ABSTRACT
Like primary research, meta-analysis of survival data (SD) involves a stepwise approach to arrive at statistically justifiable conclusions. It has the potential to provide an accurate, quantitative appraisal of the literature. It may objectively resolve controversies. The greatest challenge in conducting a meta-analysis on a clinical topic is often the lack of available data on the subject, because there are few high-quality published studies with an acceptable degree of heterogeneity. With regard to a meta-analysis of SD, the extraction of accurate data in order to derive a log hazard ratio is difficult. This can be explained by the process: (a) estimations via mathematical conversions from the provided summary data of which there might not be enough of sufficient quality in trial reports to derive out hazard ratio; and (b) direct measurements from a Kaplan-Meier survival curve, which can possibly introduce random measurement error and hence reduces the accuracy of results. If meta-analyses are to continue to have a role in surgical decision making, a key area in gynecologic oncology, clinicians need to be able to perform, assess, compare, and communicate the quality of meta-analyses, particularly in areas where several meta-analyses are available.
