Bhandari et al. (1) studied the use of baseline comparability tests and subgroup analyses in surgical randomized controlled trials. Published surgical randomized controlled trials in four medical journals and in the Journal of Bone and Joint Surgery (American and British volumes) were identified.
The study identified 72 randomized controlled trials. Of 166 significance tests, 17 (10%) were statistically significant. Twenty-seven (38%) trials included 54 subgroup analyses, and inappropriate emphases on subgroup analyses were frequent. Forty-nine (91%) analyses were performed without prior hypotheses. Investigators reported differences between subgroups in 31 (57%) of the analyses, all of which were featured in the summary or conclusion.
The authors conclude that the inferences may be misleading.
Bhandari M, Deveraux PJ, Li P, Mah D, Lim K, Schünemann HJ, Tornetta P. Misuse of baseline comparison tests and subgroup analyses in surgical trials. Clin Orthop Res 2006;447:247-251.