The Impact of Outcome Measures for Responder Analyses in Placebo-Controlled Trials of Pharmacological Treatments for Neuropathic Pain.
Nanna Brix Finnerup, Sandra Sif Gylfadottir, Xavier Moisset et al.
✦ AI-curated · Sources linked
In 30 seconds
This meta-analysis examined how dichotomous responder definitions affect estimated treatment effects in placebo-controlled pharmacological trials for neuropathic pain. Secondary analyses covered 91 double-blind randomised trials selected from 222 trials; included studies reported at least two responder outcomes. Across 76 trials, the risk difference was higher for ≥ 30% than ≥ 50% pain reduction (difference 0.023, 95% CI 0.012-0.034; p < 0.001), with corresponding NNTs of 8.8 vs. 11.1. PGIC-based outcomes had NNTs similar to ≥ 30% pain reduction on average, while ≥ 50% pain reduction consistently yielded higher NNTs.
Key findings
- Across 76 trials, risk differences were higher for ≥ 30% than ≥ 50% pain reduction (difference 0.023, 95% CI 0.012-0.034; p < 0.001).
- NNTs were 8.8 for ≥ 30% pain reduction and 11.1 for ≥ 50% pain reduction.
- In 31 trials reporting all outcomes, ≥ 30% pain reduction and PGIC had comparable average NNTs; ≥ 50% pain reduction consistently had higher NNTs.
- Comparisons involving pain-relief scales were limited by sparse data and heterogeneity.
Why it matters
Responder definitions are not interchangeable: a stricter pain-reduction threshold can produce higher NNTs and make treatment effects appear smaller. This matters when interpreting analgesic efficacy across trials and when combining results in meta-analyses. Recognising how endpoint choice shapes estimates can support more meaningful comparisons and clearer interpretation of trial findings.
What to keep in mind
The analyses included only trials reporting at least two dichotomous outcomes; comparisons involving pain-relief scales were limited by sparse data and heterogeneity.