The Tyranny of Meta-analysis and the Misuse of Randomized Controlled Trials in Maternity Care

Authors

  • Michael C. Klein MD, CCFP, FAAP, FCPS

    1. Michael C. Klein is an Emeritus Professor of Family Practice & Pediatrics at the University of British Columbia, and Senior Scientist Emeritus, Centre Developmental Neurosciences & Child Health, Child and Family Research Institute, Vancouver, British Columbia V6H 3V4, Canada.
    Search for more papers by this author

  • COMMENTARY ON:
    Wassen MMLH, Zuijlen J, Roumen FJME, Smits LJM, Marcus MA, Nijhuis JG. Early vs late epidural analgesia and the risk of instrumental delivery in nulliparous women. BJOG 2011;118:655–661.

Address correspondence to Michael C. Klein, MD, CCFP, FAAP, FCPS, Room L408, 4480 Oak Street, Vancouver, British Columbia V6H 3V4, Canada.

Abstract

Abstract:  Recent meta-analyses of key areas in maternity care have covered home birth and epidural analgesia. In each of these cases serious issues have arisen from the use of subjective inclusion and exclusion criteria, heterogeneity of included studies, and inclusion of studies that were conducted in settings that were not representative of usual maternity care. This latter flaw is especially notable for early epidural analgesia, where study environments with very low cesarean section rates are included. Such study settings lack external validity and have raised concerns about the political uses of meta-analysis. For a meta-analysis to be useful, the included studies must be broadly representative of the way that maternity care is carried out in usual birth environments. (BIRTH 39:1 March 2012)

Ancillary