EVidence Synthesis For Health Technology Assessment With Limited Studies

Abstract

Pairwise and network meta-analyses using fixed effect and random effects models are commonly applied to synthesise evidence from randomised controlled trials in health technology assessment. Fixed effect models are often used because there are too few studies with which to estimate the between-study standard deviation from the data alone. When heterogeneity is expected and inferences are required beyond the sample of studies available, then an analysis using a fixed effect model will underestimate uncertainty about the treatment effect. This research aims to provide a Bayesian approach to overcome the problem of too few studies to use a random effects model, by proposing a framework for eliciting an informative prior distribution for the between-study standard deviation in a random effects model to genuinely represent heterogeneity. We developed an elicitation method using external information such as empirical evidence and experts’ beliefs on the ‘range’ of treatment effects in order to infer the prior distribution for the between-study standard deviation. We also developed the method to be implemented in R using the SHELF package. The three-stage elicitation approach allows uncertainty to be represented by a genuine prior distribution to avoid making misleading inferences. It is flexible to what judgments an expert can provide, and is applicable to all types of outcome measure for which we can construct a treatment effect on an additive scale. The choice between using a fixed effect or random effects meta-analysis model depends on the objective of the analysis and knowledge of the included studies, but not on the number of available studies. Our elicitation framework captures external evidence about heterogeneity and overcomes the often implausible assumption that studies are estimating the same treatment effect, thereby improving the quality of inferences in decision making.