Reducing length of hospital admissions by implementing same‐day general physician review in the emergency department of an Australian tertiary hospital

BACKGROUND: The presence of a general physician assessing patients upon referral in the emergency department (ED) may improve patient outcomes compared with the traditional model of post-take ward rounds. AIM: This study examines the impact of such an intervention on patient discharge rates and readmissions in a single centre. METHODS: From July 2019 to June 2020, a general physician was rostered to review patients on the day of referral from the ED. Four cohorts were compared: pre-intervention, intervention, concurrent controls (admitted during the intervention period but not reviewed) and post-intervention. Multivariate analyses evaluated the association between same-day consultant review and the likelihood of same-day discharge, as well as other variables including median length of stay (LoS) and readmission rate. RESULTS: Among 22 620 admissions, median LoS was shorter in the intervention period than in the before or after period (87 h vs. 97 and 93 h (P < 0.001)). Same-day general physician assessment increased the odds of day zero discharge by fivefold compared with before the intervention (adjusted odds ratio (aOR) 5.47, 95% confidence interval (CI): 3.62-8.26, P < 0.001). This effect was not sustained after the intervention ended (aOR 0.47, 95% CI 0.27-0.80, P < 0.006). Lower triage acuity and younger age were associated with discharge on day zero. Notably, longer ED decision time correlated with higher day zero discharge (aOR 1.09, 95% CI: 1.03-1.15), possibly reflecting appropriate consultant involvement in cases with an unclear disposition. Day zero discharge rates varied among individual general physicians (range 4.1%-14.0%). Increased day zero discharges did not lead to higher 7-day readmission rates. CONCLUSIONS: This large trial showed that early general physician review was associated with decreased median LoS and increased rate of day zero discharge, with no increase in readmission rates or mortality. Variability between physicians suggests potential for further optimisation in practice.