Background: Percentage study weights break down the summary meta-analysis result into the relative contribution of each individual study, and are easily interpretable by patients and health professionals. They are especially important when some studies are potential outliers or at high risk of bias, to reveal how much their data contributed to the overall meta-analysis result. Similarly, if clinical decision makers are concerned that a meta-analysis result may be unreliable for translation to practice due to the inclusion of a study from a different clinical population or setting, then knowing the study’s contribution is important.

Objectives: Though percentage study weights are easily derived in traditional (e.g. inverse-variance) meta-analysis models of a single effect, they are also needed in advanced meta-analyses containing multiple effects, such as meta-regression and IPD meta-analysis. In this talk, we propose how to derive percentage study weights for such situations, in order to reveal the (otherwise hidden) contribution of each study toward the effects of interest.

Methods: We assume that studies are independent, and utilise a decomposition of Fisher’s information matrix to decompose the total variance matrix of parameter estimates into study-specific contributions, from which percentage weights are derived. This approach generalises how percentage weights are calculated in a traditional, single parameter meta-analysis model.

Results: Application will be shown to one-stage and two-stage IPD meta-analyses, meta-regression, test accuracy meta-analysis, and network (multivariate) meta-analysis of multiple treatments. These reveal percentage study weights toward clinically important estimates, such as summary treatment effects and treatment-covariate interactions, and even toward methodologically important measures, such as the magnitude of ecological bias (difference between within-study and across-study treatment-covariate interactions) and the amount of inconsistency (difference between direct and indirect evidence in a network meta-analysis).

Conclusions: Percentage study weights are immediately accessible for patients and health professionals, and thus should be routinely presented to improve transparency of each study’s contribution toward meta-analysis results.

Patient or healthcare consumer involvement: None