Assessing the potential for bias in meta-analysis due to selective reporting of subgroup analyses within studies

Subgroup analysis is frequently used to investigate heterogeneity in meta-a nalysis. Subgroup data are not always available, and researchers should rec ord what data were available for each trial. If data were not available, an d it is known that the subgroup data were collected, the potential for sele ctive reporting should be considered. Bias due to selective publishing of r eports is widely recognized in meta-analysis. In contrast, selective report ing within studies is little discussed but potentially important. We explor ed this problem by evaluating the effect of potential bias in subgroup anal ysis due to within-study selective reporting with an existing meta-analysis . The review addressed malaria chemoprophylaxis in pregnancy. The conclusio n in the original review, that benefit is limited to primigravidae, was bas ed on subgroup analysis using the three trials out of five which reported o n subgroups. We developed a method of sensitivity analysis that imputes dat a for the missing subgroups to assess the robustness of the results and the conclusions drawn. In this particular example, our analysis indicates that the estimate of effect reported in the review is most likely to overestima te the true effect and the conclusion that benefit is limited to primigravi dae may be false. Copyright (C) 2000 John Wiley & Sons, Ltd.