The characteristics and quality of randomized controlled trials in neuropathic pain: a descriptive study based on a systematic review.

BACKGROUND

Evidence from randomized controlled trials is regarded as the gold standard in clinical research and yet the quality of the conduct and reporting of trials is variable, even post-Consolidated Standards of Reporting Trials. This study arose from a systematic review and cost-effectiveness analysis of treatment for neuropathic pain. The aim was to provide a description of the included trials and investigate trends in study characteristics and measures of quality over time.

METHODS

The review provided data regarding study characteristics (patients, place, time, drugs, outcomes), methodological factors (sample size calculations, randomization, reporting baseline patient data, withdrawals, intention-to-treat (ITT), and statistical analysis (completeness and correctness of reporting of results, methods of analysis).

RESULTS

A total of 131 trials from 1969 to 2007 were included. Of these, 63% were parallel-group designs, the remainder were cross-over; 73% were placebo-controlled. Several trial features increased or improved over time: trial size, quality (using Jadad score), presentation of baseline data by group, reporting of power calculations, use of visual analogue score or numerical rating scale scales to assess pain, completeness of reporting of statistical results, use of modeling to allow for baseline pain scores. The proportion of withdrawals was constant over time with a mean of 14.3%. The proportion of studies stating the analysis as ITT, increased over time, but inspection of papers indicated that the proportion confirmed as ITT was unchanged.

CONCLUSIONS

There have been a number of improvements regarding the quality and reporting of randomized controlled trials in neuropathic pain, but some failings remain that at best make some results difficult to interpret and at worst lead to bias.