Meta-Analysis and Sparse Data Bias.

Meta-Analysis and Sparse Data Bias.
Am J Epidemiol. 2020 Sep 25;:
Authors: Richardson DB, Cole SR, Ross RK, Poole C, Chu H, Keil AP
Abstract
Meta-analyses are undertaken to combine information from a set of studies, often in settings where some of the individual study-specific estimates are based on relatively small study samples. Finite sample bias may occur when maximum likelihood estimates of associations are obtained by fitting logistic regression models to sparse data sets. We show that combining information from small studies by undertaking a meta-analytic summary of logistic regression estimates can propagate such sparse data bias. In simulations, we illustrate two challenges encountered by meta-analyses of logistic regression results in settings of sparse data: bias in the summary meta-analytic result; and, confidence interval coverage that can worsen, rather than improve, in terms of being less than nominal, as the number of studies in the meta-analysis increases.
PMID: 32975277 [PubMed - as supplied by publisher]

Source: Am J Epidemiol - September 24, 2020 Category: Epidemiology Authors: Richardson DB, Cole SR, Ross RK, Poole C, Chu H, Keil AP Tags: Am J Epidemiol Source Type: research