Individual patient- versus group-level data meta-regressions for the investigation of treatment effect modifiers: ecological bias rears its ugly head

Authors

  • Jesse A. Berlin,

    Corresponding author
    1. University of Pennsylvania School of Medicine, Department of Biostatistics and Epidemiology and Center for Clinical Epidemiology and Biostatistics, Philadelphia, PA, U.S.A.
    • University of Pennsylvania School of Medicine, Center for Clinical Epidemiology and Biostatistics, 611 Blockley Hall, Philadelphia, PA 19104-6021, U.S.A.
    Search for more papers by this author
  • Jill Santanna,

    1. University of Pennsylvania School of Medicine, Department of Biostatistics and Epidemiology and Center for Clinical Epidemiology and Biostatistics, Philadelphia, PA, U.S.A.
    Search for more papers by this author
  • Christopher H. Schmid,

    1. Biostatistics Research Center, New England Medical Center and Tufts University School of Medicine, Boston, MA, U.S.A.
    Search for more papers by this author
  • Lynda A. Szczech,

    1. Duke University Medical Center, Division of Nephrology, Department of Medicine, Durham, NC, U.S.A.
    Search for more papers by this author
  • Harold I. Feldman

    1. University of Pennsylvania School of Medicine, Department of Biostatistics and Epidemiology and Center for Clinical Epidemiology and Biostatistics, Philadelphia, PA, U.S.A.
    2. University of Pennsylvania School of Medicine, Renal Electrolyte and Hypertension Division, Department of Medicine, Philadelphia, PA, U.S.A.
    Search for more papers by this author

Abstract

When performing a meta-analysis, interest often centres on finding explanations for heterogeneity in the data, rather than on producing a single summary estimate. Such exploratory analyses are frequently undertaken with published, study-level data, using techniques of meta-analytic regression. Our goal was to explore a real-world example for which both published, group-level and individual patient-level data were available, and to compare the substantive conclusions reached by both methods. We studied the benefits of anti-lymphocyte antibody induction therapy among renal transplant patients in five randomized trials, focusing on whether there are subgroups of patients in whom therapy might prove particularly beneficial. Allograft failure within 5 years was the endpoint studied. We used a variety of analytic approaches to the group-level data, including weighted least-squares regression (N=5 studies), logistic regression (N=628, the total number of subjects), and a hierarchical Bayesian approach. We fit logistic regression models to the patient-level data. In the patient-level analysis, we found that treatment was significantly more effective among patients with elevated (20 per cent or more) panel reactive antibodies (PRA) than among patients without elevated PRA. These patients comprise a small (about 15 per cent of patients) subgroup of patients that benefited from therapy. The group-level analyses failed to detect this interaction. We recommend using individual patient data, when feasible, to study patient characteristics, in order to avoid the potential for ecological bias introduced by group-level analyses. Copyright © 2002 John Wiley & Sons, Ltd.

Ancillary