BORROWING STRENGTH FROM EXTERNAL TRIALS IN A METAANALYSIS

There exists a variety of situations in which a random effects meta-an alysis might be undertaken using a small number of clinical trials. A problem associated with small meta-analyses is estimating the heteroge neity between trials. To overcome this problem, information from other related studies may be incorporated into the meta-analysis. A Bayesia n approach to this problem is presented using data from previous meta- analyses in the same therapeutic area to formulate a prior distributio n for the heterogeneity. The treatment difference parameters are given non-informative priors. Further, related trials which compare one or other of the treatments of interest with a common third treatment are included in the model to improve inference on both the heterogeneity a nd the treatment difference. Two approaches to estimating relative eff icacy are considered, namely a general parametric approach and a metho d explicit to binary data. The methodology is illustrated using data f rom 26 clinical trials which investigate the prevention of cirrhosis u sing beta-blockers and sclerotherapy. Both sources of external informa tion lead to more precise posterior distributions for all parameters, in particular that representing heterogeneity.