Furlan · Pain research & management 2011 · systematic review and meta-regression of randomized controlled trials · n=62 trials

A comparison between enriched and nonenriched enrollment randomized withdrawal trials of opioids for chronic noncancer pain.

Cited 124 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis of randomized controlled trials

PubMed 22059206 · doi:10.1155/2011/465281 · record verified 2026-08-26

What was done

A systematic review and meta-regression of randomized controlled trials was conducted to evaluate whether enriched enrollment randomized withdrawal (EERW) designs alter efficacy and safety estimates of opioids for chronic noncancer pain compared to non-EERW designs. Searches in MEDLINE, EMBASE, and CENTRAL through July 2009 identified eligible randomized trials. Efficacy outcomes (pain and function) were evaluated using effect sizes, and adverse events were assessed using a threshold for clinical relevance of a 10% or greater mean difference.

What was found

Across 62 included randomized trials, 61 had trial durations under 16 weeks. Efficacy did not differ significantly between EERW and non-EERW trials for pain (P=0.6) or function (P=0.3). However, EERW trials failed to detect clinically relevant differences for nausea, vomiting, somnolence, dizziness, and dry skin/itching relative to non-EERW trials. Overall, opioids were more effective than placebo for nociceptive pain (effect size = 0.60, 95% CI 0.49 to 0.72) and neuropathic pain (effect size = 0.56, 95% CI 0.38 to 0.73).

Why it matters

These findings indicate that while EERW trial designs do not introduce meaningful bias into efficacy estimates, they systematically underestimate the burden of common opioid-related adverse effects.

Limits

Almost all included studies (61 of 62) lasted less than 16 weeks, limiting evidence on long-term safety, tolerance, and efficacy. The total number of randomized participants across trials was not reported in the abstract, and individual drug-level comparisons were limited.

Cited by