Wang 2012.

Study characteristics
Methods	Study design: Multicentre RCT (2 sites) Country: China Dates participants recruited: Oct 2005 to April 2007 Maximum follow‐up: 6 months
Participants	Inclusion criteria: Inclusion criteria comprised a documented diagnosis of acute MI, the ability to speak and read Chinese, a return to living at home after hospital discharge, availability for telephone follow‐up, and availability for meetings after hospital discharge. Exclusion criteria: Exclusion criteria comprised a known history of major psychiatric illness, pre‐existing mobility problems, unstable angina, severe complications such as uncontrolled arrhythmias or heart failure, and other conditions that could be aggravated by exercise, such as a resting systolic blood pressure (BP) > 200 mmHg or a resting diastolic BP > 110 mmHg. N randomised: total: 160; intervention: 80; comparator: 80 Diagnosis (% of participants): AMI: 100% Age (mean± SD): intervention: 57.3 (± 8.6); comparator: 58.3 (± 10.4) Percentage male: intervention: 85.3%; comparator: 81.5% Ethnicity: NR
Interventions	Intervention: A 6‐week, home‐based rehabilitation programme using a self‐help heart manual given to the rehab participants just before discharge from hospital. The manual was similar to the UK Heart Manual but incorporated appropriate sociocultural components such as tai chi, qi gong, and Chinese diet. Section 1 consists of 6 weekly topics on health education. Section 2 answers commonly asked questions about medication, PCI, anxiety and depression etc. Section 3 presents information on normal values of cardiac physiological risk parameters. The rehabilitation group received the manual and the introductory session in addition to usual care. The exercise component of the manual is not described in this paper, and there is no reference to its description elsewhere. Components: exercise plus education. Setting: home. Exercise programme modality: not described. Length of session: not described. Frequency: not described. Intensity: not described. Resistance training included? not described. Total duration: not described. Co‐interventions: participants in both groups were telephoned by the principal researcher 3 weeks after discharge. For the rehabilitation group, the researcher checked the participants' progress, encouraged adherence to exercise, and helped solve problems that had arisen using the manual. This consultation lasted approximately 30 minutes, with contact designed to promote participant confidence and self‐management, and minimise dependency and the possibility that the nurse could influence outcomes. Comparator: The usual care group received instructions on taking medications, information leaflets about cardiac risk factors, a healthy diet, and smoking cessation, and a follow‐up appointment. Co‐interventions: The researcher devoted an equal amount of time to telephone contact with the control group, giving general advice on any problems encountered and encouraging and supporting appropriate actions.
Outcomes	Mortality, HRQoL
Source of funding	NR
Conflicts of interest	NR
Notes	Baseline characteristics only reported for those followed up until 6 months i.e. 68 in intervention group and 65 in usual care group.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	“Patients …….were enrolled and assigned to either the experimental or the control group, using a computer‐generated random number”.
Allocation concealment (selection bias)	Unclear risk	Allocation concealment was not described.
Blinding of outcome assessment (detection bias) All outcomes	High risk	“the absence of a blinded condition may threaten its internal validity. In addition, the principal researcher played the role of both intervener and outcome assessor, which may have influenced participants to provide desired answers, and so interviewer bias cannot be excluded”
Incomplete outcome data (attrition bias) All outcomes	Low risk	12/80 (15%) lost from intervention group. 15/80 (18.8%) lost from the control group. Numbers and reasons were given and were similar for both groups.
Selective reporting (reporting bias)	Low risk	All outcomes described were reported for all time points.