport. More specificinterventions for under-represented groups(women, the elderly, ethnic minorities, patientswith heart failure or angina) includedindividualised classes, buddy systems and inclusionin the programme of a spouse or relative.Many interventions were reported and somerepresent the application of common-sensemethods. Nevertheless, studies to show thesustainable effectiveness of interventions arenecessary if the long-term benefits of interventionsare to be confirmed and the value of interventionsdisseminated more widely. The possibility existsthat a common-sense intervention may have anegative effect on attendance. For example,patients collected last and returned home firstmay value free hospital transport as part of theoverall rehabilitation package, whereas patientssubjected to an extended journey and long transittimes may find this an inconvenience thatinfluences subsequent participation.Although the RCT represents the gold standard inthe evaluation of new interventions, this may beconsidered inappropriate by a cardiacrehabilitation professional attempting to provideservices to all patients. As a possible alternative toRCTs, improvement in uptake attributable to anintervention may be identified by audit. However,reproducible audit procedures need to be in placefirst.Systematic review of the literatureTo identify studies of interventions with the aim ofimproving uptake and adherence to cardiacrehabilitation, three systematic reviews werecarried out. The issue of improving uptake andadherence was split into three major questions:how can recruitment to cardiac rehabilitation beimproved (uptake); how can patients’ adherence tocardiac rehabilitation and maintenance of lifestylechanges be improved; and how can professionalsbe encouraged to comply with guidelines andgood practice? These were designed to identifyinterventions to improve all aspects of referral andinvitation, uptake and adherence to cardiacrehabilitation.The comprehensive systematic review of literaturecovered a large range of databases andhandsearches. Studies identified were published injournals, theses and conference abstracts. It wasdisappointing to find that nearly half of thestudies reporting potentially valuable methods topromote cardiac rehabilitation were found only inthe grey literature, with little opportunity foraccess by interested healthcare professionals.Sharing of information is essential if effectivemethods are to be implemented. Of the studiesidentified, a minority were RCTs.Although some studies that looked at alteringpatient behaviour were identified, there was verylittle literature on interventions aimed atencouraging healthcare professional compliancewith guidelines or good practice regardinginvitation and support of patients’ cardiacrehabilitation. As the conduct of the healthcareprofessional is central in the recruitment ofpatients to cardiac rehabilitation it seems logical tostudy interventions relating to professionaleducation, compliance with guidelines andcoordination of services. In one RCT amultifaceted approach to transfer of care fromhospital to general practice was associated withincreased cardiac rehabilitation uptake. However,the relative importance of one specific aspect ofthe intervention directly concerning professionalcompliance could not be evaluated.The systematic literature review identified someinterventions to help improve patient uptake ofcardiac rehabilitation. Invitation letters,pamphlets, telephone calls and home visits may beused to convey a motivational message. Trainedlay volunteers providing support to patients in theperiod before an outpatient programme mayfacilitate subsequent attendance at cardiacrehabilitation.Following successful recruitment of patients tocardiac rehabilitation it is important that patientsadhere to the programme and maintain anyassociated lifestyle changes. Methods based onimprovements in self-efficacy and behaviouralfeedback showed promise in improving andsustaining risk factor management.Possible interventions suggested in the literature,but which have not been evaluated in trials, wereidentified as areas for future research. These werebased on observations in trials, reviews and patientinterviews. Interventions relating to professionalcompliance include education of healthcareprofessionals on the benefits of cardiac67© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Conclusions68rehabilitation, appointment of a programmedirector, use of clinical pathways with explicitpatient eligibility criteria, flexibility in programmestart dates, flexibility in programme times, andhome visiting to provide continuity of care.Suggested interventions to improve uptake includeearly support and planning postdischarge byhealthcare professionals, adaptation of services forunder-represented groups, use of communityfacilities, provision of services in a communitysetting, and for motivated patients withappropriately supervised delivery, home-basedmethodologies. Unevaluated interventions toimprove adherence include further approachesbased on self-efficacy, including demonstrations ofbehaviours by previous patients, which may be ofparticular value in promoting dietary change.Alternative forms of exercise or diet modificationmay help to improve adherence, and this may beespecially useful for women, elderly people andminority ethnic groups. The use of different formsof rehabilitation such as home-based programmesmay be acceptable in highly motivated patientswith less severe coronary heart disease.The identification of so many interventions inneed of evaluation suggests that there is value inthe study of factors determining attendance. Wellconductedqualitative studies in providers andparticularly in patients may be useful inidentifying the attitudes, beliefs and valuesassociated with successful cardiac rehabilitation.Since this review was completed, a qualitativestudy of factors influencing enrolment in cardiacrehabilitation has been published. 201 This studysuggested that physician recommendation,encouragement from family and friends, andaccess to transportation are important factors inpromoting enrolment.What is the potential budgetimpact of increasing uptake ofcardiac rehabilitation usingdifferent uptake interventions?Service duration and configurationThe effectiveness of different intensities andmultidisciplinarity of cardiac rehabilitation is notknown. Systematic reviews include a wide range ofinterventions both more and less intense thancurrent UK recommendations, but to date noattempt has been made to stratify effectiveness byservice model.The BACR/BHF survey of cardiac rehabilitationacross the UK showed wide variations in intensity,programme content and staffing. However, themean levels of service provision suggest that an 8-week programme with 2 hours per week ofexercise training, 1 hour per week of educationand half an hour per week of psychologicalintervention is typical.In the UK, three service configurations wereidentified, based on numbers of different types ofkey staff. A service involving three to five key staffis most commonly provided, with 73% ofprogrammes reporting this configuration. Fewprogrammes had lower staffing levels, but 20% ofprogrammes had more than five key staff. Untilevidence is available on the effectiveness of moreintensive interventions it seems reasonable to baseprojections on the moderate service configurationwith its multidisciplinary structure.Costs of cardiac rehabilitationThe average costs of cardiac rehabilitation to thehealth service per patient successfully completinga programme were estimated at about £350 (staffcosts only) and £490 (total costs) at 2000/01 prices.In the UK this equates to an NHS cost of between£15.2 and £23.6 million. This range representsthe uncertainty in identifying the total number ofpatients receiving cardiac rehabilitation in the UK.The lower figure is the number of patientscompleting a programme in centres whoresponded to the BACR/BHF survey, and thehigher figure is an extrapolation of identifiedservice levels to all known UK programmes. As theBACR/BHF database is an established and wellrespectedresource, it is likely that the nonrespondingcentres are more recent and smallerprogrammes. Consequently, an overall UK costestimate greater than £15.2 million but less than£23.6 million is probable. Again, this highlightsthe importance of consistent national audit inguiding the provision of cardiac rehabilitation.A minority of centres reported spare capacity andthis would have only a small potential impact onoverall provision. On the basis of unmet needidentified in the survey and applying theconservative eligibility criteria of acute myocardialinfarction, unstable angina and revascularisation,it was estimated that a budget increase ofapproximately 260% would be required,representing an overall annual budget of about£56 million at 2000/01 prices. Clearly, this wouldbe considerably greater if more than five key staffwere included in the programme and theeligibility criteria were extended to all patientswith a discharge diagnosis of coronary heartdisease or heart failure, or following

Health Technology Assessment 2004; Vol. 8: No. 41revascularisation. This may imply a budgetincrease of up to 630% and an annual budget ofapproximately £115 million.Other costs may be involved in the extension ofprovision, depending on local factors such ascurrent levels of staffing, the opportunity costs ofextending the role of existing staff, existing sparecapacity and facilities and, if not available, thecosts of hiring facilities, for example in communitysport centres. Conversely, economies of scale mayserve to reduce the extra budget required.Additional costs of increasing uptakeand adherenceOrder of magnitude costs of interventions toimprove uptake and adherence with cardiacrehabilitation may be inferred and these suggest awide range of implied costs. Motivationalinterventions could replace existing methods ofinvitation at minimal cost, and the incorporationof motivational elements into an establishedhome-visiting schedule may have little furtherimplication for resources. The use of layvolunteers in promoting uptake of cardiacrehabilitation is likely to be more costly, withextensive training requirements and travel costs.Similarly, the introduction of liaison nursecoordination of transfer of care would be costly ifrecruitment of new staff was required. However, itmay serve to define the role of the establishedliaison nurse in supporting patients and otherhealthcare professionals and coordination ofpostdischarge care. Strategies aimed at improvingself-management could be incorporated intooutpatient cardiac rehabilitation sessions, andtraining of rehabilitation staff in lifestyleevaluation may serve to formalise assessment andprocedures already in place in cardiacrehabilitation.69© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Chapter 13Key findingsImplications for healthcare● Provision of outpatient cardiac rehabilitation inthe UK is well below the NSF-CHD goal of 85%of patients with acute myocardial infarction andrevascularisation being offered outpatientcardiac rehabilitation.● Information on referral to and uptake of cardiacrehabilitation is incomplete, with widely varyingestimates of provision, particularly in underrepresentedgroups. Little is known about thecapacity of cardiac rehabilitation centres toincrease provision.● There is an uncoordinated approach to auditdata collection.● Reasons reported by patients for nonattendanceare amenable to intervention, butfew interventions have been formally evaluated.● Many interventions aimed at improving patientuptake, adherence and professional compliancewith guidelines and good practice have beenproposed, but few have been formallyevaluated.● Motivational communications and trained layvolunteers may help to improve uptake ofcardiac rehabilitation.● Self-management techniques may help topromote and sustain lifestyle changes associatedwith cardiac rehabilitation.● Qualitative studies in providers and patientsmay identify attitudes, beliefs and valuesassociated with cardiac rehabilitation.● Information on costs of interventions isfrequently not reported.● Experience of low-cost interventions and goodpractice exists within many cardiacrehabilitation centres.●Increased provision of outpatient cardiacrehabilitation will require additional resources.Recommendations for research anddevelopment●●●●●●●Trials comparing the cost-effectiveness ofcomprehensive multidisciplinary rehabilitationwith simpler outpatient programmes.Economic and patient preference studies of theeffects of different methods of using increasedfunding for cardiac rehabilitation, andevaluations of the impact of any increasedfunding.Evaluation of a range of interventions and goodpractice (including self-managementtechniques, motivational communication andthe use of trained lay volunteers) to promoteattendance in all patients and underrepresentedgroups.Development of standardised audit methods inthe context of modern records systems,appropriate training for dedicated staff anddialogue between service contributors.Standardisation of criteria for patient eligibility,regular and comprehensive data collection toestimate the need for and provision of cardiacrehabilitation.Identification of further areas for interventionthrough qualitative studies.Extension of low-cost interventions and goodpractice in rehabilitation.Regular updated systematic review of literaturerelating to uptake and adherence to cardiacrehabilitation to include literature not readilyavailable to providers and non-UK studies.71© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41AcknowledgementsWe are grateful to Judith Jolliffe, Jo Coast,Karl Karsch, David Thompson and BobLewin for their involvement in the early phase ofthe project. We also thank Hugh Bethell and SallyTurner for allowing us access to the BritishAssociation for Cardiac Rehabilitation/BritishHeart Foundation survey data; Roy Maxwell andDavidson Ho (HES), Farzana Nadeem (PEDW)and Ciara Kennedy (HIS) for supplying data onpopulation need; and the staff of rehabilitationcentres that contributed additional survey data,audits and relevant information.Contributors to survey of audit activityin UK cardiac rehabilitation centresWe are grateful for the assistance of rehabilitationand audit staff from the following centres: KatrinaCreedon (Birmingham Heartlands and SolihullNHS Trust), Claire Brereton-Worsman (BradfordHospitals NHS Trust), Gillian Matthews (CornwallPartnership NHS Trust), Ashley Davidson (EastSomerset NHS Trust), Mark Giles (GloucestershireHospitals NHS Trust), Alison Child (Guy’s and StThomas’ Hospital NHS Trust), Maureen Barry(Homerton Hospital NHS Trust), Judith Imich(Kings College Hospital NHS Trust), Pat Marley(The Lewisham Hospital NHS Trust), Sheila Ryan(Newham Primary Care Trust), Tony Andrews(North Devon Healthcare NHS Trust), ValerieNangle (North Middlesex University HospitalTrust), John Outhwaite and Julie Thompson(Plymouth Hospitals NHS Trust), Victoria Sievey(Royal Bournemouth and Christchurch NHSTrust), Alison Brown (Royal Cornwall NHS Trust),Alison Davey (Royal Free Hampstead NHS Trust),Margaret Wicks (Royal United Hospital Bath NHSTrust), Rosalind Leslie (Royal WolverhamptonNHS Trust), Dee Hannah (St Mary’s NHS Trust),Maggie Kelly (Salisbury Health Care NHS Trust),Margaret Pritchard (Sandwell Healthcare NHSTrust), Suzy Young and Lynne Kilner (SouthDevon Healthcare NHS Trust), Petra Haig (SouthWarwickshire General Hospitals NHS Trust),Amanda Daniel (United Bristol Healthcare Trust),Mark Walsh (Walsall Hospitals NHS Trust), BrianColeman (Whipps Cross University Hospital NHSTrust) and Linda Barratt (Worcester AcuteHospitals NHS Trust).Contributions of authorsAndrew Beswick (Research Associate) contributedto the systematic review, audit and RCT, andprepared the report. Karen Rees (Research Fellow)acted as project coordinator and contributed tothe systematic review and survey of UK provision.Ingolf Griebsch (Research Associate) preparedhealth service costs and population need. FionaTaylor (Public Health Specialist Trainee)contributed to the systematic review and auditsurvey. Margaret Burke (Trials Search Coordinatorfor the Cochrane Heart Group) carried out theliterature searches. Robert West (Reader inEpidemiology) contributed to the systematicreview and RCT attendance, design andplanning. Jackie Victory (Cardiac RehabilitationSister) contributed to the audit survey andthe UK survey of provision. Jacqueline Brown(MRC Senior Scientist) prepared healthservice costs. Rod Taylor (Senior Lecturer inPublic Health and Epidemiology) worked onthe UK survey of provision. Shah Ebrahim(Professor of Epidemiology and Ageing) workedon the systematic review and coordinated theproject.73© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41References1. Feigenbaum MS, Carter E. Cardiac RehabilitationServices. Health Technology Assessment ReportNo. 6. DHHS Publication No. PHS 88-3427.Rockville, MD: US. Department of Health andHuman Services, Public Health Service, NationalCentre for Health Services Research and HealthCare Technology Assessment; 1987.2. Scottish Intercollegiate Guidelines Network(SIGN). Cardiac rehabilitation, a national clinicalguideline. SIGN publication no. 57. Edinburgh:SIGN; 2002.3. Horgan JH, Bethell H, Carson P, Davidson C,Julian D, Mayou RA, et al. Working Party reporton cardiac rehabilitation. Br Heart J 1992;67:412–18.4. Thompson D, Bowman G, De-Bono D, Hopkins A,editors. Cardiac rehabilitation: guidelines and auditstandards. London: Royal College of Physicians ofLondon; 1997.5. Thompson DR, Meddis R. A prospectiveevaluation of in-hospital counselling for first timemyocardial infarction men. J Psychosomatic Res1990;34:237–48.6. Oldridge NB, Guyatt GH, Fischer ME, Rimm AA.Cardiac rehabilitation after myocardial infarction.Combined experience of enrolment in clinicaltrials. JAMA 1988;260:945–50.7. O’Connor GT, Buring JE, Yusuf S, Goldhaber SZ,Olmstead EM, Paffenbarger RS Jr, et al. Anoverview of randomised trials of rehabilitation withexercise after myocardial infarction. Circulation1989;80:234–44.8. Lear SA, Ignaszewski A. Cardiac rehabilitation:a comprehensive review. Current Controlled Trials inCardiovascular Medicine 2001;2:221–32.9. Ades PA. Cardiac rehabilitation and secondaryprevention of coronary heart disease. N Engl J Med2001;345:892–902.10. Jolliffe JA, Rees K, Taylor RS, Thompson D,Oldridge N, Ebrahim S. Exercise-basedrehabilitation for coronary heart disease(Cochrane Review). In The Cochrane Library(Issue 1). Oxford: Update Software; 2002.11. NHS Centre for Reviews and Dissemination.Cardiac Rehabilitation. Eff Health Care 1998;4(4).12. Department of Health. National Service Frameworkfor Coronary Heart Disease. London: Department ofHealth; 2000.13. Lloyd-Williams F, Mair FS, Leitner M. Exercisetraining and heart failure: a systematic review ofcurrent evidence. Br J Gen Pract 2002;52:47–55.14. Wenger NK, Froelicher ES, Kent-Smith L,Ades PA, Berra K, Blumenthal JA, et al. CardiacRehabilitation. Clinical Practice Guideline No. 17.AHCPR Publication No. 96–0672. Rockville, MD:US Department of Health and Human Services,Public Health Service, Agency for Health CarePolicy and Research and the National Heart,Lung, and Blood Institute; October 1995.15. National Heart Foundation of New Zealand 2002.Best practice evidence-based guidelines: cardiacrehabilitation.URL:http://www.nzgg.org.nz/library/gl_complete/Cardiac_Rehab/index.cfm. Accessed January 2003.16. National Heart Forum of Australia 2000.Recommendations for cardiac rehabilitation.URL:http://www.heartfoundation.com.au/prof/index_fr.html. Accessed January 2003.17. Hall R, More R, Camm J, Swanton H, Gray H,Flint J, et al. Fifth report on the provision ofservices for patients with heart disease. Heart2002;88 Suppl iii:1–56.18. National Assembly for Wales. Tackling CHD inWales: implementing through evidence. Cardiff:National Assembly for Wales; 2001.19. Canadian Association for Cardiac Rehabilitation.Canadian guidelines for cardiac rehabilitation andcardiovascular prevention. Winnipeg: CanadianAssociation for Cardiac Rehabilitation; 1999.20. Harlan WR, Sandler SA, Lee KL, Lam LC,Mark DB. Importance of baseline functional andsocioeconomic factors for participation in cardiacrehabilitation. Am J Cardiol 1995;76:36–9.21. Nielsen H. Cardiac rehabilitation after acutemyocardial-infarction in the elderly. Cardiology inthe Elderly 1994;2:290–4.22. Forman DE, Farquhar W. Cardiac rehabilitationand secondary prevention programs for elderlycardiac patients. Clin Geriatr Med 2000;16:619–29.23. O’Callaghan WG, Teo KK, Riordan JO, Webb H,Dolphin T, Horgan JH. Comparative response ofmale and female patients with coronary arterydisease to exercise rehabilitation. Eur Heart J1984;5:649–51.24. Cannistra LB, Balady GJ, O’Malley CJ, Weiner DA,Ryan TJ. Comparison of the clinical profile and75© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

References76outcome of women and men in cardiacrehabilitation. Am J Cardiol 1992;69:1274–9.25. Evans JA, Turner SC, Bethell HJN. Cardiacrehabilitation: are the NSF milestones achievable?Heart 2002;87 Suppl ii:41.26. Bethell HJN, Turner SC, Evans JA, Rose L.Cardiac rehabilitation in the United Kingdom:how complete is the provision? J CardiopulmRehabil 2001;21:111–15.27. Melville MR, Packham C, Brown N, Weston C,Gray D. Cardiac rehabilitation: socially deprivedpatients are less likely to attend but patientsineligible for thrombolysis are less likely to beinvited. Heart 1999;82:373–7.28. Campbell NC, Grimshaw JM, Rawles JM, RitchieLD. Cardiac rehabilitation in Scotland: is currentprovision satisfactory? J Public Health Med1996;18:478–80.29. Pell J, Pell A, Morrison C, Batchford O, Dargie H.Retrospective study of influence of deprivation onuptake of cardiac rehabilitation. BMJ 1996;313:267–8.30. Bunker S, McBurney H, Cox H, Jelinek M.Identifying participation rates at outpatientcardiac rehabilitation programs in Victoria,Australia. J Cardiopulm Rehabil 1999;19:334–8.31. Grenier O, Cambou JP, Ferrieres J, Danchin N.Secondary prevention in women with acutecoronary syndrome in France: results of thePREVENIR survey. Eur Heart J 1999;20 AbstrSuppl:656.32. Parks D, Allison M, Doughty R, Cunningham L,Ellis CJ. An audit of phase II cardiac rehabilitationat Auckland hospital. N Z Med J 2000;113:158–61.33. Thomas RJ, Miller NH, Lamendola C, Berra K,Hedback B, Durstine JL, et al. National Surveyon Gender Differences in Cardiac RehabilitationPrograms. Patient characteristics and enrolmentpatterns. J Cardiopulm Rehabil 1996;16:402–12.34. Jackson SA, Anderson RT, Clark L, Colombo K.Increasing participation in cardiac rehabilitation(the Inspire Study). Circulation 1998;98 Suppl 17:I–707.35. Blackburn GG, Foody JM, Sprecher DL, Park E,Apperson-Hansen C, Pashkow FJ. Cardiacrehabilitation participation patterns in a large,tertiary care center: evidence for selection bias.J Cardiopulm Rehabil 2000;20:189–95.36. Thompson DR, Bowman GS, Kitson AL, de BonoDP, Hopkins A. Cardiac rehabilitation services inEngland and Wales: a national survey. Int J Cardiol1997;59:299–304.37. Bittner V, Sanderson B, Breland J, Green D.Referral patterns to a university-based cardiacrehabilitation program. Am J Cardiol 1999;83:252–5.38. Hotta SS. Cardiac rehabilitation programs.Health Technology Assessment Report 1991,No. 6. Rockville, MD: National Center for HealthServices Research and Health Care TechnologyAssessment; 1991.39. King KM, Humen DP, Teo KK. Cardiacrehabilitation: the forgotten intervention. Can JCardiol 1999;15:979–85.40. Afzal A, Gunda M, Brawner CA, Havstad S,McCullough P, Keteyian SJ. Race and rate ofreferral to cardiac rehabilitation. Circulation1998;98 Suppl 17:I–810–1.41. Mosca L, Han R, Filip J, McGillen C, Rubenfire M.Barriers for physicians to refer to cardiacrehabilitation and impact of a critical care pathwayon rates of participation. Circulation 1998;98 Suppl 17:I–811.42. Bowman GS, Thompson DR, Lewin RJP. Why arepatients with heart failure not routinely offeredcardiac rehabilitation? Coronary Health Care 1998;2:187–92.43. Ades PA, Waldmann ML, McCann WJ, Weaver SO.Predictors of cardiac rehabilitation participation inolder coronary patients. Arch Intern Med 1992;152:1033–5.44. Gulanick M, Bliley A, Perino B, Keough V.Recovery patterns and lifestyle changes aftercoronary angioplasty: the patient’s perspective.Heart Lung 1998;27:253–62.45. Niederhauser HU, Schoch HK. Der alkoholkrankein der rehabilitation [The alcoholic patient inrehabilitation]. Ther Umsch 1990;47:405–11.46. Young RF, Kahana E. Age, medical advice aboutcardiac risk reduction, and patient compliance.J Aging Health 1989;1:121–34.47. Grubb NR, Simpson C, Fox KAA. Memoryfunction in patients with stable, moderate tosevere cardiac failure. Am Heart J (online)2000;140:1–5e.48. Rogers AE, Addington-Hall JM, Abery AJ,McCoy ASM, Bulpitt C, Coats AJS, et al.Knowledge and communication difficulties forpatients with chronic heart failure: qualitativestudy. BMJ 2000;321:605–7.49. Blumenthal JA, Williams RS, Wallace AG,William RB Jr, Needles TL. Physiological andpsychological variables predict compliance toprescribed exercise therapy in patients recoveringfrom myocardial infarction. Psychosom Med 1982;44:519–27.50. Blood SM, Moulton AW, Capone R, Handshaw KE,O’Sullivan PA. Factors associated with the decisionto attend outpatient cardiac rehabilitation

Health Technology Assessment 2004; Vol. 8: No. 41programs after myocardial-infarction. Clin Res1993;41:566A.51. Pell JP, Morrison CE. Factors associated with lowattendance at cardiac rehabilitation. British Journalof Cardiology 1998;5:152–5.52. Schulz DL, McBurney H. Factors which influenceattendance at a rural Australian cardiacrehabilitation program. Coronary Health Care 2000;4:135–41.53. Filip J, McGillen C, Mosca L. Patient preferencesfor cardiac rehabilitation and desired programelements. J Cardiopulm Rehabil 1999;19:339–43.54. McGee HM, Horgan JH. Cardiac rehabilitationprogrammes: are women less likely to attend?BMJ 1992;305:283–4.55. Rawles JM, Trent RJ, Jennings KP. Women andmyocardial infarction: ageism rather than sexism?Br Heart J 1995;73:87–91.56. Jackson D, Daly J, Davidson P, Elliott D, Cameron-Traub E, Wade V, et al. Women recovering fromfirst-time myocardial infarction (MI): a feministqualitative study. J Adv Nurs 2000;32:1403–11.57. Radley A, Grove A, Wright S, Thurston H.Problems of women compared with those of menfollowing myocardial infarction. Coronary HealthCare 1998;2:202–9.58. Taylor FC, Victory JJ, Angelini GD. Use of cardiacrehabilitation among patients following coronaryartery bypass surgery. Heart 2001;86:92–3.59. Lane D, Carroll D, Ring C, Beevers G, Lip GYH.Predictors of attendance at cardiac rehabilitationafter myocardial infarction. J Psychosom Res2001;51:497–501.60. Tod AM, Wadsworth E, Asif S, Gerrish K. Cardiacrehabilitation: the needs of South Asian cardiacpatients. Br J Nurs 2001;10:1028–33.61. Bethell HJN. How to do it: set up a coronaryrehabilitation programme. BMJ 1988;297:120–1.62. Ockene IS, Ockene JK. Barriers to lifestyle change,and the need to develop an integrated approachto prevention. Cardiol Clin 1996;14:159–69.63. Oldridge NB. Cardiac rehabilitation exerciseprogramme. Compliance and complianceenhancingstrategies. Sports Med 1988;6:42–55.64. Daly J, Sindone AP, Thompson DR, Hancock K,Chang E, Davidson P. Barriers to participation inand adherence to cardiac rehabilitation programs:a critical literature review. Prog Cardiovasc Nurs2002;17:8–17.65. Aikman H, McBurney H, Bunker S. Cardiacrehabilitation: the extent, reasons and predictorsof patient non-attendance. In Proceedings of VIWorld Congress of Cardiac Rehabilitation. BuenosAires, Argentina; 1996.66. Evenson KR, Fleury J. Barriers to outpatientcardiac rehabilitation participation and adherence.J Cardiopulm Rehabil 2000;20:241–6.67. Dignan MB, Young LD, Crouse JR, King JM.Factors associated with participation in apreventive cardiology service by patients withcoronary heart disease. Southern Medical AssociationJournal 1995;88:1057–61.68. Wiles, R. Patients’ perceptions of their heart attackand recovery: the influence of epidemiological‘evidence’ and personal experience. Soc Sci Med1998;46:1477–86.69. Thow M, Isoud P, White M, Robertson I, Keith E,Armstrong G. Uptake and adherence of womenpost myocardial infarction to phase III cardiacrehabilitation: are things changing? CoronaryHealth Care 2000;4:174–8.70. Nohara R, Kambara H, Mohiuddin IH, Ono S,Okuda K, Makita S, et al. Cardiac sportsrehabilitation for patients with ischemic heartdisease. Japanese Circulation Journal 1990;54:1443–50.71. Beach EK, Smith A, Luthringer L, Utz SK, AhrensS, Whitmire V. Self-care limitations of persons afteracute myocardial infarction. App Nurs Res 1996;9:24–8.72. Seddon ME, Ayanian JZ, Landrum MB, Cleary PD,Peterson EA, Gahart MT, et al. Quality ofambulatory care after myocardial infarction amongMedicare patients by type of insurance and region.Am J Med 2001;111:24–32.73. Carlson JJ, Johnson JA, Franklin BA,VanderLaan RL. Program participation, exerciseadherence, cardiovascular outcomes, and programcost of traditional versus modified cardiacrehabilitation. Am J Cardiol 2000;86:17–23.74. Young R, Kahana E, Rubenfire M. Preventivehealth behavior advice. A study with oldermyocardial infarction patients. Eval Health Prof1987;10:394–407.75. Missik E. Women and cardiac rehabilitation:accessibility issues and policy recommendations.Rehabil Nurs 2001;26:141–7.76. Thornhill M, Stevens JA. Client perceptions of arural-based cardiac rehabilitation program: agrounded theory approach. Aust J Rural Health1998;6:105–11.77. Walentiny C, Siebold C, Wassink S, Emery R,Andreassen R, Nelson R, et al. Determinants andpotential educational benefits of participation inoutpatient cardiac rehabilitation programs postcoronary-artery bypass graft-surgery. Clin Res1988;36:A327.77© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

References7878. Kubler, W, Niebauer, J, Kreuzer, J. Kosten/nutzenrelation:Bewertung der stationaren undRehabilitation [Cost-effectiveness: inpatient versusoutpatient rehabilitation]. Z Kardiol 1994;83(Suppl 6):151–8.79. Turner S. Basingstoke and Alton CardiacRehabilitation Unit: annual report 1993. Alton,Hants: Alton Health Centre; 1993.80. Gray AM, Bowman GS, Thompson DR. The costof cardiac rehabilitation services in England andWales. J R Coll Physicians Lond 1997;31:57–61.81. Taylor R, Kirby B. Cost implications of cardiacrehabilitation in older patients. Coron Artery Dis1999;10:53–6.82. Osika JS. Cardiac rehabilitation: evaluation of apilot lay volunteer home visiting service and needsassessment. Part 2 Membership Exam Thesis.London: FPHM; 2001.83. Davidson C, Reval K, Chamberlain DA,Pentecost B, Parker J. A report of a working groupof the British Cardiac Society: cardiacrehabilitation services in the United Kingdom1992. Br Heart J 1995;73:201–2.84. Office for National Statistics. Key population andvital statistics 1999. Local and health authority areas.Series VS/PP. London: HMSO; 2001.85. McKee M, James P. Using routine data to evaluatequality of care in British hospitals. Med Care1997;35(Suppl 10):OS102–11.86. Bethell HJN, Turner SC, Flint EJ, Rose L. TheBACR Database of Cardiac Rehabilitation Units inthe UK. Coronary Health Care 2000;4:92–5.87. Coats AJS, McGee HM, Stokes HC,Thompson DR, editors. BACR guidelines for cardiacrehabilitation. Oxford: Blackwell Science; 1995.88. Lewin RJ, Ingleton R, Newens AJ, Thompson DR.Adherence to cardiac rehabilitation guidelines: asurvey of rehabilitation programmes in the UnitedKingdom. BMJ 1998;316:1354–5.89. Hine D. Clinical governance and the Commissionfor Health Improvement. In West RR, Norris R,editors. Measurement of clinical performance. Practicalapproaches in acute myocardial infarction. London:Clinical Effectiveness and Evaluation Unit, RoyalCollege of Physicians; 2001. pp. 7–15.90. Scally G, Donaldson LJ. Clinical governance andthe drive for quality improvement in the new NHSin England. BMJ 1998;317:61–5.91. Johnston G, Crombie IK, Davies HT, Alder EM,Millard A. Reviewing audit: barriers andfacilitating factors for effective clinical audit.Quality in Health Care 2000;9:23–36.92. Lewin B, Roebuck A, Thompson D. Coronaryheart disease national service framework: cardiacrehabilitation – meeting the information needs.Northern & Yorkshire Public Health ObservatoryOccasional Paper No. 4; April 2002. URL:http://www.nepho.org.uk/index.php?i=11&rs=1#rpt_7. Accessed 23 October 2003.93. Auditing cardiac rehabilitation: pilot study to beset up. British Journal of Cardiology 2003;10:256.94. Georgiou A, Pearson M. The role of healthinformatics in clinical audit: part of the problemor key to the solution? J Eval Clin Pract 2002;8:183–8.95. West RR. 2001. Rehabilitation after acutemyocardial infarction: multicentre randomisedcontrolled trial. URL: http://www.doh.gov.uk/research/rd3/nhsrandd/timeltdprogs/cvd/rm19.htm. Accessed 18 December 2003.96. Stokes H, Turner S, Farr A. Cardiac rehabilitation:programme, structure, content, management andadministration. In Coats AJS, McGee HM, StokesHC, Thompson DR, editors. BACR guidelines forcardiac rehabilitation. Oxford: Blackwell Science;1995. pp. 12–39.97. Song F, Eastwood A, Gilbody S, Duley L, Sutton A.Publication and related biases. Health Technol Assess2000;4(10):1–115.98. Hyde C, Parkes J, Deeks J, Milne R. Systematicreview of studies of effectiveness of teachingcritical appraisal. UK National R&D Programme:evaluating methods to promote theimplementation of R&D. URL:http://www.bham.ac.uk/arif/SysRevs/TeachCritApp.PDF. Accessed January 2003.99. Kunz R, Oxman AD. The unpredictabilityparadox: review of empirical comparisons ofrandomised and non-randomised clinical trials.BMJ 1998;317:1185–90.100. Clarke M, Oxman AD, editors. CochraneReviewers Handbook 4.1.4 (updated October2001). In The Cochrane Library (Issue 4). Oxford:Update Software; 2001 (updated quarterly).101. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D,Stroup DF. Improving the quality of reports ofmeta-analyses of randomised controlled trials: theQUOROM statement. Lancet 1999;354:1896–900.102. Bradley F, Morgan S, Smith H, Mant D. Preventivecare for patients following myocardial infarction.Fam Pract 1997;14:220–6.103. Jolly K, Bradley F, Sharp S, Smith H, Mant D.Follow-up care in general practice of patients withmyocardial infarction or angina pectoris: initialresults of the SHIP trial. Fam Pract 1998;15:548–55.104. Jolly K, Bradley F, Sharp S, Smith H, Thompson S,Kinmonth AL, et al. Randomised controlled trial offollow up care in general practice of patients withmyocardial infarction and angina: final results ofthe Southampton heart integrated care project(SHIP). BMJ 1999;318:706–11.

Health Technology Assessment 2004; Vol. 8: No. 41105. Wyer SJ. Increasing attendance rates at a cardiacrehabilitation programme using the Theory ofPlanned Behaviour: an intervention study. PhDthesis. University of Warwick; 2001.106. Wyer SJ, Earll L, Joseph S, Harrison J, Giles M,Johnston M. Increasing attendance at a cardiacrehabilitation programme: an intervention studyusing the Theory of Planned Behaviour. CoronaryHealth Care 2001;5:154–9.107. Ajzen I, Madden TJ. Prediction of goal-directedbehaviour: attitudes, intentions, and perceivedbehavioural control. Journal of Experimental SocialPsychology 1986;22;453–74.108. Hillebrand T, Frodermann H, Lehr D, Wirth A.Vermehrte teilnahme an ambulaten herzgruppendurch poststationare nachsorge [Increasedparticipation in coronary groups by means of anoutpatient care program]. Herz Kreislauf1995;27:346–9.109. Krasemann EO, Busch T. Was kann eineaufklarungsbroschure bewirken? [What can aninformative pamphlet really achieve?]. OffentlGesundheitswes 1988;50:96–8.110. Imich J. Home-based cardiac rehabilitation.Nursing Times 1997;93(50):48–9.111. Scott IA, Eyeson-Annan ML, Huxley SL, West MJ.Optimising care of acute myocardial infarction:results of a regional quality improvement project. JQual Clin Prac 2000;20:12–19.112. Pasquali SK, Alexander KP, Lytle BL, Coombs LP,Peterson ED. Testing an intervention to increasecardiac rehabilitation enrollment after coronaryartery bypass grafting. Am J Cardiol 2001;88:1415–16.113. Keck M, Budde HG, Hamerle A. Medizinischeund sozialmedizinische einflussgrossen auf dasactive nachsorgeverhalten von ahb-patienten[Medical and sociomedical parameters of theparticipation in coronary training groups].Herz Kreislauf 1991;23:163–7.114. Keck M, Budde HG. Praxis der motivation zurambulanten herzgruppe [Practice of motivation forout-patient ‘heart group’]. Herz Kreislauf 1996;28:311–15.115. Millar B. Supporters’ clubs. Health Ser J 1993;103:39.116. Levknecht L, Schriefer J, Schriefer J, Maconis B.Combining case management, pathways, andreport cards for secondary cardiac prevention.Joint Commission Journal on Quality Improvement1997;23:162–74.117. Tod AM, Pearson K, McCabe M. Cardiacrehabilitation: integrating primary and secondarycare. Coronary Health Care 1998;2:150–5.118. Johnson K. Use of telephone follow-up for postcardiacsurgery patients. Intensive and Critical CareNursing 2000;16:144–50.119 McCarney R, Shapley M, Goodwin R, Croft P. Theintroduction of a routine monitoring system inprimary care for patients with a first episode ofcardiovascular disease. J Adv Nurs 2000;31:1376–82.120. Foresman JE. Phase I referrals to outpatientcardiac rehab programs: are we doing enough?J Cardiopulm Rehabil 1997;17:341.121. Mehta RH, Das S, Tsai TT, Nolan E, Kearly G,Eagle KA. Quality improvement initiative and itsimpact on the management of patients with acutemyocardial infarction. Arch Intern Med2000;160:3057–62.122. Kalayi C, Rimmer F, Maxwell M. Improvingreferral for cardiac rehabilitation – an interfaceaudit. Journal of Clinical Governance 1999;7:143–5.123. Caulin-Glaser T, Schmeizel R. Impact ofeducational initiatives on gender referrals tocardiac rehabilitation. Journal of CardiacRehabilitation 2000;20:302.124. Suskin N, Wisenberg G, Barnett P, Pellizzari J,Murray-Parsons N. Physician (MD) endorsementimproves cardiac rehabilitation (CR) participation.Journal of Cardiac Rehabilitation 2000;20:306.125. Campbell NC, Thain J, Deans HG, Ritchie LD,Rawles JM, Squair JL. Secondary preventionclinics for coronary heart disease: randomised trialof effect on health. BMJ 1998;316:1434–7.126. Campbell NC, Ritchie LD, Thain J, Deans HG,Rawles JM, Squair JL. Secondary prevention incoronary heart disease: a behavioural trial of nurseled clinics in primary care. Heart 1998;80:447–52.127. Feder G, Griffiths C, Eldridge S, Spence M.Effect of postal prompts to patients and generalpractitioners on the quality of primary care after acoronary event (POST): randomised controlledtrial. BMJ 1999;318:1522–6.128. Cannistra LB, O’Malley CJ, Balady GJ.Comparison of outcome of cardiac rehabilitationin black women and white women. Am J Cardiol1995;75:890–3.129. Tack BB, Gilliss CL. Nurse-monitored cardiacrecovery: a description of the first 8 weeks. HeartLung 1990;19:491–9.130. Hershberger PJ, Robertson KB, Markert RJ.Personality and appointment-keeping adherencein cardiac rehabilitation. J Cardiopulm Rehabil1999;19:106–11.131. DeBusk RF, Haskell WL, Miller NH, Berra K,Taylor CB, Berger WE, et al. Medically directed athomerehabilitation soon after clinicallyuncomplicated acute myocardial infarction: a newmodel for patient care. Am J Cardiol 1985;55:251–7.132. Lewin B, Robertson IH, Cay EL, Irving JB,Campbell M. Effects of self-help post-myocardialinfarctionrehabilitation on psychologicaladjustment and use of health services. Lancet1992;339:1036–40.79© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

References80133. Shaw DK. Telemedicine and cardiopulmonaryrehabilitation: where do we stand? J CardiopulmRehabil 1999;19:59–61.134. Ades PA, Pashkow FJ, Fletcher G, Pina IL,Zohman LR, Nestor JR. A controlled trial ofcardiac rehabilitation in the home setting usingelectrocardiographic and voice transtelephonicmonitoring. Am Heart J 2000;139:543–8.135. Roitman JL, LaFontaine T, Drimmer AM. A newmodel for risk stratification and delivery ofcardiovascular rehabilitation services in the longtermclinical management of patients with coronaryartery disease. J Cardiopulm Rehabil 1998;18:113–23.136. Bethell HJN, Mullee MA. A controlled trial ofcommunity based coronary rehabilitation. Br HeartJ 1990;64:370–5.137. Contractor A, English R, Salmon R, Leighton R,Willoughby M, Gordon N. Clinical effectiveness ofa community-based cardiovascular risk reductionprogram (Intervent USA) in lower risk patientswith coronary artery disease. J Cardiopulm Rehabil2000;20:294.138. Moore SM. Women’s views of cardiac rehabilitationprograms. J Cardiopulm Rehabil 1996;16:123–9.139. Brezinka V, Dusseldorp E, Maes S. Genderdifferences in psychosocial profile at entry intocardiac rehabilitation. J Cardiopulm Rehabil1998;18:445–9.140. Toobert DJ, Glasgow RE, Nettekoven LA,Brown JE. Behavioral and psychosocial effects ofintensive lifestyle management for women withcoronary heart disease. Patient Educ Couns1998;35:177–88.141. Egger M, Dickersin K, Davey Smith G. Problemsand limitations in conducting systematic reviews.In Egger M, Davey Smith G, Altman DG, editors.Systematic reviews in health care: meta-analysis incontext. London: BMJ Publishing Group; 2001.pp. 43–68.142. Rogutski S, Berra K, Haskell W. Home-basedcardiac rehabilitation: variations on a theme. InWenger NK, Smith LK, Froelicher ES, Comoss PM,editors. Cardiac rehabilitation: a guide to practice inthe 21st century. New York: Marcel Dekker; 1999.pp. 343–59.143. Miller P, Wikoff R, McMahon M, Garrett MJ,Ringel K. Influence of a nursing intervention onregimen adherence and societal adjustmentspostmyocardial infarction. Nurs Res 1988;37:297–302.144. Miller P, Wikoff R, McMahon M, Garrett MJ,Ringel K, Collura D, et al. Personal adjustmentsand regimen compliance 1 year after myocardialinfarction. Heart Lung 1989;18:339–46.145. Oldridge NB, Jones NL. Improving patientcompliance in cardiac exercise rehabilitation:effects of written agreement and self-monitoring.Journal of Cardiac Rehabilitation 1983;3:257–62.146. Daltroy LH. Improving cardiac patient adherenceto exercise regimens: a clinical trial of healtheducation. Journal of Cardiac Rehabilitation 1985;5:40–9.147. Mahler HIM, Kulik JA, Tarazi RY. Effects of avideotape information intervention at discharge ondiet and exercise compliance after coronary bypasssurgery. J Cardiopulm Rehabil 1999;19:170–7.148. Aish AE, Isenberg M. Effects of Orem-basednursing intervention on nutritional self-care ofmyocardial infarction patients. Int J Nur Stud1996;33:259–70.149. Ashe ED. The effects of a relapse preventionprogram on adherence to a phase II cardiacexercise program. PhD thesis. Florida: FloridaState University; 1993.150. Hopper DL. The effects of telephone contact onthe exercise adherence of cardiac patients. PhDthesis. Denver: University of Denver, 1995.151. Duncan K, Pozehl B, Rosado K. The effects ofbehavioral feedback on adherence to dietarysodium intake for patients with congestive heartfailure. J Cardiopulm Rehabil 2001;21:304.152. Leslie M, Schuster PA. The effect of contingencycontracting on adherence and knowledge ofexercise regimens. Patient Educ Couns1991;18:231–41.153. Lack ER. Group psychotherapy and the cardiacrehabilitation process: a study in improvingpatient compliance. PhD thesis. Philadelphia:Temple University; 1985.154. Marshall J, Penckofer S, Llewellyn J. Structuredpostoperative teaching and knowledge andcompliance of patients who had coronary arterybypass surgery. Heart Lung 1986;15:76–82.155. Huerin M, Rosario A, Bergman G, Belardi J,Trivi M, Guzman L, Rubinstein F. Compliancewith cardiovascular rehabilitation: comparison ofstandard and a new adherence strategies. InProceedings of the XIIIth World Congress of Cardiology,26–30 April 1998, Rio de Janeiro, Brazil. Abstract2419.156. McKenna M, Dewar P, McCreary L, Fyfe T,Murdoch DL. Women and the lower intensitywithin cardiac rehabilitation: the forgotten groups.British Association for Cardiac Rehabilitation Newsletter1998;1(4):14.157. Erling J, Oldridge NB. Effect of a spousal-supportprogram on compliance with cardiacrehabilitation. Med Sci Sports Exerc 1985;17:284.158. Sparks KE, Shaw DK, Eddy D, Hanigosky P,Vantrese J. Alternatives for cardiac rehabilitationpatients unable to return to a hospital-basedprogram. Heart Lung 1993;22:298–303.

Health Technology Assessment 2004; Vol. 8: No. 41159. Blumenthal JA, Rejeski WJ, Walsh-Riddle M,Emery CF, Miller H, Roark S, et al. Comparison ofhigh- and low-intensity exercise training early afteracute myocardial infarction. Am J Cardiol 1988;61:26–30.160. Brubaker PH, Warner JG Jr. Rejeski WJ,Edwards DG, Matrazzo BA, Ribisl PM, et al.Comparison of standard- and extended-lengthparticipation in cardiac rehabilitation on bodycomposition, functional capacity, and blood lipids.Am J Cardiol 1996;78:769–73.161. Lee JY, Jensen BE, Oberman A, Fletcher GF,Fletcher BJ, Raczynski JM. Adherence in thetraining levels comparison trial. Med Sci SportsExerc 1996;28:47–52.162. Dracup K, Meleis AI, Clark S, Clyburn A,Shields L, Staley M. Group counseling in cardiacrehabilitation: effect on patient compliance.Patient Educ Couns 1984;6:169–77.163. Penckofer S, Llewellyn J. Adherence to risk-factorinstructions one year following coronary arterybypass surgery. J Cardiovasc Nurs 1989;3:10–24.164. Linde BJ, Janz NM. Effect of a teaching programon knowledge and compliance of cardiac patients.Nurs Res 1979;28:282–6.165. Unden AL, Schenck-Gustafsson K, Axelsson PO,Karlsson I, Orth-Gomer K, Ydrefors AM. Positiveeffects of increased nurse support for malepatients after acute myocardial infarction. Qual LifeRes 1993;2:121–7.166. van Elderen T, Maes S, Seegers G. Effects of apost-hospitalization group health educationprogramme for patients with coronary heartdisease. Psychol Health 1994;9:317–30.167. Barnason S, Zimmerman L. A comparison ofpatient teaching outcomes among postoperativecoronary artery bypass graft (CABG) patients.Prog Cardiovasc Nurs 1995;10:11–20.168. Gordon NF, Haskell WL. Comprehensivecardiovascular disease risk reduction in a cardiacrehabilitation setting. Am J Cardiol 1997;80(8B):H69–73.169. Labrador MP, Merz CNB, Pass R. A randomisedtrial of risk factor case management in coronaryartery disease patients following cardiacrehabilitation. Circulation 1998;98 Suppl 17:I–811.170. Vale MJ, Jelinek MV, Santamarie JD, Best JD.Do patients with coronary heart disease achievebetter cholesterol results with a coach?Randomised controlled trial. Eur Heart J 2000;21 Abstr Suppl:220.171. Senaratne MP, Griffiths J, Mooney D, Kasza L,Macdonald K, Hare S. Effectiveness of a plannedstrategy using cardiac rehabilitation nurses for themanagement of dyslipidemia in patients withcoronary artery disease. Am Heart J 2001;142:975–81.172. Skof E, Span M, Keber I. Secondary prevention inpatients several years after myocardial infarction:comparison of an outpatient and an inpatientrehabilitation programme. J Cardiovasc Risk2001;8:119–26.173. Baile WF, Engel BT. A behavioural strategy forpromoting treatment compliance followingmyocardial infarction. Psychosom Med 1978;40:413–19.174. Stern MJ, Cleary P. National Exercise and HeartDisease Project. Psychosocial changes observedduring a low-level exercise program. Arch InternMed 1981;141:1463–7.175. Starkey C, Michaelis J, de Lusignan S.Computerised systematic secondary prevention inischeamic heart disease: a study in one practice.Public Health 2000;114:169–75.176. Orem D. Nursing concepts of practice. 4th ed.St. Louis, MO: Mosby; 1991.177. Marlatt GA, Gordon JR, editors. Relapse prevention:maintenance strategies in the treatment of addictivebehaviors. New York: Guilford Press; 1985.178. Tooth L, McKenna K. Contemporary issues incardiac rehabilitation: implications foroccupational therapists. British Journal ofOccupational Therapy 1996;59:133–40.179. Koikkalainen M, Lappalainen R, Mykkanen H.Why cardiac patients do not follow thenutritionist’s advice: barriers in nutritional adviceperceived in rehabilitation. Disabil Rehabil1996;18:619–23.180. Knapp D, Blackwell B. Emotional and behavioralproblems in cardiac rehabilitation patients. Journalof Cardiac Rehabilitation 1985;5:112–23.181. Edgren L. Co-production – an approach to cardiacrehabilitation from a service managementperspective. J Nurs Manage 1998;6:77–85.182. Oldridge NB. Compliance and dropout in cardiacexercise rehabilitation. Journal of CardiacRehabilitation 1984;4:166–77.183. DeBusk RF, Miller NH, Superko HR, Dennis CA,Thomas RJ, Lew HT, et al. A case-managementsystem for coronary risk factor modification afteracute myocardial infarction. Ann Intern Med1994;120:721–9.184. Hoepfel-Harris JA. Rehabilitation of the cardiacpatient. Improving compliance with an exerciseprogram. Am J Nurs 1980;80:449–50.185. Comoss PM. Nursing strategies to improvecompliance with life-style changes in a cardiacrehabilitation population. J Cardiovasc Nurs1988;2:23–36.186. Emery CF. Adherence in cardiac and pulmonaryrehabilitation. J Cardiopulm Rehabil 1995;15:420–3.81© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

References187. Moore SM, Kramer FM. Women’s and men’spreferences for cardiac rehabilitation programfeatures. J Cardiopulm Rehabil 1996;16:163–8.188. Allen JK, Redman BK. Cardiac rehabilitation inthe elderly: improving effectiveness. Rehabil Nurs1996;21:182–6.189. Eftekhari H, Marson P, Mattu R, Trevelyan J,Needham T. Improving a cardiac rehabilitationprogramme for Asian patients in Coventry.Coronary Health Care 2000;4:210–1.190. Dickerson SS. Cardiac spouses’ help-seekingexperiences. Clin Nurs Res 1998;7:6–24.191. Axtell SS, Ludwig E, Lopez-Candales A.Intervention to improve adherence to ACC/AHArecommended adjunctive medications for themanagement of patients with an acute myocardialinfarction. Clin Cardiol 2001;24:114–8.192. Hillert B, Remonte S, Rodgers G, Yancy CW,Kaul AF. Improving patient outcomes by poolingresources (the Texas Heart Care Partnershipexperience). Am J Cardiol 2000;85(3A):43–51A.193. Stokes HC. Education and training towardscompetency for cardiac rehabilitation nurses in theUnited Kingdom. J Clin Nurs 2000;9:411–9.194. King KM, Teo KK. Cardiac rehabilitation referraland attendance: not one and the same. RehabilNurs 1998;23:246–51.195. Effron MB, Forbes WJ, Weber AR, Mobley NH.Proximity to cardiac event improves compliancewith a phase-II cardiac rehabilitation program in afreestanding rehabilitation center. Arch Phys MedRehabil 1986;67:665.196. Netten A, Rees T, Harrison G. Unit costs of healthand social care 2001. The University of Kent,Canterbury: Personal Social Services Research Unit(PSSRU); 2001.197. http://www.nhscareers.nhs.uk, March 2002.Accessed November 2002.198. Lambert CM, Hurst NP, Lochhead A, McGregor K,Hunter M, Forbes J. A pilot study of the economiccost and clinical outcome of day patient vs.inpatient management of active rheumatoidarthritis. Br J Rheumatol 1994;33:383–8.199. Bricker L, Garcia J, Henderson J, Mugford M,Neilson J, Roberts T, et al. Ultrasound screening inpregnancy: a systematic review of the clinicaleffectiveness, cost-effectiveness and women’s views.Health Technol Assess 2000;4(16):1–193.200. Davies A, Buxton MJ, Patterson DL,Webster-King J. Anti-coagulant monitoring servicedelivery: a comparison of costs of hospital andcommunity outreach clinics. Clin Lab Haematol2000;22:33–40.201. Jones LW, Farrell JM, Jamieson J, Dorsch KD.Factors influencing enrollment in a cardiacrehabilitation exercise program. Can J CardiovascNurs 2003;13:11–15.82

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 1Need for cardiac rehabilitation in the UK83© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 1TABLE 35 Need for cardiac rehabilitation in EnglandGender Age Population Counts Rates per 100,000group (in 1000s) AMI HF UA CABG PTCA CABGp Others AMI HF UA CABG PTCA CABGp OthersFemale 35–44 3,368.4 271 15 442 46 113 23 1,232 8.0 0.4 13.1 1.4 3.4 0.7 36.645–54 3,451.0 1,154 86 2,072 311 523 72 4,915 33.4 2.5 60.0 9.0 15.2 2.1 142.455–64 2,398.5 2,585 261 4,005 886 1,165 233 9,803 107.8 10.9 167.0 36.9 48.6 9.7 408.765–74 2,253.2 4,586 808 6,148 1,399 1,316 410 13,777 203.5 35.9 272.9 62.1 58.4 18.2 611.575–84 1,601.2 4,767 1,232 5,892 488 435 172 11,105 297.7 76.9 368.0 30.5 27.2 10.7 693.585+ 737.5 1,973 732 1,983 12 18 2 4,455 267.5 99.3 268.9 1.6 2.4 0.3 604.1Male 35–44 3,450.0 1,768 88 1,587 359 667 91 3,923 51.2 2.6 46.0 10.4 19.3 2.6 113.745–54 3,446.3 5,543 470 4,977 1,951 2,531 420 14,122 160.8 13.6 144.4 56.6 73.4 12.2 409.855–64 2,361.5 7,977 1,138 8,325 4,638 3,818 968 24,906 337.8 48.2 352.5 196.4 161.7 41.0 1,054.765–74 1,949.9 8,568 1,752 9,422 4,773 2,625 1,083 24,755 439.4 89.9 483.2 244.8 134.6 55.5 1,269.675–84 1,017.8 5,379 1,399 5,711 1,058 638 371 12,074 528.5 137.5 561.1 104.0 62.7 36.5 1,186.385+ 265.5 1,156 356 1,074 19 23 4 2,247 435.3 134.1 404.4 7.2 8.7 1.5 846.2Both 35–44 6,818.4 2,040 103 2,031 405 780 114 5,157 29.9 1.5 29.8 5.9 11.4 1.7 75.645–54 6,897.2 6,698 557 7,052 2,262 3,054 492 19,043 97.1 8.1 102.2 32.8 44.3 7.1 276.155–64 4,760.0 10,570 1,399 12,336 5,524 4,983 1,201 34,731 222.1 29.4 259.2 116.1 104.7 25.2 729.665–74 4,203.0 13,162 2,562 15,577 6,173 3,941 1,493 38,561 313.2 61.0 370.6 146.9 93.8 35.5 917.575–84 2,619.0 10,156 2,634 11,607 1,546 1,073 543 23,195 387.8 100.6 443.2 59.0 41.0 20.7 885.685+ 1,003.1 3,134 1,089 3,064 31 41 6 6,713 312.4 108.6 305.5 3.1 4.1 0.6 669.3Total 26,300.7 45,760 8,344 51,667 15,941 13,872 3,849 127,400 174.0 31.7 196.4 60.6 52.7 14.6 484.4Source: HES, England. AMI, acute myocardial infarction; HF, heart failure; CABG, coronary artery bypass graft; PTCA, percutaneous transluminal coronaryangioplasty; UA, unstable angina; CABGp, CABG with AMI or HF or PTCA or UA within one admission episode; All others: all patients who have beenadmitted more than once and with more than one discharge diagnosis.84

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 36 Need for cardiac rehabilitation in WalesGender Age Population Counts Rates per 100,000group (in 1000s) AMI HF UA CABG PTCA CABGp Others AMI HF UA CABG PTCA CABGp OthersFemale 35–44 201.9 22 0 37 1 3 3 86 10.9 0.0 18.3 0.5 1.5 1.5 42.645–54 197.8 66 9 114 12 21 5 310 33.4 4.6 57.6 6.1 10.6 2.5 156.755–64 160.8 205 32 258 48 42 15 662 127.5 19.9 160.4 29.9 26.1 9.3 411.765–74 144.2 374 61 351 67 38 25 998 259.4 42.3 243.4 46.5 26.4 17.3 692.175–84 108.5 347 112 350 19 8 15 962 319.8 103.2 322.6 17.5 7.4 13.8 886.685+ 44.2 152 76 97 0 0 0 406 343.9 171.9 219.5 0.0 0.0 0.0 918.6Male 35–44 203.2 99 6 61 15 17 7 208 48.7 3.0 30.0 7.4 8.4 3.4 102.445–54 196.5 408 34 311 85 98 31 916 207.6 17.3 158.3 43.3 49.9 15.8 466.255–64 156.6 610 74 507 205 146 72 1,657 389.5 47.3 323.8 130.9 93.2 46.0 1,058.165–74 125.2 603 162 570 183 93 63 1,614 481.6 129.4 455.3 146.2 74.3 50.3 1,289.175–84 69.3 420 121 329 48 14 22 938 606.1 174.6 474.7 69.3 20.2 31.7 1,353.585+ 15.9 91 22 65 0 1 0 211 572.3 138.4 408.8 0.0 6.3 0.0 1,327.0Both 35–44 405.1 122 6 98 16 20 10 294 30.1 1.5 24.2 3.9 4.9 2.5 72.645–54 394.2 476 43 426 97 119 36 1,226 120.8 10.9 108.1 24.6 30.2 9.1 311.055–64 317.4 816 106 765 253 188 87 2,320 257.1 33.4 241.0 79.7 59.2 27.4 730.965–74 269.4 978 223 922 250 131 88 2,612 363.0 82.8 342.2 92.8 48.6 32.7 969.675–84 177.9 768 233 680 67 22 37 1,903 431.7 131.0 382.2 37.7 12.4 20.8 1,069.785+ 60.1 243 98 162 0 1 0 618 404.3 163.1 269.6 0.0 1.7 0.0 1,028.3Total 1,624.1 3,403 709 3,053 683 481 258 8,973 209.5 43.7 188.0 42.1 29.6 15.9 552.5Source: PEDW.85© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 186TABLE 37 Need for cardiac rehabilitation in Northern IrelandGender Age Population Counts Rates per 100,000group (in 1000s) AMI HF UA CABG PTCA CABGp Others AMI HF UA CABG PTCA CABGp OthersFemale 35–44 120.1 26 3 19 2 7 0 85 21.6 2.5 15.8 1.7 5.8 0.0 70.845–54 98.7 82 11 102 11 47 0 258 83.1 11.1 103.3 11.1 47.6 0.0 261.455–64 80.3 181 36 189 31 95 0 550 225.4 44.8 235.4 38.6 118.3 0.0 684.965–74 67.6 271 105 276 38 79 0 693 400.9 155.3 408.3 56.2 116.9 0.0 1,025.175–84 46.3 277 144 236 5 24 0 477 598.3 311.0 509.7 10.8 51.8 0.0 1,030.285+ 17.5 94 54 55 0 0 0 101 537.1 308.6 314.3 0.0 0.0 0.0 577.1Male 35–44 115.9 108 6 74 16 63 0 252 93.2 5.2 63.8 13.8 54.4 0.0 217.445–54 96.7 331 35 270 74 247 1 819 342.3 36.2 279.2 76.5 255.4 1.0 846.955–64 75.6 468 117 514 183 358 3 1,401 619.0 154.8 679.9 242.1 473.5 4.0 1,853.265–74 54.1 478 171 494 126 227 1 1,214 883.5 316.1 913.1 232.9 419.6 1.8 2,244.075–84 28.4 249 138 215 25 31 0 436 876.8 485.9 757.0 88.0 109.2 0.0 1,535.285+ 6.1 40 43 37 0 0 0 59 655.7 704.9 606.6 0.0 0.0 0.0 967.2Both 35–44 236 134 9 93 18 70 0 337 56.8 3.8 39.4 7.6 29.7 0.0 142.845–54 195.3 413 46 372 85 294 1 1,077 211.5 23.6 190.5 43.5 150.5 0.5 551.555–64 155.9 649 153 703 214 453 3 1,951 416.3 98.1 450.9 137.3 290.6 1.9 1,251.465–74 121.7 749 276 770 164 306 1 1,907 615.4 226.8 632.7 134.8 251.4 0.8 1,567.075–84 74.8 526 282 451 30 55 0 913 703.2 377.0 602.9 40.1 73.5 0.0 1,220.685+ 23.6 134 97 92 0 0 0 160 567.8 411.0 389.8 0.0 0.0 0.0 678.0Total 807.3 2,605 863 2,481 511 1,178 5 6,345 322.7 106.9 307.3 63.3 145.9 0.6 786.0Source: HIS, Northern Ireland.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 2Need for and estimated level of cardiacrehabilitation provision in the UK87© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 288TABLE 38 Need and estimated level of provision in England (total number of cardiac rehabilitation centres n = 220)Mean SD Median IQR n Sum of Non-responding Total sum of patientsall patients centres (n) (responding plus nonrespondingcentres)Mean Median IQRNo. of patients referred to CR 349.0 244.33 279.0 180–450 117 40,840 103 76,793 69,577 59,380–87,190No. of patients joining CR 208.6 141.8 177.5 112–277 114 23,781 106 45,893 42,596 35,653–53,143Eligible patients (source: HES) No. of Proportion of patients referred to Proportion of patients joiningpatients CR (%) Data for non-responders CR (%) Data for non-respondersimputed imputedMean Median IQR Mean Median IQRAll patients with primary diagnosis of IHD 266,833 29 26 22–33 17 16 13–20All patients discharged with AMI, CABG, PTCA, UA 131,089 59 53 45–67 35 33 27–41All patients with primary diagnosis of IHD

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 39 Need and estimated level of provision in Wales (total number of cardiac rehabilitation centres n = 18)Mean SD Median IQR n Sum of Non-responding Total sum of patients (responding plus non-respondingall patients centres (n) centres)Mean Median IQRNo. of patients referred to CR 316.7 170.7 311 180–400 10 3,167 8 5,701 5,655 4,607–6,367No. of patients joining CR 185.4 93.0 150 119–248 13 2,410 5 3,337 3,160 3,005–3,650Eligible patients (source: PEDW) No. of Proportion of patients referred to CR (%) Proportion of patients joining CR (%)patients Data for non-responders imputed Data for non-responders imputedMean Median IQR Mean Median IQRAll patients with primary diagnosis of IHD 17,650 33 32 26–36 19 18 17–21All patients discharged with AMI, CABG, PTCA, UA 7,878 72 72 59–81 42 40 38–46All patients with primary diagnosis of IHD

Appendix 290TABLE 41 Estimated level of provision in Scotland (total number of cardiac rehabilitation centres n = 36)Mean SD Median IQR n Sum of Non-responding Total sum of patients (responding plus non-respondingall patients centres (n) centres)Mean Median IQRNo. of patients referred to CR 207.3 150.4 198.5 65–340 24 4,975 12 7,462 7,357 5,755–9,055No. of patients joining CR 132.2 101.3 100 47–218 21 2,777 15 4,761 4,277 3,482–6,047

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 3British Association for Cardiac Rehabilitationadditional postal questionnaireDear Cardiac Rehab Coordinator,We have recently been funded by the NHS Health Technology Assessment Research & Developmentprogramme to look at the provision, uptake and adherence to cardiac rehabilitation programmes incurrently under-represented groups, which includes women, elderly people, Black and Asian groups, andpatients with diagnoses of angina, heart failure and post-revascularisation. As part of this work, we needan up-to-date picture of current service provision. We are aware that you have recently completed aquestionnaire for the BHF/BACR survey, and we are collaborating with Dr Bethell and Sally Turner to usethe data you have kindly provided. There are a few additional questions we need to ask that were notcovered in the recent survey. We know you are extremely busy people so we have put together a shortquestionnaire that complements the BHF/BACR survey, that should only take a few minutes to complete.Any data will be added to the main BHF/BACR database. To be consistent, we are asking for informationrelevant to the period 1st January to 31st December 2000. Please contact us if you have any queriesregarding this. Please could you return the completed questionnaire in the SAE or fax through toFAO K Rees.Thank you so much for your help with this important workVery best wishesDr Karen Rees, BHF Research Fellow91© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 3IDADDITIONAL SHORT QUESTIONNAIRE TO COMPLEMENT THE BHF/BACRDATABASE OF UK CARDIAC REHABILITATION CENTRES 20011. How long is your supervised phase 3 programme for each patient? (please give anaverage figure for each component of CCR, relevant to your programme)Exercise componentNo. weeksNo. sessions/weekAverage lengthof sessionsAverage no.patients/sessionHealth education component (e.g. Healthy diet, smoking cessation)No. weeks No. sessions/week Average length Average no.of sessions patients/sessionPsychological component (e.g. Relaxation/stress management) if givenNo. weeks No. sessions/week Average length Average no.of sessions patients/sessionTotal for whole programme – if unable to break down into component sessionsNo. weeks No. sessions/week Average length Average no.of sessions patients/session92

Health Technology Assessment 2004; Vol. 8: No. 412. Do you make any special efforts to promote attendance at rehab sessions for anyof the following groups? (TICK ANY THAT APPLY)WomenPatients with heart failurePeople aged 65+Patients with anginaBlack and Asian groupsPatients who have hadCABG/PTCAPlease provide details of the methods you use:3. Do you have any spare capacity within your current service foradditional patients?YesNoIf yes, please indicate the number of additional patientsthat could be included without any increase in resources:patients/week4. During the last year approximately how many patients were referred?Total Male Female 65+ years post-MI CABG/PTCAHeart Angina Black and Asian groupsFailure93© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 3During the year how many patients joined the programme?Total Male Female 65+ years post-MI CABG/PTCAHeart Angina Black and Asian groupsFailureOf these patients, how many (eventually) completed the programme?Total Male Female 65+ years post-MI CABG/PTCAHeart Angina Black and Asian groupsFailureWith many thanks for your help in completing this questionnaire. Please could youpost back in the envelope provided, or fax through (FAO K Rees)by 31 st October 2001CR Coordinator/contact.......................................................................................CR Programme, where based................................................................................Region.....................................................................................................................94

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 4Literature search strategiesSearch terms for major databases are given.These terms were adapted appropriately forother databases.Search strategy for MEDLINE1 exp Heart diseases/2 coronary.tw.3 cardiac.tw.4 CABG.tw.5 myocardial.tw.6 angina.tw.7 heart failure.tw.8 heart disease$.tw.9 or/1-810 exp Rehabilitation/11 exp Rehabilitation centers/12 exp Rehabilitation nursing/13 rehabil$.tw.14 Aftercare/15 aftercare.tw.16 Convalescence/17 convalescen$.tw.18 recuperat$.tw.19 or/10-1820 9 and 1921 exp Heart diseases/rh [Rehabilitation]22 20 or 2123 Patient education/24 exp Counseling/25 exp Exercise therapy/26 Exercise/27 exp Psychotherapy/28 (patient adj2 educat$).tw.29 counsel$.tw.30 (behavi$ adj2 therap$).tw.31 psychosocial$.tw.32 ((lifestyle or life-style) adj2 intervent$).tw.33 ((exercise$ or fitness) adj5 (treatment orintervent$ or program$)).tw.34 ((lifestyle or life-style) adj5 (intervent$ orprogram$ or treatment$)).tw.35 Nurse practitioners/36 "nurse practitioner$".tw.37 or/23-3638 9 and 3739 (secondary adj5 prevent$).tw.40 Survival rate/41 (reduc$ adj5 (morbid$ or mortal$)).tw.42 Patient readmission/43 rehospitali$.tw.44 ((improv$ or increase$ or decrease$) adj5(recover$ or function)).tw.45 Disease management/46 (disease adj2 manage$).tw.47 Recovery of function/48 exp "Costs and cost analysis"/49 compliance.tw.50 adheren$.tw.51 non-compliance.tw.52 costs.tw.53 Patient compliance/54 or/39-5355 37 and 54 and 956 22 or 55Search strategy for EMBASE1 exp Heart disease/2 coronary.tw.3 cardiac.tw.4 CABG.tw.5 myocardial.tw.6 angina.tw.7 heart failure.tw.8 heart disease$.tw.9 or/1-810 exp rehabilitation/11 exp rehabilitation center/12 rehabil$.tw.13 exp convalescence/14 convalescen$.tw.15 recuperat$.tw.16 or/10-1517 9 and 1618 exp Heart disease/rh19 Heart rehabilitation/20 or/17-1921 exp patient education/22 exp counseling/23 exp kinesiotherapy/24 exp exercise/25 exp psychotherapy/26 (patient adj2 educat$).tw.27 counsel$.tw.28 (behavi$ adj2 therap$).tw.29 psychosocial$.tw.30 ((lifestyle or life-style) adj5 (intervent$ or95© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 4program$ or treatment$)).tw.31 ((exercise$ or fitness$) adj5 (treatment$ orintervent$ or program$)).tw.32 Nurse practitioner/33 "nurse practitioner$".tw34 or/21-3335 9 and 3436 exp survival/37 (patient$ adj2 readmi$).tw.38 rehospitali$.tw.39 ((secondary or tertiary) adj5 prevent$).tw.40 (reduc$ adj5 (mortal$ or morbid$)).tw.41 ((improv$ or increase$ or decrease$) adj5(recover$ or function$)).tw.42 (disease$ adj2 manag$).tw.43 exp aftercare/44 aftercare.tw.45 exp economic evaluation/56 costs.tw.57 Patient compliance/58 (compliance or non-compliance).tw.59 adheren$.tw.50 Patient satisfaction/51 or/36-5052 35 and 5145 20 or 52The results of all searching were downloaded intoa reference management database and thensearched across all fields for the following terms:AdherenceDropout*Drop-out*ComplyComplianceNoncomplianceParticipant*ParticipationReferral*Nonattend*Attend*Refusal*Patient attitude*Patient satisfaction*Barrier*Nonparticipant*Non-participant*Treatment refusalMotivat*CostCostsEconom*96

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 5Inclusion/exclusion formUptake and adherence tocardiac rehabilitationFirst authorSource and dateDatabasenumberDate assessedExcluded? Included?Uptake A AProf compliance B BAdherence C CAudits/descriptive I IEconomic E EReviewer (initials)Intervention toincrease uptake bypatientsIntervention toimproveprofessionalcomplianceInterventionto improveadherence1. Is an interventionevaluated?2. Patients:AMI, CABG, PTCA,Angina, Heart Failure,Other CVD (Specify)3. Outcome:people attending,losses to follow up,adherence (medical advice,therapy, clinical events,rehospitalisation, costs)If 1, 2 and 3 yes then include study4. Reason for exclusionOther information5. Audit/descriptive information6. Economic information97© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 6Data extraction form99© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 6Data extraction form – HTA SR of uptake, patient adherence and professional complianceReference:Database ID:Reviewer: (initials)Subject relevant to: (circle all that apply)A – Intervention to increase uptake by patientsB – Intervention to improve professional complianceC – Intervention to improve patient adherenceDate:1. Data Source: (circle) Published only Unpublished only mixedCountry of publication/recruitment –2. Study type: (circle)Between group comparisons:RCT (adequate allocation concealment)Quasi RCT (inadequate allocation concealment, e.g. alternate allocation,by hospital No., DoB etc.)Non-randomised trial (e.g. Allocation to groups but no attempt atrandomisation)Before and after study (comparing outcomes in different groups ofpatients before and after an intervention)Within group comparisons:Before and after study (comparing outcomes in the same patients beforeand after an intervention)3. Quality of studies:Creation of comparison groupsa) generation of random sequence method_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _b) concealment of allocation method_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _c) how allocation occurred (e.g. patient or doctor preference) detail _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ d) balance groups by design (e.g. matching) detail _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ e) within group comparisons (circle if applies)Comparability of groupsa) Were comparison groups similar at baseline? Yes No Unclearb) Were prognostic factors identified? Yes No Unclearc) Was case-mix adjustment used to account for differences between groups? Yes No Uncleard) (For within group comparisons only) — were only paired responses analysed? Yes No Unclear100

Health Technology Assessment 2004; Vol. 8: No. 41Blinding of outcomesa) Were outcomes assessed blind/independently of intervention? Yes No UnclearFollow-upa) Was there equal follow-up between groups? Yes No Unclearb) What was the overall loss to follow-up? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Not reportedSamplea) Prospective or retrospective sampling? Prospective Retrospectiveb) Were inclusion and exclusion criteria specified? Yes No Unclearc) Was the sample size planned (e.g. sample size calculation included)? Yes No Uncleard) Is representativeness of the sample assessed? (add comments) Yes No Unclear_ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _4. Participants studied: (circle all that apply)DiagnosisPost MI CABG/PTCA Heart Failure (chronic / secondary to MI) AnginaParticipantsMen Women Age limited (specify) Ethnic Minority GroupsFor the whole sample:Mean age (range) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _Percentage men _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _Case mix (specify, e.g. 100% MI or mixed diagnoses and proportions) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _Percentage white (if known) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __Inclusion/Exclusion criteria (if stated) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _5. Intervention (investigator’s description in as much detail as possible, including theoretical basis,intensity and duration, group or individual, setting, etc.)A – Intervention to increase uptake by patients_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _B – Intervention to improve professional compliance_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _C – Intervention to improve patient adherence_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _101© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 6Description of comparison group (treatment or usual care) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ __ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _Duration of follow-up (not duration of intervention) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _6. Outcomes: (circle all that apply)Specified primary outcome (specify) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _Attendance rates(give criteria used)Adherence to rehabilitation/medical advice/therapy(give criteria used)Clinical events Rehospitalisation Costs Changes in risk factorsOther (specify)Number of follow-up measurement points (give time intervals, e.g. 6 months, 1 year)_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _7. Study comparisons: (if multiple time points, use longest duration of follow-up)CharacteristicsBaseline characteristics:Number randomisedAge (mean SD range)Sex (% male)Outcomes:Attendance Nos (%)Adherence to medication/therapy –Nos adhering (%)Adherence to medication/therapy –Nos adhering (%)Losses to follow-up Nos (%)Specified primary outcome: N(%)Clinical events: N(%)Total Mortality N(%)Cardiac Mortality N(%)Non-fatal MI N(%)Revascularisation N(%)CVD event (stroke/TIA) N(%)Other N(%)Hospitalisation/RehospitalisationNumber (%) of patientsNumber of occasionsCosts (specify what)Other outcomes:Intervention/BeforeControl/After(additional columnfor 3 arm trial or 3time phase studies)1028. Notes (what did the investigators find? Interesting features?)

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 7Flow diagram of the systematic review ofinterventions to improve uptake of cardiacrehabilitation (QUOROM statement flow diagram)Potentially relevant publications identifiedand screened for retrieval3261Publications retrieved for more detailedevaluation957Publications excluded on the basis of titleand abstract (clear evidence that sourcepaper did not describe intervention inappropriate patient group)2304Publications excluded on the basis of titleand abstract:No intervention evaluated 776No outcome pertaining to uptakeof cardiac rehabilitation 154Publications included for data extraction27 (22 studies)Studies excluded from review:Published after June 2001 1No outcome 6No comparison group 1Retrospective design 1Outcome is referral 2Outcome is patient commitment 1Outcome relates to secondaryprevention 2__Total 14Studies included in review8103© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 8Studies evaluating interventions to improve theuptake of cardiac rehabilitation105© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 8106Studies are listed in terms of study design and the hierarchy of evidence, with RCTs first.Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedA significantimprovement in uptakeof outpatient CR wasobserved in the groupthat received theletters designed toinfluence acceptanceand attendance. Theauthors note that theintervention may haveworked by acting as afear message (wordingin letter: “research hasshown that attendancecan reduce thechances of dying fromanother heart attack”),rather than throughimplementation oftheory of plannedbehaviourWomen wereless likely toattend theprogramme,but neitherage nordistance livedfrom theprogrammepredictedattendanceUptake defined asattendance at theoutpatient CRprogramme. Uptakewas 86% in theintervention groupand 57% (authorsstate 59%) in thecontrol group( 2 = 7.91, df = 1,p < 0.0025)Nominal letterof thanks givento patients at3 days post-MIand thestandard letterdetailingcourse datesLetters based on thetheory of plannedbehaviour (Ajzen andMadden, 1986 107 )designed to increaseattendance at outpatientCR were given topatients 3 days post-MIand sent 3 weeks post-MI. The first letter wasdesigned to influenceacceptance and thesecond was designed toinfluence attendance.Patients also received anominal letter of thanksat 3 days and thestandard letter detailingcourse dates as sent tocontrol patients. Afterallocation to groups theCR nurse saw all patientsfor routine assessmentand personal invitation tothe programme. Forpatients who declinedthe offer of a place abrief second letter wassent wishing them welland informing them thatthey were still welcometo contact the teamAll patientspost-MI. Meanage 63 years,87% menPatients were handed asealed numberedenvelope with a nominalletter. Half of theenvelopes also containedan intervention letter.Envelope contentsknown to a researchassistant only and hadbeen allocated tointervention or controlby random numberassignment. Envelopesgiven to patients innumerical order. CRnurse not aware of groupassigned to; however, noprocedure in place tostop patients tellingnurse which letterreceived. Comparisongroups similar at baselineParallelgroup RCT,87 patientsrandomisedWyer et al.,2001 105,106UK(journal andthesis)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe authors note themotivating effect of anoutpatient careprogramme as a linkbetween inpatient CRand cardiac groupsAttendance at cardiacgroup after12 months. In thegroup with specialoutpatient care 57%of patients attended acardiac groupcompared with 27%of controls(p < 0.005)No outpatientcareprogrammeSpecial outpatient careprogramme to supportblue-collar workers afterMI to join coronarygroups. The programmeconsisted of fourdifferent conversationsbetween patients and asocial worker: at end ofrehabilitationprogramme, telephonecontact after 4 weeks,home visit after3 months and telephonecontact after 6 monthsPost-MIpatientsattendinginpatient CRprogramme.Mean age 52(33–60) years,89% menMethod of randomisationand allocationconcealment unclear.Comparison groupswere similar at baselineParallelgroup RCT,94 patientsrandomised.Resultsreportedfor 87(4 patientsdied, threerefusedfollow-up)Hillebrandet al., 1995 108Germany(journal)Relates toattendance atoutpatientheart groupafter inpatientCRcontinued107© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 8108Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe programmeproviding coordinatedfollow-up care byspecialist cardiac liaisonnurses did not improvehealth outcomes, butwas effective inpromoting at least oneoutpatient CR sessionattendanceSerum cholesterol,blood pressure,distance walked in 6minutes and smokingcessation did notdiffer betweengroups. Body massindex was slightlylower in theintervention group.More patients in theintervention groupattended at least oneoutpatient CR sessioncompared withcontrols (42% vs24%, p < 0.001). Thedifference was mostmarked in anginapatients (42% vs10%)No contactbetweenspecialistcardiac liaisonnurses andgeneralpractices. Notexplicitlystated, butunderstood tobe norecommendationto seepractice nurseand no patientheldrecordSpecialist cardiac liaisonnurses coordinated thetransfer of care betweenhospital and generalpractice. The liaisonnurse saw patients inhospital and encouragedthem to see the practicenurse after discharge.Support was provided topractice nurses byregular contact, includinga telephone call shortlybefore patient dischargeto discuss care and booka first follow-up visit tothe practice. Practicenurses were encouragedto telephone the liaisonnurse to discussproblems or to seekadvice on clinical ororganisational issues.Each patient was given apatient-held record cardwhich prompted andguided follow-up atstandard intervalsPatientsregistered with67 generalpractices in aspecifiedgeographicalarea. Patientsadmitted tohospital withMI (71%) orwith angina ofrecent onset(

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe lay volunteerhome visiting servicewas associated withincreased uptake ofoutpatient CR servicesAttendance at firstoutpatient CRappointment. In thedistrict with layvolunteer visiting 71%of patients attended afirst appointment atoutpatient CR. In thecontrol district 47%attended (p = 0.02).No difference inseveral clinicaloutcomes andindicators at first CRattendanceDistrict withno layvolunteervisitingHome visiting by layvolunteers. Visitors weretrained 5 hours per dayfor up to 7 days. Patientswere visited forapproximately30 minutes each weekfor 6 weeks. Layvolunteers providedadvice relating torehabilitation and offeredto accompany the patientto the first outpatient CRappointmentPost-MIpatients invitedto attend anoutpatient CRprogramme.Mean age72 years, 78%menGroups similar atbaselineNonrandomisedstudy. Onedistrict withintervention(98 patients)and onedistrictwithout(77 patients)Osika, 2001 82UK(thesis)The authors reportthat the informativepamphlet used tomotivate patients wasassociated withincreased attendanceat outpatient heartgroups. Patientsranked sources ofinformation: personalconversation withdoctor, pamphlet,lectures, personal talkswith caring personsand talks with otherpatientsNumber of patientswho joined anoutpatient heart groupafter 6 months. 78patients werefollowed up in boththe intervention andcontrol groups. 66.5%of patients whoreceived theintervention attendeda heart groupcompared with 31.0%in the control group( 2 = 20, df = 1,p < 0.001)Aftercompletion ofan inpatient CRprogrammepatients weregiven theaddresses oflocal outpatientheart groupsAfter completion of aninpatient CR programmepatients were given apamphlet withinformation aboutoutpatient heart groupsdesigned to motivatepatients to join. Thebooklet containedgeneral informationabout heart disease,including nutrition,exercise, relaxation andmedication. Patientswere also given theaddresses of localoutpatient heart groupsAll malepatients post-MIGroups of patientsattending in differentperiods. No informationcomparing groups atbaseline. 156 patientsfollowed up. Loss tofollow-up 22%Nonrandomisedtrial.200 patientsstudiedKrasemann andBusch, 1988 109Germany(journal)Relates toattendance atan outpatientheart groupafter aninpatient CRprogrammecontinued109© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 8110Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe authors concludethat a systemsapproach, involving aprompt for outpatientCR referral as part of adischarge critical carepathway, maypotentially increaserates of participation inCR for womenOverall participationat outpatient CR was54%, as determinedby patient self-report.The critical carepathway wasassociated with a nonsignificantincrease inoutpatient CRparticipation(OR 1.9, 95%CI 0.6 to 5.5)Beforeimplementationof critical carepathwayCritical care pathwayprompting referral foroutpatient CRAll patientspost-MI. Meanage 61 years,with 68% menNo information onbaseline characteristics ofthe two groupsBefore- andafter-study.Comparisonof outpatientCRparticipationrates in two6-monthperiods witha total of199 patientsMosca et al.,1998 41USA(conferenceabstract)Introduction ofpostdischarge homevisits by trainedcommunity nursesreported to beassociated with anincreased attendanceat outpatient CRPatient perception ofthe postdischargehome visiting service.Attendance atoutpatient CR byinvited patientsincreased from 55%before to 75% afterinstigation of theprogramme. Aspatient numbers arenot given, thesignificance of thiscannot be determinedPatientsdischargedfrom hospitalbefore thenurse homevisitingprogrammewas instigatedPostdischarge, preoutpatientCR homevisits by trainedcommunity nurses. ForMI patients this was2–3 weeks afterdischarge and for heartsurgery patients4–5 weeks. Visits wereaimed at: reduction ofanxiety levels, enablingpatients to make andmaintain lifestyle changes,identifying problems orpotential problems, andproviding patients withsupport, education andcounselling. The numberand frequency of homevisits were dependent onindividual requirementPost-MI andpostcardiacsurgery patientsNo information on thenumbers of patients orbaseline characteristics ofthe two groupsBefore- andafter-study.Comparisonof 12-monthaudit periodsImich, 1997 110UK(journal)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe authors suggestthat clinical guidelinescombined withfeedback of clinicalindicators to healthprofessionals wereuseful in improvingquality of care,including outpatient CRutilisation in MIpatients. However, theimprovement may bedue to parallel changesin levels of provisionClinical indicatorchanges pre- topostintervention. Nochanges were seen in-blocker, aspirin orangiotensin convertingenzyme inhibitor use.Lipid-lowering druguse increased from23% to 56%(p < 0.003).Outpatient CR servicebecame operational atstart of interventionperiod and showed asteady increase inutilisation rate from24% to 54%(p = 0.003)Beforedisseminationof evidencebasedclinicalguidelinesDissemination ofevidence-based clinicalguidelines for themanagement of AMI tohospital staff and GPs.Information on clinicalindicators was fed backto all hospital consultantphysicians, senioremergency staff, medicalservice directors andsenior clinicians. As partof the feedback theobserved proportion ofpatients receiving thetreatments wascompared with a qualitythreshold or minimumlevel of utilisationindicative of a reasonablestandard of care. Localproviders could compareand improve their ownpracticePost-MIpatients. Patientcharacteristicsonly availablefor pre- andpostinterventiongroups. Meanage 66 years,66% menPre- andpostintervention groupswere similar at baseline,but no information wasreported for theintervention period.Outpatient CRprogramme was not fullyoperational in thepreintervention period.Authors assumed thatpatient characteristics,diagnostic methods andtreatment modalitieswould remain essentiallyunchanged throughoutthe studyBefore- andafter-study.Comparisonof clinicalindicators inthree periods:preintervention(133 patients),implementationofintervention(271 patients),and postintervention(245 patients)Scott et al.,20001 111Australia(journal)111© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 9Studies excluded from the review of interventionsto improve uptake of cardiac rehabilitationAuthors, year Intervention Reason for exclusionCampbell et al., 1998 125,126Nurse-led clinic giving secondary prevention Outcome: use of secondaryassessment and adviceprevention, not attendance at CRCaulin-Glaser & Schmeizel, 2000 123 Education of health professionals about CR Outcome: referral not attendanceFeder et al., 1999 127 Leaflets promoting secondary prevention. Outcome: attendance at a generalAlso general practices received letters practice and drug prescribing, notsummarising effective secondary prevention CRForesman, 1997 120 Telephone invitation to CR programme No comparison groupJohnson, 2000 118 Nurse telephone follow-up No data: descriptiveKalayi et al., 1999 122 Computerised referral pathway Outcome: referral not attendanceKeck et al., 1991, 113 Keck &Comprehensive motivation programmeBudde,1996 114 integrated into CR to improve attendance No data: descriptiveat heart group after inpatient CRLevknecht et al., 1997 116 Outpatient clinical pathway No outcome data: descriptiveMcCarney et al., 2000 119General practice database identifies patients No data: descriptivefor home visit by health visitor to improvesecondary preventionMehta et al., 2000 121 Quality improvement initiative: critical care Retrospective study. Allocation topathway, patient education tool and staff groups according to physicianeducationpreferenceMillar 1993 115 Home visit by cardiac support worker No data: descriptivePasquali et al., 2001 112 Telephone call describing CR benefits and Out of review periodassistance with referralSuskin et al., 2000 124 Physician endorsement Outcome: commitment toparticipate, not attendanceTod et al., 1998 117 Integration of primary and secondary care No outcome data: descriptive113© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 10Flow diagram of the systematic review ofinterventions to improve adherence to cardiacrehabilitation (QUOROM statement flow diagram)Potentially relevant publications identifiedand screened for retrieval3261Publications retrieved for more detailedevaluation957Publications excluded on the basis of titleand abstract (clear evidence that sourcepaper did not describe intervention inappropriate patient group)2304Publications excluded on the basis of titleand abstract:No intervention evaluated 776No outcome pertaining toadherence with cardiac rehabilitation 143Publications included for data extraction38 (37 studies)Studies included in reviewStudies excluded from reviewEffectiveness of rehabilitationformats 9No outcome relating to adherence 10No comparison group 3Retrospective design 1__Total 2314115© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 11Studies evaluating interventions to improveadherence to cardiac rehabilitation117© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 11118Studies are listed in terms of study design and the hierarchy of evidence, with RCTs first.Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: details measure and results outcomes authors’country % men, mean involved, intensity, of usual care and results conclusions,(publication age (SD), follow-up period for interesting featurestype) ethnicity if stated outcome assessmentOverall, no improvementin compliance withrehabilitation sessions wasseen in patientsrandomised to aprogramme of selfmanagementtechniques,including signedagreement and diaries. Ina subgroup analysis, theauthor found highercompliance rates in thosepatients in theintervention group whosigned the agreement toparticipate than in thosewho did not, andsuggested the need forfurther investigation ofself-managementcompliance-improvingstrategiesCompliance defined asattendance at 60% ormore of the scheduled 48supervised cardiacrehabilitation sessions. (Ifany patient failed to attendfour consecutive sessionsthey were contacted bytelephone and urged tocontinue.) Data presentedas intention to treat.Compliance rate was 54%in the intervention group,and 42% in the controlgroup; these rates werenot statistically significant.Attendance of dropoutswas similar in theintervention and controlgroups (21% vs 16%) andwas also similar forcompliers (74% vs 76%).Not all patients in theintervention group signedthe agreement toparticipate. Compliancewas significantly higher inthe 48 subjects who signed(65%), than in the 15 whorefused (20%)UsualcomprehensivecardiacrehabilitationprogrammeUsual comprehensive cardiacrehabilitation programme, plusself-management techniquesincluding an agreement toparticipate in the programmefor 6 months to be signed bythe patient and coordinator,and self-report diaries tocomplete and be discussedwith the coordinator at regularintervals. Diaries includedsix graphs for plotting selfmonitoredsubmaximal heartrates each month, at 33%,50% and 75% of the maximumpower output achieved in theprevious exercise test, and six24-hour recall questionnaires ofdaily activities on a randomlychosen day to be completedeach month. In addition, aweight loss diary to fill in eachweek was given to those whoinitially agreed to lose weight,and similar diaries to recordnumber of cigarettes smokedeach day. Follow-up at the endof the intervention period of6 monthsMixed CHDpatients, MI 73%,CABG 16%,angina 12%.Mean age 50.5years, all menPatients randomisedusing a list of randomnumbers. Method ofallocationconcealment unclear.Blind assessment ofoutcomes unclear.Unclear whethercomparison groupswere similar atbaseline. Losses tofollow-up (defined asnon-attendance ateight consecutiverehabilitationsessions) were similarin the interventionand control groups(21% and 16%)Parallel groupRCT. 120patientsrandomised.Patientsstratified bysmokingstatus,occupation,leisure habitsand number ofpriorinfarctionsbeforerandomisation.Thesevariables wereshown to bepredictors ofdropout basedon previousexperience ofthis groupOldridge &Jones, 1983 145Canada(journal)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedOverall, no significantimprovement inattendance at exercisesessions with theintervention. There issome suggestion thateducational level is afactor associated withattendanceAttendance at exercisesessions over 3 months.Attendance for patients inthe intervention group was63.8%, and 62.2% in thecomparison group.Subgroup analysis revealedthat among theintervention group,attendance was greateramong better educatedpatients. Spouseparticipation, age, genderand occupation were notassociated with attendance,although the numbers inthese subgroups are likelyto be too small to drawfirm conclusionsComparisongroup patientsand spousesreceived thesame pamphletwith informationon the benefitsand drawbacksof exercise, asthe interventiongroup. This wasdone so allpatients wouldhave the sameinducement toenter theprogramme. Itwas thoughtunlikely that thissingleinterventionwould producelastingbehaviouralchangeOral persuasive communicationand education intervention toimprove patient adherence toexercise regimens. Interventiondeveloped from interviews withprevious patients and their spousesto elicit the most common beliefsof benefits and drawbacks to theexercise programme. Patients inthe intervention group received anoral persuasive communication onthe telephone in scriptedcounselling format to: convincethem of the benefits of regularexercise, warn them of likelydrawbacks so that expectationswould be realistic, acquaint themwith methods used by otherpatients to cope with drawbacks,and elicit an oral commitment toattend at least two classes perweek for the first 6 weeks. Inaddition, patients received a mailedwritten persuasive communicationto reinforce these points. Spousesalso received telephone counsellingto encourage the patient to attendand discuss methods that otherpatients spouses found useful. Awritten communication toreinforce these points was also sentto the spouse to increase thespouse’s support. Patients alsoreceived a pamphlet withinformation on benefits anddrawbacks of exercise. Allcommunication was tailored toindividual patients based on datacollected by questionnaire atbaselineMixed CHDpatients, 81% MI,63% with ahistory of angina,17% post-CABG.Mean age 53.8years, 88% men,95% whiteMethod ofrandomisation andallocationconcealmentunclear. Outcomeassessors wereblind to groupallocation, andcomparisongroups weresimilar at baseline.No losses tofollow-upreported at 3months, theperiod at whichmost data arepresented. Studywas powered todetect a 14%difference inattendance over3 monthsParallel groupRCT, 174patientsrandomisedDaltroy, 1985 146USA(journal)continued119© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 11120Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe authors hypothesisedthat either tape wouldresult in improvedoutcomes compared withcontrol, but also that theCoping Tape wouldexhibit the highest selfefficacybeliefs andgreatest compliance. Theyconcluded that viewingeither tape proved to bean effective method forincreasing dietary andexercise complianceduring the first 3 monthspostsurgeryAnxiety andself-efficacybeliefs.Hypothesisedthat thesewould affectcomplianceCompliance with lifestylechanges. Exercisecompliance assessed with asimple questionnaire ofleisure-time exercise (howmany times in the past 7days they participated inlight, moderate andstrenuous exercise).Dietary compliance wasassessed from thecholesterol–saturated fatsubscale of the diet habitsurvey. At 3 monthsexercise compliance wasimproved with bothinterventions (p < 0.05),but improvements weregreatest for the CopingTape at 1 month formoderate exercise, and at3 months for strenuousexercise. Both interventiongroups showed a reductionin dietary cholesterol andsaturated fat at 1 month(p < 0.05), but nodifferences were seen at3 monthsStandarddischargepreparation fromone of the twohospitals whererecruitment tookplace. Noattempt wasmade tointerfere withthis. Consistedalmostexclusively ofadvice not to lifttoo much, andinstructionsregarding theimportance ofexercise andmaintaining alow-fat dietVideotape interventionpostsurgery, shortly beforedischarge to enhance compliancewith exercise and diet. Twovideotape interventions:(1) Mastery Tape: depicts patientsas calm and confident at the timeof hospital discharge, makingsteady progress with nocomplications during 6 months,and adjusting to the recommendedexercise and low-fat diet withrelative ease; (2) Coping Tape:edited so the same patientsmention concerns they areexperiencing about hospitalrelease and cope with effort, butsuccessfully, with a variety ofdifficulties (e.g. fatigue, dietchanges), so the recovery isportrayed as a steady forwardprogression of ups and downs.Exercise and dietary complianceassessed at 1 and 3 monthspostdischargeConsecutiveelective first timeCABG, mean age61.4 (3.2 years),86.5% men,83.3% whiteMethod ofrandomisation andallocationconcealmentunclear. Blindassessment ofoutcomes unclear.Unclear whethercomparisongroups weresimilar at baseline,as demographicdata are notpresentedseparately foreach group.Overall losses tofollow-up at3 months were9%Parallel groupRCT, twointerventiongroups andone controlgroup,215 patientsrandomisedMahler et al.1999 147USA(journal)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe authors conclude thatthis nursing interventionwas effective atsupporting healthy eatingand patients’ self-careagencyAppraisal of selfcare and eatinghabits whichshowedimprovementwith theintervention,and self-efficacyfor healthyeating, whichshowed nochange withtheinterventionAdherence to dietaryadvice assessed fromdietary record and foodhabits questionnaire. Totalfat and saturated fatpercentage of calories wassignificantly reduced in theintervention group(p < 0.01). Significantimprovement on the foodhabits questionnaire wasseen in the interventiongroup (p < 0.05)Control groupreceived followuptelephonecalls and homevisit at 7 weeksto collect data on3-day dietaryrecord andassess foodhabits. Adviceabout diet wasnot given unlessthe patientintroduced thesubjectNursing intervention of nutritionalself-care based on Orem’stheory. 176 Home visit to collectbaseline 3-day diet record, assessfood habits and provideinformation about nutritional goalsfor healthy hearts, and determinewhether these were being met.Suggestions for changes given anda commitment on the part of thepatient sought to make thesechanges. Three follow-uptelephone calls over 6 weeks anda further home visit to collect3-day diet record and assess foodhabits at 7 weeks postdischarge.Data compared between baselineand 7 weeks postdischargeMI patients,majority first MI.Patients recruitedwhile in hospital.Mean age 62 (11)years, 60% menMethod ofrandomisation andallocationconcealmentunclear.Comparisongroups weresimilar at baseline.Blind assessmentof outcomesunclear. Losses tofollow-up notreportedParallel groupRCT,104 patientsrandomisedAish & Isenberg,1996 148Canada(journal)continued121© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 11122Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedMany hypotheses testedwith relatively smallnumber of patients. Theauthors conclude thatthey found no differencesbetween groups onmeasures of adherence,self-motivation, selfefficacyor internal healthlocus of controlVariouspsychologicalmeasures usedto determinerelationshipswithattendanceTotal adherence to themaximum number ofexercise sessions. This was90% in the interventiongroup and 89% in thecontrol group (notsignificant)During thecourse of theexerciseprogrammepatients receiveda ‘benign’educationintervention,which coveredbasic exerciseconcepts,guidelines forproper exerciseparticipation,exercise tips andhandouts, andthe benefits ofexerciseMotivational relapse preventionintervention received during thecourse of the cardiac rehabilitationprogramme. The cardiacrehabilitation programme consistedof three weekly exercise sessionsof 30–40 minutes over 2–3 months.The intervention was started afterfour or five exercise sessions. Theintervention was based on Marlattand Gordon’s model. 177 Patientsreceived individual sessions, one aweek for 3 weeks. Session 1: usingpretest information, factors foundto interfere with adherence wereintroduced. Patients discussed theirperceptions on the value ofexercise, listed their goals for theprogramme and anticipatedoutcomes. Session 2: patients wereintroduced to decision-makingconcepts and cognitive interferencefactors. Discussion with regard tocoping with ‘slips’ and introductionto appropriate ways to reframeperspectives. Patients filled in dailyactivity sheets. Session 3: focusedon the importance of lifestylebalance. Patients were asked torefer to daily activity sheets tointroduce concepts of shoulds andwants. Stressors were identifiedthat may impact on lifestyle balanceand discussed, as was theimportance of positive thinking anduse of medication. Patients alsotook part in a stress managementexercise and relaxation procedureSubjects recruitedfrom a phase 2cardiacrehabilitationprogramme.Mixed cardiacpatients includingMI, CABG, anginaand patients withvalve problems.Mean age 62(range33–77) years,gender notmentioned, 95%whiteAllocation togroups bypresentingpatients with apacket containinga form coded Aor B. Blindassessment ofoutcomes unclear.Similarity ofgroups at baselineunclear. Overalllosses to followup22%Quasi RCT,41 patientsrandomisedAshe, 1993 149USA(PhD thesis)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe author concludes thatpersonal independence ofparticipants after a CRprogramme can beimproved by means oftelephone contactconducted afterconclusion of theprogrammeExerciseconfidencescale (selfefficacy),personalattitudes scale(depression),resting andexercise bloodpressure, bloodlipids andgraded maximalexercise test.No differencesbetweeninterventionand controlExercise habit, intention toexercise and conditionsthat facilitated ordiscouraged theperformance of exerciseassessed by self-completedquestionnaire. Nodifference in exercise habitor intention to exerciseassociated with theintervention. Conditionsthat facilitated theperformance of exercisewere improved in theintervention groupcompared with control(p

Appendix 11124Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedVery small feasibility studywith only four patientsrandomised to eachgroup. Intervention ofbehavioural feedback hadpositive effects on sodiumintake and attitudestowards adherence torecommended sodiumlevelsRepeat of the 3-day dietarylog of sodium intake at theend of the interventionperiod, and completion of thebeliefs about diet instrument(measure of benefits andbarriers related to dietarysodium adherence). Sodiumintake was significantly less inthe intervention group (1569vs 2836 mg, p < 0.05), andthese patients identifiedsignificantly fewer barriers toadhering to sodiumrecommendationsPatients in therehabilitationprogrammecompleted the3-day dietary logon sodium intakeCardiac rehabilitation plusbehavioural feedback interventionon dietary sodium intake.Feedback given on a 3-day dietarylog of sodium intake, discussion ofproblem-solving strategies toreduce future sodium intakeHeart failurepatients who werecurrentlyparticipating in acardiacrehabilitationprogramme. Nodetails of age orgender, meanejection fraction24%No details ofmethod ofrandomisation,allocationconcealment,blinding ofoutcomeassessors,comparability ofgroups at baselineor losses tofollow-upParallel groupRCT, eightpatientsrandomised.Feasibilitystudy, hencesmallnumbersDuncan et al.,2001 151USA(conferenceproceedings)There was no effect ofthe intervention onprogramme attendance.The authors conclude thatthe increase in exerciseknowledge suggests thatthe contingencycontracting processmay be an interventionworthy of considerationin addition to traditionalgroup lecturesExercise knowledge test andadherence to exercisesessions. There was asignificant increase in exerciseknowledge scores at 8 weeksin the intervention group, butno difference in attendancerates between theintervention and controlgroups (90% and 89%,respectively)Patients in thecomparisongroup receivedthe same 1 hourper week offormal educationas theinterventiongroupContingency contracting.Contracting in this study is definedas a written contract negotiatedwith the patient, stating how longthey will exercise for to maintaintheir heart rate at a certain level,on how many days that week, inreturn for a reward. Patientschose rewards that were generallyacceptable providing they werenot damaging to health. Examplesincluded dietary recipes, T-shirtsor loan of exercise equipment. Allcontracting done by the educationcoordinator of the cardiacrehabilitation programme, withsessions lasting for 10 minuteseach week for 8 weeks. Contractbehaviours were designed to beattainable, and exercise sessionswere routinely scheduled threetimes a week. Upon completion ofthe contract patients receivedtheir reward. Patients alsoreceived 1 hour of formaleducation each week. Follow-upassessment at 8 weeksMixed CHDpatient: MI 25%,CABG 43%,angioplasty 18%,other coronarydisease 14%.Mean age55.6 years, 71%menAlternateallocation tointervention orcontrol. Blindassessment ofoutcomes unclear.The comparisongroups weresimilar at baselinein terms of ageand gender butthere were moreCABG and fewerMI patients in thecontrol group.There was a 7%overall loss tofollow-up over an8-week periodQuasi RCT,30 patientsrandomisedLeslie & Schuster,1991 152USA(journal)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe authors conclude thatno differences werefound between theintervention or controlgroups for medicalregimen adherence. Theauthors observed thatattitudes and perceivedbeliefs of others werepredictive of adherenceand should be included inany rehabilitationprogramme. Repeatedself-evaluationquestionnaires and nursevisits may have acted asan intervention in thecontrol groupCompliance with medicalregimen, divided into diet,smoking, activity, stress andmedications. Various healthbehaviour and attitudescales used to assess this.No significant differenceswere seen betweenintervention and controlgroups at 60 days or 1 yearpostdischargeAll patients had10–15 days ofcardiacrehabilitation inhospital,including adviceon diet, activity,smoking andmedication.Patientsinterviewedbefore dischargeand visited athome 30 dayspostdischarge tocollectassessment dataas for theinterventiongroup.Assessment ofcompliance withmedical regimenat follow-uphome visits at60 days and1 yearAll patients had 10–15 days ofcardiac rehabilitation in hospital,including advice on diet, activity,smoking and medication. At 30days postdischarge nurses visitedpatients at home to deliver anintervention to improve medicalregimen adherence whichconsisted of three steps:(1) Assessment: data obtained onpatient self-assessment of attitudesand perceived beliefs of otherstowards regimen compliance,personal psychological and societaladjustments, and reportedregimen compliance by patient andspouse, predischarge and 30 dayspostdischarge. (2) Problemidentification: data from step 1were evaluated by the patient,spouse and nurse. Problem areaswere defined and factorscontributing to non-compliancewere discussed. (3) Goal setting:on the basis of problemsidentified, alternative actions werediscussed and a health plan withspecific goals was developed.Assessment of compliance withmedical regimen at follow-uphome visits at 60 days and 1 yearAll patientsrecruited from inpatient cardiacrehabilitationprogrammes (oneof three similarprogrammes)following a firstuncomplicated MI.Age range30–65 years, 81%men. Dataavailable for onlythose patientsfollowed up, notall thoserandomisedAlternateallocation tointervention orcontrol. Blindassessment ofoutcomes unclear.The comparisongroups weresimilar at baseline.There was an11% overall lossto follow-up overa 60-day period,and 30% loss tofollow-up at1 yearQuasi RCT.115 patientsrandomisedMiller et al.,1988 143Miller, et al.,1989 144USA(journals)continued125© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 11126Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe authors conclude thatthe analyses showed noeffect of thepsychotherapyintervention on patientcompliance measured byphysiological or selfreportmeasures. Theydid, however,demonstrate that patientsreceiving the interventionattended significantlymore prescribed exercisesessionsPhysiological markers ofcompliance with exercise(heart rate and bloodpressure). Attendance atprescribed exercisesessions, and self-report ofexercise at home: rated aspoor (once a week for lessthan 20 minutes or none,score 0–1), good (threetimes a week for20–30 minutes, score 2) orexcellent (more than threetimes a week for morethan 30 minutes, score 3).Mean scores for self-reportmeasures of compliancewere 2.57 and 2.37 for theintervention and controlgroups, respectively, overthe 3 month period. Theintervention groupattended 88.4% of theprescribed exercisesessions, the control group75.7% (p < 0.05)Control groupreceived usualcardiacrehabilitation12-week psychotherapyintervention and its impact onpatient compliance withprescribed regimen in the cardiacrehabilitation setting. Patientsreceived the intervention whenthey first started the cardiacrehabilitation session. Interventionconsisted of 90-minute sessionsonce a week for 3 months, or12 or 15 sessions of insightorientatedgroup psychotherapy.The primary focus was to help thegroup to communicate thoughtsand feelings with the associatedaffects. The therapist created anatmosphere that was supportive,cooperative and goal directed. Thesecondary focus was to highlightand promote change in thosebehaviours and styles of relatingthat reflected a maladaptivecontrol orientation, e.g. hesitancyor refusal to comply with physicianrecommendations, or resistance togroup participation. Period offollow-up at end of interventionperiod of 3 monthsPatients referredto cardiacrehabilitation.Mixed CHDpatients: 62% MI,24% CABG, 15%both. Mean age 59(range28–72 years),86.8% menPatients randomlyassigned tointervention orcontrol as theypresentedthemselves. Onoccasions wherethere wereinsufficientnumbers to forman interventiongroup, the authorsdeparted fromtheir standardrandomisation andpriority was givento theinterventiongroup, hence thisgroup is larger(n = 22 vsn = 12). Blindingof outcomeassessors unclear.Comparisongroups weresimilar at baseline.Overall losses tofollow-up 29%Quasi RCT,48 patientsrandomised,datapresented for34 completersLack, 1985 153USA(PhD thesis)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe authors conclude thatoverall compliance andactivity levels weresignificantly higherpostsurgery in thestructured teaching groupKnowledgescores postsurgery weresimilar for thestructured andunstructuredteaching groupsCompliance with diet,medication, smokingcessation and activity, andknowledge. Complianceassessed by self-report.Patients were assigned ascore on follow-up visit ofcompliance based onindividual risk factors.Overall compliance scoreswere 86.8 in theintervention group, and79.5 in the control group(p < 0.05). Compliancewas better in theintervention than in thecontrol group for activity(15.6 vs 7 blocks walked,p < 0.005). None of thepatients in the interventiongroup reported noncompliancewith any of therisk factors measuredReceivedteaching byunstructuredmethodStructured teaching programme toincrease patients’ knowledge andcompliance to medication, diet,smoking cessation and exercisepost surgery. Teaching guidedeveloped by nurses withconsiderable experience with thispatient group, and used by nurses.Follow-up assessment at4–6 weeks postsurgeryCABG patients.Elective CABGperformed by thesame group ofsurgeons. Meanage 59 (range46–78 years),70% menData werecollected fromcontrols first(n = 30) and theninterventionpatients (n = 30).Comparisongroups weresimilar at baseline.Sampling wasprospective andinclusion criteriawere specified.Losses to followupnot reportedNonrandomisedtrial,60 patientsstudiedMarshall et al.,1986 154USA(journal)The authors conclude thatat each time-point, theadherence strategy groupshowed significantlybetter compliance withcardiac rehabilitationCompliance expressed asrelative risks, defined asattendance at 66% ormore sessions. 12 weeks,RR 2.3 (95% CI 1.8 to 2.9),24 weeks, RR 2.9 (2.3 to3.7), 52 weeks, RR 4.25(3.2 to 5.6) (log-rank testbetween strategiesp < 0.001)TraditionalrehabilitationTraditional rehabilitation and anadherence strategy which includeda signed commitment torehabilitation, family involvement,sports and recreational activities toincrease patient commitment andinvolvement in a variety ofactivities, and three weekly talkson CHD-related topics. Follow-upat 12, 24 and 52 weeksPatients withCHD. Mean age56 (SD 10) years,67.5% menComparisongroups weresimilar at baselineand sampling wasprospective. Nofurther details aregivenNonrandomisedtrial,509 patientsstudied,interventiongroupn = 229,control groupn = 280Huerin et al.,1998 155Argentina(conferenceproceedings)continued127© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 11128Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff involved, group: details measure and results outcomes authors’country % men, mean intensity, follow-up period of usual care and results conclusions,(publication age (SD), for outcome assessment interesting featurestype) ethnicity if statedThe authors conclude thattailored exercise fordifferent patient groupsincreased compliancewith the exercisecomponent of standardrehabilitation. Patientnumbers not reported.The magnitude of changescannot be assessedAttendance at cardiacrehabilitation. Attendancewas 82% in the lowintensityexercise groupand 34% in the standardrehabilitation comparisongroup. Attendance in thewomen-only group was75%StandardrehabilitationTwo interventions examined.(1) Low-intensity exercise toincrease compliance, offered topatients who were previouslyunable to take part in standardexercise owing to co-morbidity.The comparison group werestandard rehabilitation. (2)Women-only groups: compliancewith rehabilitation. Historicallycompliance in women withstandard rehabilitation was 6%MI patients, meanage 62 years, 75%men. Mean age ofthe women-onlygroup was60 yearsNo more detailspresented abouthow the studieswere conductedTwo studydesigns: nonrandomisedtrial, exercisevs. standardrehabilitation,and anevaluation ofa women-onlyprogrammecomparedwith historicalcontrols. Noinformationon patientnumbersMcKenna et al.,1998 156UK(conferenceproceedings)The authors conclude thateven for patients whosespouses did not attendrehabilitation there werebeneficial effects of groupsupport when the spousalprogramme was runningcompared with theprespousal programmeAttendance at cardiacrehabilitation sessionsdefined as those whoattended at least 50% ofthe sessions. Before thespousal interventionattendance was 44%, afterthe spousal supportprogramme where thespouse participated inrehabilitation patientattendance was 90%(p < 0.001), and where thespouse did not attend thiswas 67%Patientcompliancebeforeintroduction ofthe spousesupportprogrammeSpousal support programme:spousal participation in cardiacrehabilitation. Not all spousesattended, so comparators arebaseline before spouseparticipation (n = 30), spouseparticipation in cardiacrehabilitation (n = 30) and nospouse participation (n = 30).Follow-up at 6 monthsCHD patients. Noother detailsProspectivesampling, inclusioncriteria notspecified. Noother detailsabout how thestudy wasconductedBefore- andafter-studyErling &Oldridge,1985 157Canada(conferenceproceedings)

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 12Studies excluded from the review of methods toimprove adherence to cardiac rehabilitationAuthors, year Intervention Reason for exclusionAdes et al., 2000 134 Home-based telephone-monitored Authors report on effectiveness of CR formats. PatientsCR compared with outpatient CR were not randomised but allocated to home-basedtelephone monitored CR if living in remote area orunable to attend outpatient CR due to work or timeconstraintsBaile & Engel 1978 173 Follow-up of non-compliant No comparison groupcoronary care unit patientsBarnason & Postdischarge telephone follow-up No relevant outcomeZimmerman, 1995 167 or group teachingBlumenthal et al., 1988 159 Comparison of high- and low- Comparison of rehabilitation intensities. Authors reportintensity exercise trainingon effectiveness. RCT of patients representative ofmost US CR programmesBrubaker et al., 1996 160 Comparison of standard and Comparison of rehabilitation duration. Authors reportextended length CRbenefit from continuing CR for >1 year. Retrospectivenon-randomised comparison of patients who attended3 months or >1 year of CRCampbell et al., 1998 126 Nurse-run clinics in general practice No relevant outcomeCarlson et al., 2000 73 Comparison of traditional and partly Comparison of rehabilitation formats. Authors reporthome-based CRincreased total exercise sessions with partly homebasedprogramme. Randomised trialDeBusk et al., 1985 131 Comparison of home and group Comparison of rehabilitation formats. Authors reportexercise trainingon effectiveness. Randomised trialDracup et al., 1984 162 Group counselling No relevant outcomes.Gordon & Haskell, 1997 168 Physician-supervised, nurse No comparison group. No relevant outcomescase-manager CR modelLabrador et al., 1998 169 Physician-directed, nurse-supervised No relevant outcomescase-management programmeLee et al., 1996 161 Comparison of high- and low- Comparison of rehabilitation intensities. Authors reportintensity exercise trainingon effectiveness. Randomised trialLinde & Janz, 1979 164 Postoperative teaching programme No relevant outcomesMehta et al., 2000 121 Quality improvement initiative: Retrospective study. Allocation to groups according tocritical care pathway, patient physician preferenceeducation tool and staff educationPenckofer & Llewellyn, Comparison of education by Comparison of education interventions. Authors report1989 163 structured and unstructured methods little extra benefit from structured teaching. NotrandomisedSenaratne et al., 2001 171 Lipid management by cardiac Outside search period. No relevant outcomerehabilitation nurseSkof et al., 2001 172 Comparison of late outpatient and No relevant outcomeinpatient CRcontinued129© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 12Authors, year Intervention Reason for exclusionSparks et al., 1993 158 Home-based telephone-monitored Comparison of rehabilitation formats. Authors reportCR compared with outpatient CR on effectiveness. Randomised trial.Starkey et al., 2000 175 Computer-facilitated secondary No comparison groupprevention programmeStern & Cleary, 1981 174 Low-level exercise programme No comparison groupUnden et al., 1993 165 Nurse support No relevant outcomeVale et al., 2000 170 Telephone coaching by dietitian No relevant outcomevan Elderen et al., 1994 166 Group health education programme No relevant outcome130

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 13Flow diagram of the systematic review ofinterventions to improve professional compliancewith cardiac rehabilitation (QUOROM statementflow diagram)Potentially relevant publications identifiedand screened for retrieval3261Publications retrieved for more detailedevaluation957Publications excluded on the basis of titleand abstract (clear evidence that sourcepaper did not describe intervention inappropriate patient group)2304Publications excluded on the basis of titleand abstract:No intervention evaluated 776No outcome pertaining to professionalcompliance with cardiac rehabilitation 163Publications included for data extraction18 (17studies)Studies included in reviewStudies excluded from review:No outcome 7No comparison group 2Retrospective design 1Descriptive only 1__Total 116131© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 14Studies evaluating interventions to improveprofessional compliance with cardiac rehabilitation133© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 14134Studies are listed in terms of study design and the hierarchy of evidence, with RCTs first.Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe programmeproviding coordinatedfollow-up care byspecialist cardiac liaisonnurses did not improvehealth outcomes, butwas effective inpromoting at least oneoutpatient CR sessionattendanceSerum cholesterol,blood pressure,distance walked in 6minutes and smokingcessation did notdiffer betweengroups. Body massindex was slightlylower in theintervention group.More patients in theintervention groupattended at least oneoutpatient CR sessioncompared withcontrols (42% vs24%, p < 0.001). Thedifference was mostmarked in anginapatients (42% vs10%)No contactbetweenspecialistcardiac liaisonnurses andgeneralpractices. Notexplicitlystated, butunderstood tobe norecommendationto seepractice nurseand no patientheldrecordSpecialist cardiac liaisonnurses co-ordinated thetransfer of care betweenhospital and generalpractice. The liaisonnurse saw patients inhospital and encouragedthem to see the practicenurse after discharge.Support was provided topractice nurses byregular contact, includinga telephone call shortlybefore patient dischargeto discuss care and booka first follow-up visit tothe practice. Practicenurses were encouragedto telephone the liaisonnurse to discussproblems or to seekadvice on clinical ororganisational issues.Each patient was given apatient-held record cardwhich prompted andguided follow-up atstandard intervalsPatientsregistered with67 generalpractices in aspecifiedgeographicalarea. Patientsadmitted tohospital withMI (71%) orwith angina ofrecent onset(

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe authors suggestfurther study requiredto evaluate whetherphysician endorsementimproves CRparticipationNo effect ofmethod ofdelivery (inpersonvsnot)Self-reportedcommitment toparticipate in CR was62% in those whoreceived the physicianendorsementcompared with 38%in those receiving ageneric endorsement(p = 0.08)GenericendorsementWritten endorsement byattending physicianPost-MIpatients. 76%menNo information onrandomisation procedureor baselinecharacteristics of the twopatient groupsRCT, 50patientsrandomisedSuskin et al.,2000 124Canada(conferenceabstract)The authors note thatas well as improvingpatient care the use ofelectronic referral andfeedback of referralrates was of benefit tostaff, saving timereferring andidentifying patientsMonthlyreferral rateat start ofstudy 15/37(40%).Within3 months ofinterventionmonthlyreferral 35/39(90%)After interventionreferral increasedfrom 194/298 (65%)to 208/263 (79%)(p = 0.0002)Beforeimplementationof referralpathwayinterventionElectronic referralpathway with feedbackto ward staff on referralrates. The referralpathway was initiatedwhen a CR referralscreen was automaticallyflagged up on theelectronic patient recordof those patients with adischarge diagnosis of MIPost-MIpatients. Noinformation onage or genderof groupsNo information onbaseline characteristics ofthe two patient groupsBefore- andafter-study.Comparisonof CRreferral ratesbeforeintervention(298 patients)and afterintervention(263 patients)Kalayi et al.,1999 122UK(journal)The authors concludethat a systemsapproach, involving aprompt for outpatientCR referral as part of adischarge critical carepathway, maypotentially increaserates of participation inCR for womenOverall participation atoutpatient CR was54%, as determined bypatient self-report. Thecritical care pathwaywas associated with anon-significant increasein outpatient CRparticipation (OR =1.9, 95% CI 0.6 to 5.5)Beforeimplementationof critical carepathwayCritical care pathwayprompting referral foroutpatient CRAll patientspost-MI. Meanage 61 years,68% menNo information onbaseline characteristics ofthe two groupsBefore- andafter-study.Comparisonof outpatientCRparticipationrates in two6-monthperiods witha total of199 patientsMosca et al.,1998 41USA(conferenceabstract)continued135© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 14136Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe authors concludethat appropriatelydesigned educationalinterventions aresuccessful in alteringbehaviours ofhealthcare personneland improving referralto CRIn-hospital referralincreased by 50%(p < 0.05). Physicianoffice referralincreased by 61%(p < 0.05). Male andfemale referral ratesincreased by 41%(p < 0.05) and 65%(p < 0.05),respectivelyBeforeimplementationof educationalinterventionEducational interventionfor healthcare providerson the comprehensivenature and benefits ofCR, with particularemphasis on women.Consisted of grandrounds for medical andnursing staff, monthlyreview of healthoutcomes and costeffectivenessof CR withclinical cardiologycouncil, placement of CRinformation inexamination offices, andinstructions for nurses todiscuss CR with patientsand encourage discussionof referral withphysiciansPostrevascularisationpatients.No informationon age orgender ofgroupsNo information onbaseline characteristics ofthe two patient groupsBefore- andafter-study.Comparisonof outpatientCRparticipationrates in two5-monthperiods.Patientnumbers notspecifiedCaulin-Glaser& Schmeizel,2000 123USA(conferenceabstract)continued

Health Technology Assessment 2004; Vol. 8: No. 41Authors, Study type Study quality Participants’ Intervention: details Comparison Principal outcome Other Comments:year and and size case-mix, including setting, staff group: measure and outcomes authors’country % men, involved, intensity, details of results and results conclusions,(publication mean age follow-up period for usual care interesting featurestype) (SD), outcome assessmentethnicity ifstatedThe authors suggestthat clinical guidelinescombined withfeedback of clinicalindicators to healthprofessionals wereuseful in improvingquality of care,including outpatient CRutilisation in MIpatients. However, theimprovement may bedue to parallel changesin levels of provisionClinical indicatorchanges pre- topostintervention. Nochanges were seen in-blocker, aspirin orangiotensin convertingenzyme inhibitor use.Lipid-lowering druguse increased from23% to 56%(p < 0.003).Outpatient CR servicebecame operational atstart of interventionperiod and showed asteady increase inutilisation rate from24% to 54%(p = 0.003)Beforedisseminationof evidencebasedclinicalguidelinesDissemination ofevidence-based clinicalguidelines for themanagement of AMI tohospital staff and GPs.Information on clinicalindicators was fed backto all hospital consultantphysicians, senioremergency staff, medicalservice directors andsenior clinicians. As partof the feedback theobserved proportion ofpatients receiving thetreatments wascompared with a qualitythreshold or minimumlevel of utilisationindicative of a reasonablestandard of care. Localproviders could compareand improve their ownpracticePost-MIpatients.Patientcharacteristicsonly availablefor pre- andpostinterventiongroups.Mean age66 years, 66%menPre- andpostinterventiongroups were similar atbaseline, but noinformation wasreported for theintervention period.Outpatient CRprogramme was notfully operational in thepreinterventionperiod. Authorsassumed that patientcharacteristics,diagnostic methodsand treatmentmodalities wouldremain essentiallyunchanged throughoutthe studyBefore- andafter-study.Comparison ofclinicalindicators inthree periods:preintervention(133 patients),implementationof intervention(271 patients)and postintervention(245 patients)Scott et al.,2000 111Australia(journal)137© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 15Studies excluded from the review of interventionsto improve professional compliance withcardiac rehabilitationAuthors, year Intervention Reason for exclusionAxtell et al., 2001 191 Inclusion of pharmacist in MI care Outcome: use of medicationCampbell et al., 1998 125,126 Nurse-led clinic giving secondary prevention Outcome: use of secondaryassessment and adviceprevention, not attendance at CR.Feder et al., 1999 127 General practices received letters with summary Outcomes: attendance at a generalof effective secondary prevention with reference practice and drug prescribing, notto local guidelines. Also prompts to patients CRHillert et al., 2000 192 Risk factor management through physician No comparison groupeducation, participation and consensusdevelopmentLevknecht et al., 1997 116 Outpatient clinical pathway No outcome data: descriptiveLinde & Janz, 1979 164 Nurse training to master’s level in postoperative Inpatient programme. Outcome:teaching programmepatient knowledge and follow-up.Data not interpretableMcCarney et al., 2000 119 General practice database identifies patients for No data: descriptivehome visit by health visitor to improve secondarypreventionMehta et al., 2000 121 Quality improvement initiative: critical care Retrospective study. Allocation topathway; patient education tool and staffgroups according to physicianeducationpreferenceSenaratne et al., 2001 171 Lipid management by cardiac rehabilitation nurse Outside search period. Outcome:lipid levelsStarkey et al., 2000 175 Computer-facilitated secondary prevention No comparison groupprogrammeTod et al., 1998 117 Integration of primary and secondary care No outcome data: descriptive139© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 16Estimates for unit costs for different staff categoriesand gradesStaff category Costs (£)Salary a Salary Over- Capital Total Staff Totalon-costs heads b costs c hours costs costsworked per perper year c hour d hourStaff nurse B grade 11,820 1,507 e 2,216 1,288 1,575 8 11Staff nurse E grade 18,222 2,323 e 2,216 1,288 1,575 13 15Staff nurse F grade 21,010 2,679 e 2,216 2,263 1,575 15 18Staff nurse G grade 23,948 3,053 2,216 2,263 1,575 17 20Staff nurse H grade 26,540 3,384 e 2,216 2,263 1,575 19 22Physiotherapist helper 10,865 1,130 2,216 2,775 1,584 8 11Physiotherapist basic 17,202.50 2,193 f 2,216 4,302 1,512 13 17Physiotherapist senior I 23,452 2,990 2,216 4,302 1,512 17 22Physiotherapist senior II 20,670 2,635 f 2,216 4,302 1,512 15 20Physiotherapist superintendent III 25,832.50 3,294 f 2,216 4,302 1,512 19 24Physiotherapist superintendent IV 23,452.50 2,990 f 2,216 4,302 1,512 17 22Sport scientist 22,767.50 2,903 f 2,216 4,302 1,512 17 21Exercise physiologist (MTO 3) 20,647 2,632 f 2,216 4,302 1,512 15 20Occupational therapist basic 17,202.50 2,193 g 2,216 4,302 1,512 13 17Occupational therapist senior I/II 21,785 2,778 2,216 4,302 1,512 16 21Occupational therapist head 26,467.50 3,375 g 2,216 4,302 1,512 20 24Dietitian 21,785 2,778 e 2,216 3,606 1,554 16 20Dietitian senior I/II 23,452.50 2,990 e 2,216 3,606 1,554 17 21Pharmacist 32,983.50 4,205 e 2,216 3,606 1,554 24 28GP NA NA NA NA NA 54 62Medical consultant 67,064 9,664 24,320 4,161 1,640 47 64Clinical psychologist 38,316 5,364 3,978 2,144 1,476 30 34Cardiac technician (MTO 4) 25,118 3,203 e 2,216 4,302 1,512 19 23Social worker 19,951 2,709 3,399 2,007 1,554 15 18Secretary NA NA NA NA NA 10 13a Salaries information from www.nhscareers.nhs.uk, March 2002. 197 Salaries are based on the midpoint of the relevant scaleprevailing at 1 April 2001, except for GP, medical consultant, secretary and social worker (source: Netten et al., 2001 196 ).All costs are given as 2000/01 values (overheads, capital overheads).b Comprise estimates for indirect overheads (administrative services) (source: Netten et al. 196 ). Indirect overheads for allother staff for which Netten et al. 196 do not provide estimates are assumed to be the same as for staff in the same group.c Based on Netten et al. 196 Capital overheads for all other staff for which Netten et al. 196 do not provide estimates areassumed to be the same as for staff in the same group.d Comprises only salary and salary on-costs.e On-costs are estimated assuming the same on-costs/salary ratio as for staff nurse G grade.fOn-costs are estimated assuming the same on-costs/salary ratio as for physiotherapist senior I.g On-costs are estimated assuming the same on-costs/salary ratio as for occupational therapist senior I/II.MTO, medical technical officer.141© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 17List of equipmentCosts of equipment for outpatient cardiac rehabilitation (phase 3), are shown below. (Source: cardiacrehabilitation programme, Bristol Royal Infirmary.) All prices include VAT.Life Fitness upright cycle LC9100 £2770.65Cardiosport watches ‘Cardiosport Go’ (×15) £375.41Reebok step (×2) £78.00Physio Med Rehab Bouncer £57.57Yellow Theraband exercise roll (50 yards) £38.77York Probells in carry case (×2) £19.90Theraband exercise ball (Antiburst) (45 cm) (×2) £19.97Pro Fitness exercise mats (×15) £89.85Physio Med rehab support rails £46.94Theraband exercise ball (Antiburst) (55 cm) (×2) £27.02Duflex deluxe Gym Mate £82.19–––––––––––––––––––––––––––––––Total: £3606.27 (VAT included)143© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 18Staff input: average hours per weekTABLE 42 Average hours per week by staff category for centres in group 1 (more than five key staff)Staff category Centre Centre Centre Centre Centre Centre Centre Centre Centre Centre1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10Nursing gradeBEF 20.0 37.5 18.5 37.5G 38.0 37.5 37.5 26.0 37.5 37.5H 37.5 37.5 18.0 37.5PhysiotherapistHelperBasic 2.0Senior I 9.0 4.0 10.0 49.5 3.0 18.0 7.0Senior IISuperintendent III 26.0Superintendent IV 6.0Sport scientistExercise physiologist 37.5 0.5Occupational therapistBasic 1.0 7.0Senior 7.0 6.0 8.0 2.0Head 6.0Dietician 4.0 0.32 0.33Senior dietitian 0.33 0.16 1.0 0.2 3.0 1.0 0.125Pharmacist 0.5 0.16 1.0 0.16 1.0 0.12 0.16 1.0 0.125 0.33Physician a 0.15 0.2 0.41 3.0 0 bClinical psychologist 0.75 4.0 0.16 0 b 0.5 1.0 0.33Cardiac technician 3.0 4.0Social worker 1.0Secretary 25.0 2.0 10.0 5.0Total 70.4 122.8 83.5 67.2 95.0 67.6 39.5 103.0 41.8 56.5a Includes cardiologists, consultants, clinical assistants and GPs.b Centres indicated the contribution of these staff groups in their questionnaire, but stated that their contribution was ratheradvisory and did not include any time involvement.145© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 18TABLE 43 Average hours per week by staff category for centres in group 2 (three to five key staff)Staff categoryCentre Centre Centre Centre Centre Centre Centre Centre Centre Centre2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10Nursing gradesB 9.5E 15.0 37.5 19.5F 22.5 75.0 30.0G 75.0 10.0 2.0 37.5 37.0H 37.5 20.0PhysiotherapistHelper 6.0 14.0BasicSenior I 10.0 4.0 11.0 6.0 2.0 21.0Senior II 2.0 12.0Superintendent IIISuperintendent IVSport scientist 6.0 13.0Exercise physiologistOccupational therapistBasicSenior 0.5HeadDietitian 0.25 0.25 0.15 0.16 1.0Senior dietitian 3.0 0.62 1.0Pharmacist 0.25 1.0 0.25 0.5 0.32 0.04 1.0 0.16Physician a 0.5 7.01Clinical psychologist 3.0Cardiac technician 4.0Social worker 16.0Secretary 12.0 0.35 30.0 8.0Total 33.0 90.0 123.0 23.5 5.3 39.7 72.0 75.3 64.0 100.5a Includes cardiologists, consultants, clinical assistants and GPs.146

Health Technology Assessment 2004; Vol. 8: No. 41TABLE 44 Average hours per week by staff category for centres in group 3 (two or fewer key staff)Staff category Centre Centre Centre Centre Centre Centre Centre Centre Centre Centre3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10Nursing gradeBEF 30.0G 4.0 52.5 21.0 2.0 12.0 1.0H 4.0PhysiotherapistHelperBasicSenior I 14.0 6.0 2.0 1.0 4.0Senior II 4.0 2.0 6.0Superintendent IIISuperintendent IVSport scientist 4.0Exercise physiologistOccupational therapistBasicSenior 1.0 2.0HeadDietitianSenior dietitianPharmacistPhysician aClinical psychologistCardiac technician 5.0Social workerSecretary 3.0Total 18.0 53.5 32.0 6.0 6.0 21.0 4.0 1.0 1.0 38.0a Includes cardiologists, consultants, clinical assistants and GPs.147© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 19Referral, uptake and completion rates for30 randomly selected UK cardiac rehabilitationprogrammes in 2000Group 1 a Group 2 b Group 3 c(n = 10) (n = 10) (n = 10)Hours per patient 29.0 (8.9) 24.0 (8.6) 20.0 (12.3)27.0 21.5 17.5(21–48) (15–38) (6–48)Patients referred to CR 282.4 (169.5) 352.5 (244.5) 170.7 (142.5)289.5 255.5 150(84–578) (130–855) (3–400)Patients joined CR 157.3 (97.6) 194.3 (104.7) 97.9 (73.3)148.0 171.5 103.5(46–381) (73–429) (2–216)% of referrals 55.7 55.1 57.351.1 67.1 69.0Patients completed CR 126.3 (90.4) 158.1 (100.8) 88.8 (67.2)104.5 150.5 92.539–319 44–392 2–195% of referrals 44.7 44.8 5236.0 58.9 61.6Data are shown as mean (SD) and median (in italics) (range).a Centres with more than five key staff.b Centres with three to five key staff.c Centres with two or fewer key staff.149© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment 2004; Vol. 8: No. 41Appendix 20Average cost estimates for cardiac rehabilitation(detailed table)151© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Appendix 20152Costs (£) Group 1 a Group 2 b Group 3 c(n = 10) (n = 10) (n = 10)Staff costs only Total costs Staff costs only Total costs Staff costs only Total costsPer year/centre 53,087 (18,251) 72,676 (25,485) 42,125 (24,059) 57,353 (32,146) 12,383 (12,688) 17,638 (16,752)51,323 69,834 43,337 58,362 8,400 12,797(26,658–85,869) (37,183–118,299) (3,607–78,526) (5,932–105,763) (714–38,157) (1,888–50,886)Per patient referred 243 (143) 330 (177) 137 (92) 186 (120) 127 (94) 249 (243)175 245 113 156 105 156(88–441) (123–587) (28–357) (45–470) (10–254) (16–660)Per patient joined 421 (187) 571 (246) 236 (167) 320 (217) 174 (132) 324 (299)466 639 212 293 118 192(159–777) (221–948) (36–651) (59–856) (26–382) (44–944)Per patient completed 542 (225) 738 (298) 317 (228) 429 (295) 186 (133) 344 (295)553 750 259 354 139 230(224–846) (314–1118) (48–716) (79–947) (26–382) (44–944)Per patient completed/hour 20 (9) 27 (12) 14 (11) 20 (14) 14 (15) 30 (41)20 27 11 15 7 10(5–33) (7–45) (1–38) (2–49) (1–40) (1–110)Data are shown as mean (SD) and median (in italics) (range).a Centres with more than five key staff.b Centres with three to five key staff.cCentres with two or fewer key staff.

Health Technology Assessment 2004; Vol. 8: No. 41Health Technology AssessmentProgrammeMembersPrioritisation Strategy GroupChair,Professor Tom Walley,Director, NHS HTA Programme,Department of Pharmacology &Therapeutics,University of LiverpoolProfessor Bruce Campbell,Consultant Vascular & GeneralSurgeon, Royal Devon & ExeterHospitalProfessor Shah Ebrahim,Professor in Epidemiologyof Ageing, University ofBristolDr John Reynolds, ClinicalDirector, Acute GeneralMedicine SDU, RadcliffeHospital, OxfordDr Ron Zimmern, Director,Public Health Genetics Unit,Strangeways ResearchLaboratories, CambridgeMembersHTA Commissioning BoardProgramme Director,Professor Tom Walley,Director, NHS HTA Programme,Department of Pharmacology &Therapeutics,University of LiverpoolChair,Professor Shah Ebrahim,Professor in Epidemiology ofAgeing, Department of SocialMedicine, University of BristolDeputy Chair,Professor Jenny Hewison,Professor of Health CarePsychology, Academic Unit ofPsychiatry and BehaviouralSciences, University of LeedsSchool of MedicineDr Jeffrey AronsonReader in ClinicalPharmacology, Department ofClinical Pharmacology,Radcliffe Infirmary, OxfordProfessor Ann Bowling,Professor of Health ServicesResearch, Primary Care andPopulation Studies,University College LondonProfessor Andrew Bradbury,Professor of Vascular Surgery,Department of Vascular Surgery,Birmingham HeartlandsHospitalProfessor John Brazier, Directorof Health Economics,Sheffield Health EconomicsGroup, School of Health &Related Research,University of SheffieldDr Andrew Briggs, PublicHealth Career Scientist, HealthEconomics Research Centre,University of OxfordProfessor Nicky Cullum,Director of Centre for EvidenceBased Nursing, Department ofHealth Sciences, University ofYorkDr Andrew Farmer, SeniorLecturer in General Practice,Department of Primary HealthCare, University of OxfordProfessor Fiona J Gilbert,Professor of Radiology,Department of Radiology,University of AberdeenProfessor Adrian Grant,Director, Health ServicesResearch Unit, University ofAberdeenProfessor F D Richard Hobbs,Professor of Primary Care &General Practice, Department ofPrimary Care & GeneralPractice, University ofBirminghamProfessor Peter Jones, Head ofDepartment, UniversityDepartment of Psychiatry,University of CambridgeProfessor Sallie Lamb, ResearchProfessor in Physiotherapy/Co-Director, InterdisciplinaryResearch Centre in Health,Coventry UniversityProfessor Julian Little,Professor of Epidemiology,Department of Medicine andTherapeutics, University ofAberdeenProfessor Stuart Logan,Director of Health & SocialCare Research, The PeninsulaMedical School, Universities ofExeter & PlymouthProfessor Tim Peters, Professorof Primary Care Health ServicesResearch, Division of PrimaryHealth Care, University ofBristolProfessor Ian Roberts, Professorof Epidemiology & PublicHealth, Intervention ResearchUnit, London School ofHygiene and Tropical MedicineProfessor Peter Sandercock,Professor of Medical Neurology,Department of ClinicalNeurosciences, University ofEdinburghProfessor Mark Sculpher,Professor of Health Economics,Centre for Health Economics,Institute for Research in theSocial Services, University of YorkProfessor Martin Severs,Professor in Elderly HealthCare, Portsmouth Institute ofMedicineDr Jonathan Shapiro, SeniorFellow, Health ServicesManagement Centre,BirminghamMs Kate Thomas,Deputy Director,Medical Care Research Unit,University of SheffieldProfessor Simon G Thompson,Director, MRC BiostatisticsUnit, Institute of Public Health,CambridgeMs Sue Ziebland,Senior Research Fellow,Cancer Research UK,University of OxfordCurrent and past membership details of all HTA ‘committees’ are available from the HTA website (www.ncchta.org)163© Queen’s Printer and Controller of HMSO 2004. All rights reserved.

Health Technology Assessment ProgrammeMembersDiagnostic Technologies & Screening PanelChair,Dr Ron Zimmern, Director ofthe Public Health Genetics Unit,Strangeways ResearchLaboratories, CambridgeMs Norma Armston,Freelance Consumer Advocate,BoltonProfessor Max BachmannProfessor HealthCare Interfaces,Department of HealthPolicy and Practice,University of East AngliaProfessor Rudy BilousProfessor of Clinical Medicine &Consultant Physician,The Academic Centre,South Tees HospitalsNHS TrustDr Paul Cockcroft,Consultant MedicalMicrobiologist/LaboratoryDirector, Public HealthLaboratory, St Mary’s Hospital,PortsmouthProfessor Adrian K Dixon,Professor of Radiology,Addenbrooke’s Hospital,CambridgeDr David Elliman,Consultant in CommunityChild Health, LondonProfessor Glyn Elwyn,Primary Medical CareResearch Group,Swansea Clinical School,University of WalesSwanseaDr John Fielding,Consultant Radiologist,Radiology Department,Royal Shrewsbury HospitalDr Karen N Foster, ClinicalLecturer, Dept of GeneralPractice & Primary Care,University of AberdeenProfessor Antony J Franks,Deputy Medical Director,The Leeds Teaching HospitalsNHS TrustMr Tam Fry, HonoraryChairman, Child GrowthFoundation, LondonDr Edmund Jessop,Medical Adviser,National SpecialistCommissioning Advisory Group(NSCAG), Department ofHealth, LondonDr Jennifer J Kurinczuk,Consultant ClinicalEpidemiologist,National PerinatalEpidemiology Unit,OxfordDr Susanne M Ludgate, MedicalDirector, Medical DevicesAgency, LondonDr William Rosenberg, SeniorLecturer and Consultant inMedicine, University ofSouthamptonDr Susan Schonfield, CPHMSpecialised ServicesCommissioning, CroydonPrimary Care TrustDr Margaret Somerville,Director of Public Health,Teignbridge Primary Care TrustProfessor Lindsay WilsonTurnbull, Scientific Director,Centre for MR Investigations &YCR Professor of Radiology,University of HullProfessor Martin J Whittle,Head of Division ofReproductive & Child Health,University of BirminghamDr Dennis Wright, ConsultantBiochemist & Clinical Director,Pathology & The KennedyGalton Centre, Northwick Park& St Mark’s Hospitals,HarrowMembersPharmaceuticals PanelChair,Dr John Reynolds, ClinicalDirector, Acute GeneralMedicine SDU, OxfordRadcliffe HospitalProfessor Tony Avery,Professor of Primary HealthCare, University of NottinghamProfessor Stirling Bryan,Professor of Health Economics,Health ServicesManagement Centre,University of BirminghamMr Peter Cardy, ChiefExecutive, Macmillan CancerRelief, LondonDr Christopher Cates, GP andCochrane Editor, Bushey HealthCentreProfessor Imti Choonara,Professor in Child Health,University of Nottingham,Derbyshire Children’s HospitalMr Charles Dobson, SpecialProjects Adviser, Department ofHealthDr Robin Ferner, ConsultantPhysician and Director, WestMidlands Centre for AdverseDrug Reactions, City HospitalNHS Trust, BirminghamMrs Sharon Hart, ManagingEditor, Drug & TherapeuticsBulletin, LondonDr Christine Hine, Consultant inPublic Health Medicine,Bristol South & West PrimaryCare TrustProfessor Stan Kaye,Professor of Medical Oncology,Consultant in MedicalOncology/Drug Development,The Royal Marsden HospitalMs Barbara Meredith,Project Manager ClinicalGuidelines, Patient InvolvementUnit, NICEDr Frances Rotblat, CPMPDelegate, Medicines ControlAgency, LondonProfessor Jan Scott,Professor of PsychologicalTreatments,Institute of Psychiatry,University of LondonMrs Katrina Simister, NewProducts Manager, NationalPrescribing Centre, LiverpoolDr Richard Tiner, MedicalDirector, Association of theBritish Pharmaceutical IndustryDr Helen Williams,Consultant Microbiologist,Norfolk & Norwich UniversityHospital NHS TrustDr Karen A Fitzgerald,Pharmaceutical Adviser, Bro TafHealth Authority, Cardiff164Current and past membership details of all HTA ‘committees’ are available from the HTA website (www.ncchta.org)

Health Technology Assessment 2004; Vol. 8: No. 41MembersTherapeutic Procedures PanelChair,Professor Bruce Campbell,Consultant Vascular andGeneral Surgeon, Royal Devon& Exeter HospitalDr Mahmood Adil, Head ofClinical Support & HealthProtection, Directorate ofHealth and Social Care (North),Department of Health,ManchesterDr Aileen Clarke,Reader in Health ServicesResearch, Public Health &Policy Research Unit,Barts & the London School ofMedicine & Dentistry,Institute of Community HealthSciences, Queen Mary,University of LondonMr Matthew William Cooke,Senior Clinical Lecturer andHonorary Consultant,Emergency Department,University of Warwick, Coventry& Warwickshire NHS Trust,Division of Health in theCommunity, Centre for PrimaryHealth Care Studies, CoventryDr Carl E Counsell, SeniorLecturer in Neurology,University of AberdeenDr Keith Dodd, ConsultantPaediatrician, DerbyshireChildren’s HospitalProfessor Gene Feder, Professorof Primary Care R&D, Barts &the London, Queen Mary’sSchool of Medicine andDentistry, University of LondonProfessor Paul Gregg,Professor of OrthopaedicSurgical Science, Department ofOrthopaedic Surgery,South Tees Hospital NHS TrustMs Bec Hanley, FreelanceConsumer Advocate,HurstpierpointMs Maryann L. Hardy,Lecturer,Division of Radiography,University of BradfordProfessor Alan Horwich,Director of Clinical R&D, TheInstitute of Cancer Research,LondonDr Phillip Leech, PrincipalMedical Officer for PrimaryCare, Department of Health,LondonDr Simon de Lusignan,Senior Lecturer, Primary CareInformatics, Department ofCommunity Health Sciences,St George’s Hospital MedicalSchool, LondonDr Mike McGovern, SeniorMedical Officer, Heart Team,Department of Health, LondonProfessor James Neilson,Professor of Obstetrics andGynaecology, Dept of Obstetricsand Gynaecology,University of Liverpool,Liverpool Women’s HospitalDr John C Pounsford,Consultant Physician, NorthBristol NHS TrustDr Vimal Sharma,Consultant Psychiatrist & HonSnr Lecturer,Mental Health Resource Centre,Victoria Central Hospital,WirrallDr L David Smith, ConsultantCardiologist, Royal Devon &Exeter HospitalProfessor Norman Waugh,Professor of Public Health,University of AberdeenCurrent and past membership details of all HTA ‘committees’ are available from the HTA website (www.ncchta.org)165

Health Technology Assessment ProgrammeMembersExpert Advisory NetworkProfessor Douglas Altman,Director of CSM & CancerResearch UK Med Stat Gp,Centre for Statistics inMedicine, University of Oxford,Institute of Health Sciences,Headington, OxfordProfessor John Bond,Director, Centre for HealthServices Research,University of Newcastle uponTyne, School of Population &Health Sciences,Newcastle upon TyneMr Shaun Brogan,Chief Executive, RidgewayPrimary Care Group, AylesburyMrs Stella Burnside OBE,Chief Executive,Office of the Chief Executive.Trust Headquarters,Altnagelvin Hospitals Health &Social Services Trust,Altnagelvin Area Hospital,LondonderryMs Tracy Bury,Project Manager, WorldConfederation for PhysicalTherapy, LondonMr John A Cairns,Professor of Health Economics,Health Economics ResearchUnit, University of AberdeenProfessor Iain T Cameron,Professor of Obstetrics andGynaecology and Head of theSchool of Medicine,University of SouthamptonDr Christine Clark,Medical Writer & ConsultantPharmacist, RossendaleProfessor Collette Mary Clifford,Professor of Nursing & Head ofResearch, School of HealthSciences, University ofBirmingham, Edgbaston,BirminghamProfessor Barry Cookson,Director,Laboratory of HealthcareAssociated Infection,Health Protection Agency,LondonProfessor Howard Stephen Cuckle,Professor of ReproductiveEpidemiology, Department ofPaediatrics, Obstetrics &Gynaecology, University ofLeedsProfessor Nicky Cullum,Director of Centre for EvidenceBased Nursing, University of YorkDr Katherine Darton,Information Unit, MIND – TheMental Health Charity, LondonProfessor Carol Dezateux,Professor of PaediatricEpidemiology, LondonMr John Dunning,Consultant CardiothoracicSurgeon, CardiothoracicSurgical Unit, PapworthHospital NHS Trust, CambridgeMr Jonothan Earnshaw,Consultant Vascular Surgeon,Gloucestershire Royal Hospital,GloucesterProfessor Martin Eccles,Professor of ClinicalEffectiveness, Centre for HealthServices Research, University ofNewcastle upon TyneProfessor Pam Enderby,Professor of CommunityRehabilitation, Institute ofGeneral Practice and PrimaryCare, University of SheffieldMr Leonard R Fenwick,Chief Executive, Newcastleupon Tyne Hospitals NHS TrustProfessor David Field,Professor of Neonatal Medicine,Child Health, The LeicesterRoyal Infirmary NHS TrustMrs Gillian Fletcher,Antenatal Teacher & Tutor andPresident, National ChildbirthTrust, HenfieldProfessor Jayne Franklyn,Professor of Medicine,Department of Medicine,University of Birmingham,Queen Elizabeth Hospital,Edgbaston, BirminghamMs Grace Gibbs,Deputy Chief Executive,Director for Nursing, Midwifery& Clinical Support Servs,West Middlesex UniversityHospital, IsleworthDr Neville Goodman,Consultant Anaesthetist,Southmead Hospital, BristolProfessor Alastair Gray,Professor of Health Economics,Department of Public Health,University of OxfordProfessor Robert E Hawkins,CRC Professor and Director ofMedical Oncology, Christie CRCResearch Centre, ChristieHospital NHS Trust, ManchesterProfessor F D Richard Hobbs,Professor of Primary Care &General Practice, Department ofPrimary Care & GeneralPractice, University ofBirminghamProfessor Allen Hutchinson,Director of Public Health &Deputy Dean of ScHARR,Department of Public Health,University of SheffieldDr Duncan Keeley,General Practitioner (Dr Burch& Ptnrs), The Health Centre,ThameDr Donna Lamping,Research Degrees ProgrammeDirector & Reader in Psychology,Health Services Research Unit,London School of Hygiene andTropical Medicine, LondonMr George Levvy,Chief Executive, MotorNeurone Disease Association,NorthamptonProfessor James Lindesay,Professor of Psychiatry for theElderly, University of Leicester,Leicester General HospitalProfessor Rajan Madhok,Medical Director & Director ofPublic Health, Directorate ofClinical Strategy & PublicHealth, North & East Yorkshire& Northern Lincolnshire HealthAuthority, YorkProfessor David Mant,Professor of General Practice,Department of Primary Care,University of OxfordProfessor Alexander Markham,Director, Molecular MedicineUnit, St James’s UniversityHospital, LeedsDr Chris McCall,General Practitioner,The Hadleigh Practice,Castle MullenProfessor Alistair McGuire,Professor of Health Economics,London School of EconomicsDr Peter Moore,Freelance Science Writer,AshteadDr Andrew Mortimore,Consultant in Public HealthMedicine, Southampton CityPrimary Care TrustDr Sue Moss,Associate Director, CancerScreening Evaluation Unit,Institute of Cancer Research,SuttonProfessor Jon Nicholl,Director of Medical CareResearch Unit, School of Healthand Related Research,University of SheffieldMrs Julietta Patnick,National Co-ordinator, NHSCancer Screening Programmes,SheffieldProfessor Robert Peveler,Professor of Liaison Psychiatry,University Mental HealthGroup, Royal South HantsHospital, SouthamptonProfessor Chris Price,Visiting Chair – Oxford,Clinical Research, BayerDiagnostics Europe,CirencesterMs Marianne Rigge,Director, College of Health,LondonDr Eamonn Sheridan,Consultant in Clinical Genetics,Genetics Department,St James’s University Hospital,LeedsDr Ken Stein,Senior Clinical Lecturer inPublic Health, Director,Peninsula TechnologyAssessment Group,University of ExeterProfessor Sarah Stewart-Brown,Director HSRU/HonoraryConsultant in PH Medicine,Department of Public Health,University of OxfordProfessor Ala Szczepura,Professor of Health ServiceResearch, Centre for HealthServices Studies, University ofWarwickDr Ross Taylor,Senior Lecturer,Department of General Practiceand Primary Care,University of AberdeenMrs Joan Webster,Consumer member, HTA –Expert Advisory Network166Current and past membership details of all HTA ‘committees’ are available from the HTA website (www.ncchta.org)

FeedbackThe HTA Programme and the authors would like to knowyour views about this report.The Correspondence Page on the HTA website(http://www.ncchta.org) is a convenient way to publishyour comments. If you prefer, you can send your commentsto the address below, telling us whether you would likeus to transfer them to the website.We look forward to hearing from you.The National Coordinating Centre for Health Technology Assessment,Mailpoint 728, Boldrewood,University of Southampton,Southampton, SO16 7PX, UK.Fax: +44 (0) 23 8059 5639 Email: hta@soton.ac.ukhttp://www.ncchta.org ISSN 1366-5278