Li 2002.

Study characteristics
Methods	Study design: RCT  Duration of study; 9 months. Recruitment between June 1999 and March 2001. Intervention lasted 50 to 51 days. Follow‐up assessments 6 months post intervention
Participants	Country: China  Income classification: lower‐middle income between 1999 and 2002   Geographical scope: rural  Healthcare setting: home and community  Mental health condition: schizophrenia (specifically late‐onset)  Population (mention if patient, carer, or dyad)  1. Age: 50 years and older  2. Gender: both  3. Socioeconomic background: over 80% were farmers  4. Inclusion criteria  a. Met Chinese Classification of Mental Disorder (CCMD)‐2‐R criteria for schizophrenia b. Age 50 and above c. First episode of illness  5. Exclusion criteria: nil
Interventions	Stated purpose: to determine the efficacy of community rehabilitation for patients with late‐onset schizophrenia  INTERVENTION (n = 38)  Name: community rehabilitation  Delivered by: PHP Title/name of PW and number: village and street‐level doctors ‐ number not specified   1. Selection: existing infrastructure  2. Educational background: medical degree  3. Training (contents, duration, by whom): briefed by a hospital community medicine doctor regarding general mental health prevention and treatment and specific issues to take note of    4. Supervision: hospital community medicine doctor in charge of patient  Intervention details: home visits and telephone enquiries  1. Duration/frequency: duration 50.12 ± 28.19 days, frequency of home visits at least weekly, telephone enquiries as needed   2. Content of intervention (by types of health workers and per patients/carers): medication following specific guidelines (see below), counselling, requested work, social activities  CONTROL (n = 38) Specialist care ‐ inpatient care; doctors in charge were psychiatrists, treated mainly with medication according to guidelines (same as intervention group), supplemented by occupational therapy and psychotherapy. Duration 51.03 ± 29.17 days  CO‐INTERVENTIONS: medication guidelines were to begin with a low dose of a typical antipsychotic such as perphenazine, gradually increased as needed up to 29.05 ± 9.83 mg/d, to add on sulpiride up to 200mg/d for depressed mood, apathy, or withdrawal, and to add on alprazolam if there was obvious insomnia
Outcomes	Patients  1. Recovery (BPRS reduced by ≥ 80%)  2. BPRS  3. SDSS  4. Relapse (reappearance of symptoms or worsening of original symptoms requiring medication modification) 5. Disease much improved (BPRS reduced by ≥ 60%)# 6.Disease improved (BPRS reduced by ≥ 30%)#  7. Disease showing no improvement (BPRS reduction <3 0%)#  Carers  Nil  Process/health workers Nil  Economic outcomes  Direct costs of treatment (page 2 of article (page 712 of journal), right column, paragraph numbered “4”)  (asterisk for study's primary outcomes; star: outcomes that we have not reported in this review)  Time points: 0 months, 6 months
Notes	Article in Chinese Source of funding: none  Notes on validation of instruments (screening and outcomes): validated  Additional information: declarations of interest ‐ none Handling the data: none  Prospective trial registration number: none
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Judgement comment: randomised based on sequence of admission to hospital
Allocation concealment (selection bias)	High risk	Judgement comment: no description of any efforts to conceal allocation
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No effort to blind participants was mentioned in the paper
Blinding of outcome assessment (detection bias) all outcomes	High risk	No attempt at blinding researchers was mentioned in the paper
Baseline outcome measurements similar	Low risk	Baseline outcome measures (table 2 ‐ BPRS, table 3 ‐ SDSS) were similar in both groups
Baseline characteristics similar?	Low risk	Baseline characteristics (age, gender, occupation, education, marital status, family history) were similar in both arms
Incomplete outcome data (attrition bias) Efficacy data	Low risk	There is no attrition in this study
Protection against contamination	Low risk	Judgement comment: groups were treated in different settings (hospital, community); therefore unlikely to contaminate each other
Selective reporting (reporting bias)	Unclear risk	All outcomes planned in materials and methods section were reported in results section. No clinical trial protocol is available
Other bias	Low risk	No other sources of bias have been found