Asynchronous Electronic Screening for Unhealthy Alcohol Use Among Veterans in Primary Care

Importance: Screening for unhealthy alcohol use is recommended in primary care; however, completion and quality are inconsistent especially during telemedicine visits. Little is known about optimal workflows incorporating electronic screening (e-screening). Objective: To evaluate whether use of previsit asynchronous e-screening is associated with improved completion and detection of unhealthy alcohol use via the Alcohol Use Disorders Identification Test (AUDIT-C) questionnaire compared with usual staff-administered screening during telemedicine primary care visits. Design, Setting, and Participants: Pragmatic cluster randomized quality improvement trial conducted at 2 primary care clinics in the Veterans Health Administration (VHA) from June 24 to August 1, 2024. Primary care clinicians (PCCs) were randomized 1:1, stratified by site, to intervention or control. Intervention: For PCCs in the control arm, patients received usual care including staff-administered AUDIT-C at telemedicine visits. For PCCs in the intervention arm, 24 to 48 hours before visits patients additionally received an invitation to asynchronous self-administered e-screening. Veterans who did not complete e-screening were still eligible for staff completion of screening during their clinic visits. Main Outcomes and Measures: The primary outcome was completion of AUDIT-C; secondary outcome was positive screen result (AUDIT-C ≥5). The exploratory outcome was brief intervention after positive screen result. All statistical models were clustered by PCC and adjusted for patient age, sex, race and ethnicity, comorbidity, prior primary care use, and site. Results: Among 848 veterans in the primary analysis (mean [SD] age, 55.4 [16.1] years; 729 [86.0%] male), use of e-screening was associated with increased telemedicine visit screening completion rates by 30.5 percentage points (74.4% [95% CI, 68.5%-80.3%] for e-screening vs 43.9% [95% CI, 26.6%-61.2%] for usual care; P < .001) and with increased likelihood of a positive screen result (10.6% [95% CI, 8.0%-13.2%] for e-screening vs 2.7% [95% CI, 0.7%-4.7%] for usual care; P < .001). Exploratory analysis identified the proportion of veterans receiving a brief intervention after a positive screen result (2.3% [10 of 442] for usual care vs 5.9% [24 of 406] for e-screening; P = .01). Conclusions and Relevance: In this study, use of asynchronous e-screening was associated with improved completion and screen-positive results for unhealthy alcohol use in primary care, with the greatest gains for telemedicine encounters. Overall, this approach may close the implementation gap for population-based screening, improve disclosure, and reduce staff burden, particularly in hybrid care models. Trial Registration: isrctn.org Identifier: ISRCTN16316660.