Randomized clinical trials (RCTs) stopped early for benefit often receive great attention and affect clinical practice, but pose interpretational challenges for clinicians, researchers, and policy makers. Because the decision to stop the trial may arise from catching the treatment effect at a random high, truncated RCTs (tRCTs) may overestimate the true treatment effect. The
Study Of Trial Policy Of Interim Truncation (STOPIT-1), which systematically reviewed the epidemiology and reporting quality of tRCTs, found that such trials are becoming more common, but that reporting of stopping rules and decisions were often deficient. Most importantly, treatment effects were often implausibly large and inversely related to the number of the events accrued. The aim of STOPIT-2 is to determine the magnitude and determinants of possible bias introduced by stopping RCTs early for benefit. Methods/Design
We will use sensitive strategies to search for systematic reviews addressing the same clinical question as each of the tRCTs identified in STOPIT-1 and in a subsequent literature search. We will check all RCTs included in each systematic review to determine their similarity to the index tRCT in terms of participants, interventions, and outcome definition, and conduct new meta-analyses addressing the outcome that led to early termination of the tRCT. For each pair of tRCT and systematic review of corresponding non-tRCTs we will estimate the ratio of relative risks, and hence estimate the degree of bias. We will use hierarchical multivariable regression to determine the factors associated with the magnitude of this ratio. Factors explored will include the presence and quality of a stopping rule, the methodological quality of the trials, and the number of total events that had occurred at the time of truncation.
Finally, we will evaluate whether Bayesian methods using conservative informative priors to "regress to the mean" overoptimistic tRCTs can correct observed biases.
A better understanding of the extent to which tRCTs exaggerate treatment effects and of the factors associated with the magnitude of this bias can optimize trial design and data monitoring charters, and may aid in the interpretation of the results from trials stopped early for benefit.