Objectives
To assess both in publications and trial registries the reporting of reasons of participant withdrawals in randomized controlled trials (RCTs) evaluating targeted therapies for chronic inflammatory skin diseases, to evaluate the discrepancies between these sources and to classify the reported reasons for withdrawal.
Study Design and Setting
We included RCTs published between September 27, 2009, and October 1, 2024, from four systematic reviews assessing the efficacy or safety of targeted therapies in adult patients with plaque psoriasis, hidradenitis suppurativa, atopic dermatitis, or alopecia areata. The main outcome was the proportion of RCTs presenting at least one discrepancy in the number of withdrawn participants (WP) in total and per treatment arm between the published article and the trial registry. For trials reporting the same number of WP in both sources, we also evaluated discrepancies in reasons for withdrawal. We classified the reasons for withdrawal (e.g. inefficacy, adverse events) and the mechanism (e.g. missing at random, missing not at random).
Results
Among the 96/209 included RCTs reporting WP data for the primary endpoint in both sources, 25 (26%) had discrepancies between sources in total WP number. Then, among the 71 with the same WP number, 5 did not report reasons and 34 report a different number and/or reasons between sources. Thus, overall, 32/96 trials (33%) were fully concordant for both WP numbers and reasons. Among the 73 and 59 different reasons reported in publications and trial registries respectively, 70% (51/73) and 68% (40/59) were not informative enough to be classified.
Conclusion
Our study revealed significant shortcomings in the reporting of participant withdrawals, marked by incomplete data, discrepancies between sources, and a lack of standardized terminology in RCTs evaluating systemic targeted therapies for chronic inflammatory skin diseases.
Robin Guelimi