Characterizing emergency department referral trends using an electronic health record-integrated referral order

Ambulatory clinics increasingly refer patients to the emergency department (ED) when outpatient resources are limited, exacerbating workflow strain and delaying emergent evaluation. An electronic referral process was implemented at a large academic medical center to centralize requests. We sought to characterize the ambulatory-to-ED referral population and evaluate temporal patterns relevant to operational improvement.

Electronic referral orders placed between September 1, 2024, and September 30, 2025, were retrospectively analyzed. Outcomes included arrival timing, consultation, and disposition. Analysis of variance and Chi-squared tests with standardized residuals evaluated temporal trends.

Of 5380 orders, 4.896 (91.0%) matched an ED encounter. Order volume increased over the project period, across specialties and locations. Among matched encounters, most occurred on weekdays (90.5%) and during daytime hours (81.5%), with bimodal peaks at 1200 and 1600. Arrival volume differed across day, evening, and night shifts (P < 0.001). Of unmatched orders (9.0%), 25% occurred on Fridays. Disposition differed significantly between weekends and weekdays (χ^2 = 23.8, P < 0.001), with higher weekend admissions (64.8% vs 53.9%). Admission rates were lowest on Fridays (52.7%). Consultation-related outcomes did not vary by weekday (P = 0.209).

Referral arrivals demonstrated a bimodal distribution, differing from commonly reported ED arrival patterns. Friday referral patterns suggest opportunities to expand outpatient management capacity. Analysis of electronic referral data may inform resource allocation and improve coordination between ambulatory and emergency care, particularly in medical centers with integrated referral systems.