Get access

Estimation using all available covariate information versus a fixed look-back window for dichotomous covariates

Authors

  • Steven M. Brunelli,

    Corresponding author
    • Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    Search for more papers by this author
  • Joshua J. Gagne,

    1. Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    Search for more papers by this author
  • Krista F. Huybrechts,

    1. Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    Search for more papers by this author
  • Shirley V. Wang,

    1. Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    Search for more papers by this author
  • Amanda R. Patrick,

    1. Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    Search for more papers by this author
  • Kenneth J. Rothman,

    1. Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    2. RTI Health Solutions, Research Triangle Park, NC, USA
    3. Departments of Epidemiology and Medicine, Boston University Medical Center
    Search for more papers by this author
  • John D. Seeger

    1. Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School
    Search for more papers by this author

Correspondence to: S. M. Brunelli, Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital and Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA. E-mail: sbrunelli@partners.org

ABSTRACT

Purpose

When using claims data, dichotomous covariates (C) are often assumed to be absent unless a claim for the condition is observed. When available historical data differs among subjects, investigators must choose between using all available historical data versus data from a fixed window to assess C. Our purpose was to compare estimation under these two approaches.

Methods

We simulated cohorts of 20 000 subjects with dichotomous variables representing exposure (E), outcome (D), and a single time-invariant C, as well as varying availability of historical data. C was operationally defined under each paradigm and used to estimate the adjusted risk ratio of E on D via Mantel–Haenszel methods.

Results

In the base case scenario, less bias and lower mean square error were observed using all available information compared with a fixed window; differences were magnified at higher modeled confounder strength. Upon introduction of an unmeasured covariate (F), the all-available approach remained less biased in most circumstances and rendered estimates that better approximated those that were adjusted for the true (modeled) value of C in all instances.

Conclusions

In most instances considered, operationally defining time-invariant dichotomous C based on all available historical data, rather than on data observed over a commonly shared fixed historical window, results in less biased estimates. Copyright © 2013 John Wiley & Sons, Ltd.

Ancillary