Recursive cumulative meta-analysis: A diagnostic for the evolution of total randomized evidence from group and individual patient data

Meta-analyses of randomized evidence may include published, unpublished, an d updated data in an ongoing estimation process that continuously accommoda tes more data. Synthesis may be performed either with group data or with me ta-analysis of individual patient data (MIPD). Although MIPD with updated d ata is considered the gold standard of evidence, there is a need for a care ful study of the impact different sources of data have on a meta-analysis a nd of the change in the treatment effect estimates over sequential informat ion steps. Unpublished data and late-appearing data may be different from e arly-appearing data. Updated information after the end of the main study fo llow-up may be affected by cross-overs, missing information, and unblinding . The estimated treatment effect may thus depend on the completeness and up dating of the available evidence. To address these issues, we present recur sive cumulative meta-analysis (RCM) as an extension of cumulative metaanaly sis. Recursive cumulative meta-analysis is based on the principle of recalc ulating the results of a cumulative meta-analysis with each new or updated piece of information and focuses on the evolution of the treatment effect a s a more complete and updated picture of the evidence becomes available. An examination of the perturbations of the cumulative treatment effect over s equential information steps may signal the presence of bias or heterogeneit y in a meta-analysis. Recursive cumulative meta-analysis may suggest whethe r then is a true underlying treatment effect to which the meta-analysis is converging and how treatment effects are sequentially altered by new or mod ified evidence. The method is illustrated with an example from the conduct of an MIPD on acyclovir in human immunodeficiency virus infection. The rela tive strengths and limitations of both meta analysis of group data and MIPD are discussed through the RCM perspective. (C) 1999 Elsevier Science Inc.