Crofford 2008.

Methods	26‐week randomised, double‐blind, placebo‐controlled, parallel‐group, EERW trial. Participants initially screened for response (≥ 50% decrease in pain and PGIC of much or very much improved). Responders to initial titration selected for randomisation to placebo or continued use of tolerated effective dose
Participants	Fibromyalgia according to ACR classification and pain of at least 40/100 mm in week before randomisation, with 6 months of follow‐up N = 1051 entered open‐label phase (6 weeks); 566 randomised to double‐blind phase (26 weeks) Mean age 49 years, 93.5% female, 90% white Baseline mean pain: 78/100
Interventions	Pregabalin titrated to a maximum of 600 mg daily, n = 279 (300 mg = 63; 450 mg = 73; 600 mg = 143) Placebo daily, n = 287
Outcomes	Loss of therapeutic response (worsening of pain or other symptoms, pain reduction less than 30% of baseline on several occasions, withdrawal) measured in days, converted to maintenance of therapeutic response for analysis Adverse events Withdrawals
Notes	Pfizer Global Research and Development sponsored Oxford Quality Score: R2, DB2, W1. Total = 5/5
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"A tele‐randomisation system randomised responders to either matching placebo or optimal open label dosage of pregabalin (1:1)" Review authors judged this low risk, assuming computer generation
Allocation concealment (selection bias)	Low risk	Tele‐randomisation system
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"matching placebo or optimal open‐label dosage of pregabalin"
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"matching placebo or optimal open‐label dosage of pregabalin"
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants included in analysis
Selective reporting (reporting bias)	Low risk	All relevant outcomes reported
Size	Low risk	> 200 participants per treatment arm