A framework for the pre-specification of statistical analysis strategies in clinical trials (Pre-SPEC)
share
Bias can be introduced into clinical trials if statistical methods are chosen based on subjective assessment of the trial data. Pre-specification of the planned analysis approach is essential to help reduce such bias. However, many trials fail to adequately pre-specify their statistical analysis approach, thereby allowing analysts to choose the method which gives the most favourable result. We propose a five-point framework for the pre-specification of the statistical analysis strategy for a clinical trial's primary outcome (the Pre-SPEC framework); this framework is designed to ensure that methods cannot be chosen based on the trial data in order to give a more favourable result. The five points are: (1) Pre-specify before recruitment to the trial begins; (2) Specify a single primary analysis strategy; (3) Plan all aspects of the analysis; (4) Enough detail should be provided so that a third party could independently perform the analysis; and (5) Adaptive analysis strategies should use deterministic decision rules. This framework could be used as a template to help plan an analysis strategy, or to evaluate whether an existing strategy is at risk of bias due to inadequate pre-specification.
READ FULL TEXT
