Evaluation of gestational age and admission date assumptions used to determine prenatal drug exposure from administrative data

Authors

  • Marsha A. Raebel PharmD,

    Corresponding author
    1. HMO Research Network, Center for Education and Research on Therapeutics (CERTs), Denver, CO, USA
    2. Kaiser Permanente of Colorado Clinical Research Unit, Denver, CO, USA
    3. University of Colorado, School of Pharmacy, Denver, CO, USA
    • Kaiser Permanente of Colorado Clinical Research Unit, PO Box 378066, Denver, CO 80237-8066, USA.
    Search for more papers by this author
  • Jennifer L. Ellis MSPH,

    1. HMO Research Network, Center for Education and Research on Therapeutics (CERTs), Denver, CO, USA
    2. Kaiser Permanente of Colorado Clinical Research Unit, Denver, CO, USA
    Search for more papers by this author
  • Susan E. Andrade ScD

    1. HMO Research Network, Center for Education and Research on Therapeutics (CERTs), Denver, CO, USA
    2. Meyers Primary Care Institute, Worcester, MA, USA
    Search for more papers by this author

  • No conflict of interest was declared.

Abstract

Objective

Our aim was to evaluate the 270-day gestational age and delivery date assumptions used in an administrative dataset study assessing prenatal drug exposure compared to information contained in a birth registry.

Study Design and Setting

Kaiser Permanente Colorado (KPCO), a member of the Health Maintenance Organization (HMO) Research Network Center for Education and Research in Therapeutics (CERTs), previously participated in a CERTs study that used claims data to assess prenatal drug exposure. In the current study, gestational age and deliveries information from the CERTs study dataset, the Prescribing Safely during Pregnancy Dataset (PSDPD), was compared to information in the KPCO Birth Registry. Sensitivity and positive predictive value (PPV) of the claims data for deliveries were assessed. The effect of gestational age and delivery date assumptions on classification of prenatal drug exposure was evaluated.

Results

The mean gestational age in the Birth Registry was 273 (median = 275) days. Sensitivity of claims data at identifying deliveries was 97.6%, PPV was 98.2%. Of deliveries identified in only one dataset, 45% were related to the gestational age assumption and 36% were due to claims data issues. The effect on estimates of prevalence of prescribing during pregnancy was an absolute change of 1% or less for all drug exposure categories. For Category X, drug exposures during the first trimester, the relative change in prescribing prevalence was 13.7% (p = 0.014).

Conclusion

Administrative databases can be useful for assessing prenatal drug exposure, but gestational age assumptions can result in a small proportion of misclassification. Copyright © 2005 John Wiley & Sons, Ltd.

Ancillary