Meta-analysis of published studies or meta-analysis of individual data? Caesarean section in HIV-positive women as a study case

Very few comparative evaluations of meta-analysis of published data and meta-analysis using individual patient data have appeared in the medical literature. The association between type of delivery and HIV perinatal transmission appears to be an excellent study case, given the recent publication of the meta-analysis of individual patient data by the International Perinatal, HIV Group. In this paper, we report the results of a meta-analysis of the published studies, which show a statistically significant reduction of perinatal HIV transmission rate with Caesarean section. The results are surprisingly similar to those of the recently published meta-analysis of individual patient data, indicating that, in the absence of significant confounding, the two meta-analytic methods are likely to give consistent results. In this era of constrained resources for biomedical research, caution should be taken in abandoning meta-analysis of published data for studying epidemiological associations of public health interest. (C) 2003 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.