Bolton 2003.

Study characteristics
Methods	Study design: cluster‐randomised parallel‐group gender‐stratified controlled clinical trial (unit of randomisation: village; unit of analysis: individual. 15 villages in each arm) Duration of study: February 2002 to July 2002; 6‐month follow‐up completed in January 2003
Participants	Country: Uganda Income classification: low income Geographical scope: 30 villages in Rakkai Province and contiguous half of Masaka Province in South West Uganda; rural Healthcare setting: community (community centres, churches, open spaces) Mental health condition: depression (DSM‐IV depression and sub‐syndromal depression) Population: patients 1. Age: adults (> 18 years); mean age ranged from 27 years (SD 13.5) to 66 years (SD 10.5) 2. Gender: both (stratified for gender) 3. Socioeconomic background: not stated except for education (mean 4.7 years (SD 2.8) Intervention; 3.9 years (SD 3.3) control) 4. Inclusion criteria: 3‐stage screening: Stage 1 (by trained local World Vision staff) identified 20 people from the selected 15 villages (8 for males and 5 for females) with depressive symptoms in local idiom; Stage 2: same interviews visited identified people, and if they admitted to having 1 of 2 locally approximate depressive conditions, informed consent was sought; Stage 3: eligibility expanded to include sub‐syndromal depression by DSM‐IV criteria (less 1 DSM criterion); screening for depression was done by 10 trained and experienced local World Vision staff using a composite instrument (Bolton 2004) consisting of HSCL (to assess depressive symptoms and to diagnose DSM‐IV major depression (excluding criteria related to exclusion of medical causes and drug effects), a previously validated algorithm), a locally developed culturally appropriate instrument to assess functional impairment (separately for women and for men), and ethnographically validated questions that assessed significant distress and duration of depression 5. Exclusion criteria a. Absence of symptoms of depression b. Age < 18 years c. Unwillingness to meet weekly (additional criteria revised after screening commenced) d. People very different in age from the rest included in a village e. Those appearing currently suicidal
Interventions	Stated purpose: to test the efficacy of a manual‐based, time‐limited group psychotherapeutic approach in relieving depressive symptoms and improving functioning; and to demonstrate that psychotherapy trials are feasible in sub‐Saharan Africa INTERVENTION (n = 107) Name: Group Interpersonal Therapy for Uganda (IPT‐G‐U), 116 people (of 163 in 15 villages originally randomised, and 139 invited to participate; 107 completed intervention and follow‐up) Delivered by: LHWs Title/name of PW and number: group leaders ‐ 9/10 who completed training 1. Selection: local person of the same sex as the sex‐segregated group; non‐clinicians fluent in English and Luganda employed by World Vision 2. Educational background: completed high school (college‐level) 3. Training: duration: 2 weeks intensive training. Trained by 2 faculty members of the New York State Psychiatric Institute (members of the team led by Myrna Weissman that developed IPT and the group adaptation of IPT) assisted by a trained psychologist and an experienced group therapist employed by World Vision. Content of training: participating in local adaptations of the IPT manual; explanations of treatment process and contract; explanations of the role of group leaders in helping members to identify problem areas and in discussing locally acceptable variations to absolute confidentiality; identification of, and agreement about, interpersonal problem areas likely to be encountered in group work according to the 4 domains in IPT; using the principles of IPT to identify personal problems and to support one other to find options and to facilitate implementation. Format: didactic teaching and experiential group processes with role‐plays and group exercises 4. Supervision: by local World Vision mental health professionals involved in training; format and duration not described 5. Incentives/remuneration: weekly payment for 16 weeks (amount not stated) Intervention details 1. Duration/frequency: 16 weekly 90‐minute sessions 2. Content of intervention: group work led by group leader who first diagnoses depression; works with group member to identify problem areas associated with current symptoms and to identify the 4 areas of interpersonal difficulties that served as triggers for depression; conducts weekly review of mood and encouragement of participant's description of events that could link to the mood; and facilitates support and solutions from group members CONTROL (n = 117) Treatment as usual (treatment by local traditional healers, no treatment, or, in rare cases, hospitalisation), 138 people (of 178 randomised in 15 villages and 145 invited to participate; 117 completed intervention and follow‐up) CO‐INTERVENTIONS: no restrictions on additional interventions (utilisation and nature of any not described)
Outcomes	Patients 1. Screening: HSCL and local functional impairment scale 2. Prevalence of DSM‐IV major depression (excluding criteria related to exclusion of medical causes and drug effects ‐ using Mollica DSM‐IV algorithm for A, C, and E criteria)* 3. HSCL mean scores 4. Functional Impairment scores (sex‐specific 9‐item questionnaire) 5. Depression in subgroups continuing informal group meetings between 2 weeks and 6 months vs subgroup not meeting after group intervention Carers Not applicable Process/health workers No direct outcomes reported: indirect outcomes are the results of the trial Economic outcomes Not reported Time points: initial assessment 2 weeks after intervention, follow‐up at 6 months (*: primary outcomes)
Notes	Source of funding: supported by World Vision, Washington, DC; Psychotherapy Core of the Child Intervention Research Center Columbia University (NIMH grant #5P30 MH60570); Center for International Emergency Disaster and Refugee Studies; Johns Hopkins Bloomberg School of Public Health; Mellon Foundation Notes on validation of instruments (screening and outcomes): all screening instruments and outcome measures locally adapted and validated in previous exercises and published; the HSCL scale consisted of 14 items, with 4 responses for each item related to the degree of distress due to a particular symptom (range 0 to 42 points); higher scores indicate more severe depression; function scale consisted of 9 items, with 5 responses for each item, indicating degree of difficulty in completing the activity (range 0 to 36 points); higher scores indicate more dysfunction Additional information: IPT attendance was high: 54% attended at least 14/16 sessions; 4% attended ≤ 10 sessions. Declarations of interest ‐ none Handling the data: as per footnotes in data and analysis Prospective trial registration number: not prospectively registered
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from report: "random assignment was performed by enumerating the villages and using a random number table to determine study allocation" Comment: cluster‐randomisation of 30 villages to 15 in each arm was done using a random numbers table; the 30 villages of 154 eligible villages were chosen for a previous prevalence study ‐ Bolton 2002, unpublished ‐ that used weighted random sampling based on government census data
Allocation concealment (selection bias)	Unclear risk	Quote from report: "each list began with those who met the original diagnostic criteria, followed by those who fell short by a single criterion, in order of decreasing depression score. Interviewers visited all persons in the order they appeared on the list. The interviewer re‐read the consent form, advised persons about the study group to which their village had been allocated, and asked them to confirm their willingness to continue in the study. Interviewers continued down the list until they had at least 8 participants (at which point they did not contact the remainder of the list) or until they reached the end of the list" Comment: allocation of participants was not concealed although cluster‐randomisation of villages was the unit of randomisation; eligibility criteria were modified to exclude people whose age varied widely from the rest of those selected in each village, to ensure better outcomes with group IPT (based on previous experiences)
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Comment: participants and personnel were not blinded; however, cluster‐randomisation would ensure minimal risk of performance bias because villages where intervention was given were separate from villages randomised to usual care
Blinding of outcome assessment (detection bias) all outcomes	Low risk	Quote from report: "the baseline assessments were conducted in the villages, with the randomisation of village groups to intervention or control or control status done afterwards to ensure that interviewers were not aware of participant trial status at baseline. In an effort to keep interviewers unaware of the participants’ intervention status, the post‐intervention and 6‐month follow‐up assessments were conducted at a centrally located community centre. At these assessments, trial participants were transferred from their villages and were asked not to divulge either their village of origin or their treatment assignment status. To reduce measurement error that might have arisen from different interviewing styles, study participants were interviewed by the same interviewer at each stage of the study" Comment: the period from recruitment to first assessment was 18 weeks and to second assessment was a further 6 months. There is a possibility that the recruiter (who did not administer the intervention) may have guessed allocation for the first assessment in a few instances, but this is unlikely to have altered results significantly, given the magnitude of the differences in results between groups
Baseline outcome measurements similar	Low risk	Quote from 6‐month follow‐up report: "at baseline 86% of participants in the intervention group met the modified diagnostic criteria for major depressive disorder and 94% of those in the control group met these criteria (prevalence difference was not significant)" Quote from primary report: "however, there was a significant difference in the proportions who met the original depression diagnostic criteria, both among those who completed the study and all those on the original lists of eligible participants (TABLE 1). (Tests for differences in baseline characteristics were performed using standard significance tests and were not adjusted for cluster effects. However, because we found a positive correlation between clusters, adjusting for cluster effects would tend to reduce variance and cause group differences to be even less significant than the values reported herein)" Comment: discrepancy in interpretation of baseline differences in the 2 reports; results adjusted for clustering and for baseline outcome differences
Baseline characteristics similar?	Low risk	Comment: no differences in age or education nor in symptom duration
Incomplete outcome data (attrition bias) Efficacy data	Low risk	Quote from 6‐month follow‐up report: "six months after the post‐intervention 103 (96%) of the 107 participants in the intervention group who completed the trial and 113 (97%) of the 117 completed the trial and 113 (97%) of the 117 controls were reassessed" Comment: attrition was high in both groups due to the 3‐stage screening process; however, 116/163 eligible and randomised to IPT consented to participate; 132/178 randomised to control consented to participate; results did not differ in completer analyses and in 2 sets of ITT analyses
Protection against contamination	Low risk	Comment: cluster‐randomisation of intervention arms precluded contamination
Selective reporting (reporting bias)	Unclear risk	Comment: trial was not prospectively registered, but all pre‐stated outcomes were reported
Other bias	Low risk	Comment: no other biases were detected