General Practitioner Referral in Emergency Department Presentations with Acute Infection: A Retrospective Observational Study

Patients with infection present to emergency departments (EDs) through multiple referral pathways, including general practitioner (GP) referral and self/ambulance presentation. The impact of referral sources on presentation characteristics, ED processes, and clinical outcomes remains poorly characterised. A retrospective observational study examined ED presentations with infection across eight sampling points in 2025 in an inner-city ED located in an area of severe deprivation. Patients were stratified into general practitioner (GP)/GP out-of-hours (GP-OOH) referrals and other referral pathways. Demographic characteristics, physiological observations, ED process measures, and clinical outcomes were compared using unadjusted analyses, with multivariable median regression used to adjust the principal exploratory ED process outcomes for clinically relevant covariates. Among 277 patients, 55 (19.9%) were referred by GP/GP-OOH and 222 (80.1%) via other routes. GP-referred patients more frequently presented on weekdays and arrived by private transport (both p < 0.001). No statistically detectable differences were observed in most demographic, clinical, and physiological measures, although the study was not designed or powered to establish equivalence. GP-referred patients experienced longer times to triage (p < 0.001) and first physician assessment (p = 0.003). GP-referred patients represented a minority of infection presentations. No statistically significant differences were detected in most clinical outcomes, although admission was less frequent among GP-referred patients (36.4% vs. 51.8%; absolute difference −15.4 percentage points; risk ratio 0.70, 95% CI 0.48–1.02), and clinically important differences cannot be excluded. GP referral was associated with longer waits for early ED assessment, while differences in downstream treatment and clinical outcomes should be interpreted cautiously given the exploratory nature of these comparisons and the small number of GP-referred patients.