Reporting of minimum clinically important differences in surgical trials

Abstract

Background The minimum clinically important difference (MCID) is the smallest difference in outcome between the groups that would be of clinical interest. It influences the estimates that are made to determine the required sample side. The aim of this study was to explore the reporting of the MCID in surgical trials. Method Surgical trials that were published between January 1981 and December 2006 in five prestigious surgical journals were evaluated. Selected for study were trials that studied two groups and reported the main outcome event as a proportion. Results Only 21% (100/486) of the admissible surgical trials mentioned a value for the MCID when estimating the sample size. There was a trend, however, for compliance with these factors to increase during the study period. The present post-hoc calculations of the required sample size, which were based on the observed differences between the groups at the end of the study, suggested that one-third of the trials should have accrued at least fivefold the number of patients. Although reporting an estimate of the sample size was associated with the study of more patients (median sample size 145 vs 100), it was not associated with the reporting of more positive results, that is, 61% (95/155) versus 65% (214/331). Conclusion There has been an improvement in the proportion of surgical trials reporting formal estimates of sample size during the last three decades. But the construct of these estimates is often suspect because of a failure to provide realistic values for the MCID. © 2009 Royal Australasian College of Surgeons.