Purpose: It has been suggested that nonrandomized studies of interventions can neither discriminate between the effect of an intervention and that of bias nor accurately estimate the magnitude of measured effects. This study was designed to compare the results of an historic control study of a surgical procedure with those of a subsequent randomized control trial conducted under similar circumstances.
Methods: The results of an historic control study and a randomized, controlled trial of the safety and efficacy of laparoscopic rectopexy for rectal prolapse that were conducted 17 months apart by the same group of surgeons at the same institution were compared in terms of direction and magnitude of measured effects.
Results: The historic control study was reliable in determining the direction of measured effects in six of six (100 percent) outcomes common between the two studies, and there was agreement on the statistical significance (or lack of it) in five (83 percent); however, the magnitude of measured effects for all but one outcome assessed was 33 to 107 percent larger than in the randomized, controlled trial. There was no agreement in the medical literature on the effect of the historic control design on the direction and magnitude of measured effects.
Conclusions: The results of a surgical historic control trial compared favorably with those of a randomized, controlled trial conducted under similar circumstances in determining the direction of measured effects but tended to yield larger estimates of effect magnitudes. The medical literature is divided on the effect of the historic control study design on study outcomes and more research is required to further define its role in evidence-based surgery.