The Birmingham Rehabilitation Uptake Maximisation Study (BRUM). Home-based compared with hospital-based cardiac rehabilitation in a multi-ethnic population: cost-effectiveness and patient adherence.

OBJECTIVES: To evaluate the relative effectiveness and cost-effectiveness of a home-based programme of cardiac rehabilitation using the Heart Manual, with centre-based programmes. It also sought to explore the reasons for non-adherence to cardiac rehabilitation programmes.
DESIGN: An individually randomised trial, with minimisation for age, gender, ethnicity, initial diagnosis and hospital of recruitment. Participants were followed up after 6, 12 and 24 months by questionnaire and clinical assessment. Individual semistructured interviews were undertaken in the homes of a purposive sample of patients who did not adhere to their allocated programme, and focus groups were undertaken with groups of patients who adhered to the programmes.
SETTING: Four hospitals in predominantly inner-city, multi-ethnic, socio-economically deprived areas of the West Midlands in England, for 2 years from 1 February 2002. PARTICIPANTS: A total of 525 patients who had experienced a myocardial infarction (MI) or coronary revascularisation within the previous 12 weeks.
INTERVENTIONS: All the rehabilitation programmes included exercise, relaxation, education and lifestyle counselling. All patients were seen by a cardiac rehabilitation nurse prior to hospital discharge and provided with information about their condition and counselling about risk factor modification. The four centre-based programmes varied in length from nine sessions at weekly intervals of education, relaxation and circuit training to 24 individualised sessions over 12 weeks of mainly walking, fixed cycling and rowing with group-based education. The home-based programme consisted of an appropriate version of the Heart Manual, home visits and telephone contact. The Heart Manual was introduced to patients on an individual basis, either in hospital or on a home visit. Home visits by a nurse took place at approximately 1, 6 and 12 weeks after recruitment, with a telephone call at 3 weeks. At the final visit, patients were encouraged to maintain their lifestyle changes and to continue with their exercise programme. Where needed, follow-up was made by a rehabilitation nurse who spoke Punjabi. An audiotape of an abridged version of the Heart Manual in Punjabi accompanied the manual for patients with a limited command of English. MAIN OUTCOME MEASURES: Primary outcomes were smoking cessation, blood pressure, total and high-density lipoprotein cholesterol, exercise capacity measured by the incremental shuttle walking test and psychological status measured by the Hospital Anxiety and Depression Scale (HADS). Secondary outcomes included self-reported diet, physical activity, cardiac symptoms and quality of life. Health service resource use and costs of rehabilitation programmes from health service and societal perspectives were also measured. Adherence to the physical activity element of the rehabilitation programmes was measured by questionnaire 6, 9 and 12 weeks.
RESULTS: No clinically or statistically significant differences were found in any of the primary or secondary outcome measures between the home- and centre-based groups. Significant improvements in total cholesterol, smoking prevalence, the HADS anxiety score, self-reported physical activity and diet were seen in both arms between baseline and the 6-month follow-up. Five or more contacts with a cardiac rehabilitation nurse were received by 96% of home-based participants, whereas only 56% of centre-based participants attended this many rehabilitation classes. The direct rehabilitation costs to the health service were significantly higher for the home-based programme (mean cost 198 pounds versus 157 pounds for the centre-based programme), but when patient costs were included the mean cost of the centre-based arm rose to 182 pounds. Patients' reasons for not taking up or adhering to cardiac rehabilitation were multifactorial and very individual. Other health problems limited some patients' ability to exercise. Most non-adherers found some aspects of their cardiac rehabilitation programme helpful. Many had adapted advice on rehabilitation and were continuing to exercise in other ways and had made lifestyle changes, particularly to their diet. The home-based patients' lack of motivation to exercise on their own at home was a major factor in non-adherence. The focus groups revealed little diversity of views among patients from each programme. Patients in the hospital programme enjoyed the camaraderie of group exercise and the home-based patients valued the wealth of information and advice in the Heart Manual.
CONCLUSIONS: A home-based cardiac rehabilitation programme for low- to moderate-risk patients does not produce inferior outcomes compared with the traditional centre-based programmes. With the level of home visiting in this trial, the home-based programme was more costly to the health service, but with the difference in costs borne by patients attending centre-based programmes. Different reasons were given by home and hospital cardiac rehabilitation patients for not taking up or adhering to cardiac rehabilitation, with home-based patients often citing a lack of motivation to exercise at home. Social characteristics, individual patient needs and the location of cardiac rehabilitation programmes need to be taken into account in programme design to maximise participation. Research is recommended into cardiac rehabilitation in patients from ethnic minority groups; measurement tools to assess physical activity and dietary change; evaluating the Heart Manual in patients who decline centre-based cardiac rehabilitation; the implementation of home-based programmes in the UK; and strategies that sustain physical activity in the long term.