Differences among specialists in using the Electronic Medical Record alert system for HCV screening in outpatients at a large general hospital.

BACKGROUND & OBJECTIVES: Some hepatitis C virus (HCV) patients remain underdiagnosed at a large hospital. The electronic medical record (EMR) alert system can increase in-hospital screening. This study evaluates the effectiveness of screening among outpatients and assesses variations in physician engagement across different specialties. METHODS: From December 2022 to April 2023, outpatients aged 45-79 years, with diabetes mellitus or chronic kidney disease, who did not have anti-HCV data in their EMR were identified. A pop-up alert prompted physicians to either order anti-HCV testing or refer patients for a health check-up. The frequency of alerts, physician response rates, blood test completion rates, and HCV seropositivity were analyzed by specialty. RESULTS: Of 85,320 alerts generated, 61.4% elicited a physician response. Internal Medicine clinics had significantly higher response rates than non-Internal Medicine clinics (77.5% versus 44.6%, P < 0.01). Among patients with physician responses, 30.9% completed the study, with markedly higher completion when blood tests were directly ordered compared to referrals (88.4% versus 7.2%, P < 0.01). The monthly number of screenings increased tenfold after the system was implemented. Overall, 22,377 patients (26.2%) were screened, and 485 (2.17%) were seropositive. One-fourth of seropositive cases were identified in non-Internal Medicine clinics, where the positivity rate was higher (2.7% versus 2.04%, P = 0.01). CONCLUSIONS: Short-term use of the EMR alert system increased HCV detection among outpatients. The default test orders in the EMR improved completion rates more than referrals to a health check-up. There were differences in physician engagement and seropositive case detection among specialties. Targeted interventions, particularly among specialties with lower engagement, are necessary to improve the success of HCV screening.