Kashner 1995.

Methods	Study design: randomised controlled trial
Participants	Diagnosis: somatisation disorder Method of diagnosis: inclusion criteria were diagnosis of somatisation disorder, history of ≥ 13 unexplained symptoms, the first beginning < 30 years of age, severe enough to seek treatment, take medication, or reduce functioning. This was determined by a research psychiatrist, using a checklist of 37 items that made up diagnostic criteria of somatisation disorder, according to DSM‐III‐R criteria Exclusion criteria: the absence of a primary care provider who agreed to allow the research team to see the participant, lack of transportation to the medical centre, indication of moving out of town during study Total number randomised: 70 Age: mean age 44.2 years (no SD provided) Sex: 84% women (n = 59); 16% men (n = 11) Severity of symptoms at baseline: inclusion criterion: ≥ 13 unexplained symptoms Duration of symptoms at baseline: inclusion criterion: 1 symptom starting before age 30 years Setting: recruitment: internists, GPs, and general population (through advertisements in local media). Treatment: at medical centre Location: Central Arkansas, USA Number of treatment centres: 1 Co‐morbidities: unknown Adjunctive therapy: not mentioned Adjunctive medication: not mentioned Co‐intervention: none reported
Interventions	Participants were randomly assigned to either 1. Group therapy intervention + consultation letter (n = 44) Duration: 8 small group sessions of 2 hours every other week, during a 4‐month period Treatment protocol: the overall goals were to develop a source of peer support, share methods of coping with physical problems, enable participants to increase their ability to perceive and express emotion, and enjoy the experience of participating in the group. Treatment in groups of 4‐6, according to a structured protocol, with a class atmosphere, including didactic presentations, small group discussions, therapy exercises, and group discussions. See Kashner 1995 p.464‐5 for details Co‐intervention (both groups): a standard psychiatric consultation letter was sent to the primary care physician, diagnosing the participant with somatisation disorder and including recommendations for its management Therapist: master's level clinicians 2. Consultation letter only (n = 26) Co‐intervention (both groups): standard psychiatric consultation letter sent to primary care physician, diagnosing the participant with somatisation disorder and including recommendations for its management Duration: NA Treatment protocol: NA Therapist: NA
Outcomes	Time points for assessment: baseline and 4 months, 8 months and 12 months after baseline Outcomes: (unclear what is primary and what is secondary) 1. RAND Health Status (4 domains) 2. days in bed (count) 3. healthcare utilisation (costs)
Notes	Date of study: started in 1987, end unclear (healthcare costs were adjusted for inflation using 1990 dollar rates) Funding sources: Robert Wood Johnson Foundation, National Institute of Mental Health, VA Health Services Research and Development Program for Mental Health Declarations of interest among the primary researchers: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and personnel not blinded to intervention
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcome assessors did not know whether participant was in experimental or control group, but most outcomes were participant reported and these were not blinded
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants completed study
Selective reporting (reporting bias)	High risk	Days in bed described as assessed but no results reported
Treatment fidelity	Low risk	Treatment was based on a structured protocol developed before the study (available from the authors)
Researcher allegiance	Low risk	No indication that researchers had a preference for 1 of the treatment modalities
Other bias	Low risk	No other sources of bias