homepage link

Limitations and Misuse of Multicenter Trials: A Critique of The Women's Health Initiative (WHI) Study II*

Timothy D. Bilash, M.D., M.S.
August 2002


    The demands for concrete evidence about important clinical issues has encouraged the use of meta-analyses (retrospective pooling from individual studies). Failure of these to provide meaningful clinical results has spawned an increasing number of multi-centered prospective studies. There are inherent limitations to this approach, and study conclusions seem to reverse almost monthly in some cases.

    Bigger is not necessarily better, and often predictably so. Significance can not be created out of flawed or incomplete data, or when data bias requires the use of plausibility arguments. These studies can guide and improve subsequent research, and to provide a background for clinical decisions by physicians, but their use to create global standards of care or government regulation is questionable.

    This paper challenges the validity of a recent publication about a carefully designed hormone replacement study (*JAMA, July 17, 2002, 288(3):321-333). It is shown that numerous problems in the study, such as high treatment crossover and dropout rates, render the outcome variables uninterpretable. Despite this, a myriad of publications and societies have echoed the study conclusions, without addressing these limitations.[1]

    The only conclusion of the paper, that the Prempro(CEE.625/MPA2.5) regimen "should not be initiated or continued for primary prevention of CHD" (Coronary Heart Disease), is not supported by the data.[2,3] Rather another important outcome, to which CHD greatly contributes, is supported by the data. The Overall Mortality Rate (OMR) shows a widening improvement beyond 5 years after hormone therapy is initiated. Unfortunately, the study was terminated at this point.

    It is important for clinicians to appreciate the limitations of research studies. Heuristic guidelines for the statistical interpretation of survival data and rate studies is provided.

    References noted January 2003:
    [1] "Preliminary Statement to ACOG Membership on the Women's Health Initiative Study", July 10, 2002
    [2] "Cardiovascular Disease and Postmenopausal Hormone Therapy", Leon Speroff, presented at Current Controversies in Obstetrics and Gynecology, Nov22-24, 2002, Newport Beach, CA
    [3] "Postmenopausal Hormone Therapy and Breast Cancer", Leon Speroff, presented at Current Controversies in Obstetrics and Gynecology, Nov22-24, 2002, Newport Beach, CA


goto homepage

page views since Sept2007