Barsky 2004

Methods	Participant blinding: no Therapist blinding: no Assessor blinding: yes Analysis: intention‐to‐treat Country: USA Ethics approval: not stated
Participants	Number: active ‐ 102 (76 female; 26 male); control ‐ 85 (67 female; 18 male); total ‐ 187 (143 female; 44 male) Mean (sd) age in years: active ‐ 40.66 (12.60); control ‐ 44.29 (13.75); total ‐ 42.31 Diagnostic criteria: Whitely Index & Somatic Symptom Inventory with cutoff score of 150; 61% of participants had a DSM‐IV diagnosis of hypochondriasis Recruited from: primary care; volunteers Mean (sd) length of illness: active ‐ (); control ‐ (); total ‐ 11 years () Exclusion criteria: 1. <18 years of age; 2. not fluent or literate in English; 3. not having seen GP in last 12 months; 4. major medical morbidity; 5. somatoform pain disorder; 6. psychosis; 7. suicide risk; 8. ongoing disability or compensation claim of litigation. Language restrictions: English only Comorbidity: not stated Number of drop‐outs: active ‐ 10; control ‐ 7; total ‐ 17
Interventions	Active treatment: cognitive behaviour therapy + letter to GP Control treatment: "usual medical care" Treatment setting: Primary care ‐ letter sent to GPs advising on management, Secondary care outpatient Discipline of therapists: "masters or doctoral degrees and prior CBT experience" Number and duration of sessions: 6 x 90min Interval between sessions: 1 week Total exposure: 9 hours
Outcomes	Length of follow up: 12 months Primary outcome measure: Whitely Index score at 12 months Secondary outcome measures: Health Anxiety Inventory Hypochondriacal Cognitions Questionnaire Somatic Symptom Inventory SCID for DSM‐IV hypochondriasis Somstosensory Amplification Scale Functional Status Questionnaire ‐ intermediate activities of daily living subscale Functional Status Questionnaire ‐ social activities subscale Covariates: Hodgkins Symptom Checklist 90 Duke Severity of Illness Scale 2 ordinal ratings of aggregate medical morbidity by primary care physician
Notes	possible selection bias ‐ only 30% of subjects participated likely performance bias ‐ GPs in active treatment group were given advice on management attrition bias unlikely detection bias unlikely Overall quality assessment: moderate (QRS score = 30)
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear