Measuring prescriptions dispensed from urgent care through the VA community care benefit

Abstract

Background

The Department of Veterans Affairs (VA) now offers eligible Veterans an urgent care benefit covering visits and 14-day prescriptions outside of VA. Prescriptions written and dispensed outside VA lack the clinical decision support of VA-issued prescriptions, raising concerns about safety and polypharmacy. To date, there has been limited analyses of prescribing patterns through the urgent care benefit.

Methods

We used a repeated cross-sectional design to examine Veterans who filled non-VA urgent care prescriptions from 07/30/2019 to 03/20/2023. Data were sourced from the Community Care Reimbursement System (CCRS), which tracks all VA-paid medications dispensed by non-VA pharmacies. We identified potentially noncompliant prescriptions as those not meeting VA urgent care benefit restrictions. We also identified prescriptions continued in VA as a “new VA medication” after 30-days from the urgent care fill.

Results

Overall, 83,862 Veterans received 271,476 non-VA urgent care prescriptions. Veterans’ average age was 55.9, with 79.3 % male, 73.0 % White, 86.7 % non-Hispanic, and 41.4 % rural dwelling. Urgent care use increased from 341 prescription fills in March 2020 to 9738 in January 2023. Frequently filled prescriptions included antimicrobials (n = 114,492, 42.2 %) and hormones/synthetics/modifiers, like steroids (n = 44,457, 16.4 %). Potentially noncompliant prescriptions accounted for 9.3 %, with 6.7 % not on the urgent/emergent formulary and 2.6 % supplied for over 14 days. Over 70,704 (26.0 %) prescriptions were continued in VA post-urgent care visit, of which 15 % had no prior VA fill (i.e., new VA medication). Veterans with new continued VA prescriptions were more likely to be male (79.4 % vs. 73.9 %) and from urban areas (59.3 % vs. 57.5 %) (All P < .001).

Conclusions

Veterans increasingly received non-VA prescriptions through urgent care centers in the community from 2019 to 2023, including drug classes of interest to VA due to potential risks of inappropriate prescribing (e.g., steroids) or drug interactions (e.g., antibiotics). The CCRS database can be integrated with other VA databases as a quality improvement tool to improve care coordination and drug safety.

Implications

This evaluation highlights the need for improved clinical decision support for non-VA prescriptions and demonstrates the potential of integrated data systems to monitor and enhance medication safety and coordination within VA.

Level of evidence

Cross-sectional analysis of national VA data.