Falcao 2008.

Methods	Study setting: South America (Brazil). Single centre, recruitment in rheumatology university clinic, outpatient based Study design: Parallel Duration therapy: 10 weeks Follow‐up: 3 months Analysis: Chi2 or student t‐test analyses between groups, analyses of variance (repeated measures) between groups at different time points
Participants	Patients: Treatment Group: 30 patients, 100% female, 80% Caucasian, mean age 45 years; disease duration 3.5 (2.4) years Control Group: 30 patients, 100% female, 77% Caucasian, mean age 46 years; disease duration 3.7 (4.8) years Inclusion: ACR 1990 criteria for FM, age 18 to 65 years, female, at least 4 years of formal education (elementary school), patients had not received any kind of treatment for their disease Exclusion: Other rheumatic diseases, known hypersensibility to amitriptyline, cyclobenzaprine or paracetamol, use of psychotropic drugs, psychiatric diseases, work‐related litigation
Interventions	Treatment Group: CBT,group: progressive muscle relaxation training, cognitive restructuring, stress management + routine medical visits (3h/week), total: 30h Control Group: TAU: Medication and routine medical visits Co‐medication allowed: Patients in both groups: amitriptyline 12.5mg/day during first week, then increase to 25mg/day, those with intolerance or side effects were given cyclobenzaprine 5mg/day, use of paracetamol was allowed 750mg if patients had pain (max. dose of 2250mg/day) Other Co‐therapies: None
Outcomes	Primary Outcomes Self reported pain: VAS pain 0‐10 Self reported negative mood: Beck Depression Inventory (BDI) 0‐54 Self reported disability: FIQ 0‐10; data provided on request Acceptability: Total dropout rate Secondary Outcomes Self reported self efficacy pain: Not assessed Self reported fatigue: FIQ fatigue 0‐10; data provided on request Self reported sleep problems: Not assessed Self reported disease‐specific health‐related quality of life: FIQ total 0‐100
Notes	1. Reasons for dropout: ‐ Experimental group:2x move to another city, 3x gave up after 1st, 5th or 6th CBT session ‐ Control group: 1x use of psychotropic drugs, 3x gave up study at different times 2: Attendance rates: All completers in the CBT group attended more than 80% of the sessions. 3. Responder analysis: None 4. Funding sources and declaration of interest of the primary researchers: No details reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Patients were randomised by drawing lots with concealed allocation
Allocation concealment (selection bias)	Low risk	Yes (see above)
Incomplete outcome data (attrition bias) All outcomes	High risk	No intention‐to‐treat analysis
Selective reporting (reporting bias)	Low risk	All outcomes reported or provided on request
Blinding of outcome assessor	Low risk	Evaluation performed by a physiotherapist who was blind to treatment allocation