Estimating Generic Drug Utilization With Electronic Health Records From a Mixed-Payer Ambulatory Health Care Delivery System

Authors

Robert J. Romanelli, Palo Alto Medical Foundation Research InstituteFollow
Vani Nimbal, Palo Alto Medical Foundation Research Institute
Trang Le, University of California, San Francisco
Jodi B. Segal, Johns Hopkins School of Medicine

Publication Date

8-15-2016

Keywords

generic drugs, electronic health records

Abstract

Background/Aims: Generic drug utilization in an outpatient setting is typically measured through reconciled pharmacy claims data; however, not all health care systems have access to these data for their populations. We sought to develop and validate an automated algorithm to estimate generic drug utilization with electronic health records (EHR) data.

Methods: We chose 25 therapeutic classes to study due to their potential for poor generic uptake. We developed a generic utilization algorithm using EHR prescribing data (i.e. medication name and indicator of a dispense-as-written request) from a mixed-payer health care system in northern California in 2013. These data were combined with information on generic product availability from Drugs@FDA and Redbook™ at the time prescriptions were written. The algorithm was validated in a cohort of ambulatory managed-care beneficiaries for whom prescribing data and pharmacy claims data (i.e. gold standard) are both available. An estimated generic utilization rate (GUR) across therapeutic classes was calculated by applying the algorithm, and its performance was assessed. We next applied the algorithm to electronic prescribing data from a cohort of non-managed-care beneficiaries for whom pharmacy claims data are not available.

Results: We identified 20,750 managed-care beneficiaries with an electronic prescription for a drug product in the therapeutic classes of interest in 2013 and a corresponding pharmacy claim. We also identified 119,153 non-managed-care beneficiaries with electronic prescriptions for products in the same therapeutic classes. Managed-care and non-managed-care beneficiaries were similar with respect to age, sex and race/ethnicity. Applying the algorithm, we estimated a GUR of 58.7% among managed-care beneficiaries; the actual GUR based on pharmacy claims was 57.2%. Sensitivity (96%) and specificity (94%) of the algorithm were high, and total percent agreement was 96%. In the cohort of non-managed-care beneficiaries, a similar GUR was estimated (56.1%).

Conclusion: A generic utilization algorithm that combined EHR data with information on generic product availability was validated and performed well in estimating the GUR for outpatient prescriptions in a cohort of managed-care beneficiaries, independent of pharmacy claims data. Health care delivery systems may apply the methods described herein to quantify generic utilization in their ambulatory populations for quality improvement and research initiatives.

Recommended Citation

Romanelli RJ, Nimbal V, Le T, Segal JB. Estimating generic drug utilization with electronic health records from a mixed-payer ambulatory health care delivery system. J Patient Cent Res Rev. 2016;3:224.