Assigning exposure duration to single prescriptions by use of the waiting time distribution

Pharmacoepidemiol Drug Saf. 2013 Aug;22(8):803-9. doi: 10.1002/pds.3459. Epub 2013 May 23.


Purpose: The purpose of this study is to develop and present a method to calculate which exposure duration should be assigned to single prescriptions for use in databases where information on expected duration is not recorded.

Methods: We propose a method, based on the waiting time distribution (WTD), which estimates how frequently prevalent users redeem new prescriptions. However, we added two steps to the WTD-approach. First, we excluded incident users, representing noise in the analysis. Second, we calculated the cumulative percentage of users that had presented themselves after a given number of days. Using a cutoff value of 80%, we thus calculated the number of days for the majority of prevalent users to present themselves, that is, the exposure duration that should be assigned to the single prescription. The primary strength of the method is that it can be applied in a standardized fashion and that it does not require information on dosage instructions. The primary weakness of the method is that it is only usable on drugs with predominantly chronic use patterns. The method was tested using four model drugs: bendroflumethiazide, warfarin, levothyroxine, and non-steroidal anti-inflammatory drugs (NSAIDs).

Results: We found exposure period estimates of 92, 86, 69, and 210 days for bendroflumethiazide, levothyroxine, warfarin, and NSAIDs prescriptions, respectively. NSAIDs were found not to comply with the requirements of the method. The calculated number of exposed days was only slightly influenced by the assumptions of the method.

Conclusions: The method provides a useful approach to generate automated estimates of exposure duration that should be assigned to each prescription.

Keywords: drug prescriptions; drug utilization; methods; pharmacoepidemiology; registries.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Databases, Factual
  • Denmark
  • Drug Prescriptions / statistics & numerical data
  • Humans
  • Pharmacoepidemiology*
  • Practice Patterns, Physicians' / statistics & numerical data*
  • Prescription Drugs / administration & dosage
  • Prescription Drugs / therapeutic use*
  • Time Factors


  • Prescription Drugs