Abstract
BACKGROUND: Suicide attempts constitute a significant public health emergency, with Emergency Departments (EDs) serving as a primary point of contact for individuals in acute psychiatric crisis. This observational, multicentric study aimed to analyse and compare the profiles of psychiatric consultations provided in the EDs of the University Hospital Systems (AOU) of Novara and Alessandria between January and December 2024, focusing on suicidal behaviours, clinical characteristics, and intervention outcomes. We hypothesised that territorial differences in patient demographics and service organisation would lead to distinct management patterns: Novara, expecting younger patients with chronic histories requiring hospitalisation, versus Alessandria, with older, self-referred cases, favouring territorial approaches. METHODS: Sociodemographic and clinical data were collected anonymously and pseudonymized from institutional software (Track Care, PsNet, REDCap). A total of 1,196 accesses requiring psychiatric evaluation were recorded. Multilevel logistic regression was performed to account for centre-level clustering (ICC = 0.12). RESULTS: The Novara cohort was significantly younger (mean age 42.5 vs. 47.1 years, p = 0.001) with more complex psychiatric histories (76.3% previous psychiatric history vs. 57.3%, p = 0.001) and higher rates of suicidal ideation and self-harm. Novara showed higher rates of acute pharmacological therapy (51.13% vs. 39.59%, p = 0.001) and voluntary hospital admissions (40.74% vs. 25.85%, p = 0.001). Conversely, Alessandria's older population (mean age 47.1 years) exhibited greater conscious suicidal intent (39.68% vs. 23.36%, p = 0.024) and used high-lethality methods. Alessandria adopted a more territorial management approach with higher discharge rates (27.23% vs. 18.89%) and community mental health centre (CMHC) referrals (21.23% vs. 17.96%). Substantial methodological differences were observed in substance use screening protocols. Novara reported 67.84% positive screens versus Alessandria's 12.22% (naive OR 14.2, 95% CI 11.0-18.3). Multilevel analysis accounting for centre-level effects reduced this estimate to OR 5.3 (95% CI 3.8-7.4, ICC = 0.12), representing a 63% attenuation. Sensitivity analysis restricted to the 537 patients uniformly screened yielded OR 4.9 (95% CI 3.5-6.8). These findings underscore that the apparent centre difference is substantially attributable to measurement bias rather than true epidemiological prevalence. Multivariate logistic regression confirmed a significant association between female sex and self-harming behaviours (OR 1.57, 95% CI 1.13-2.19, p = 0.008). CONCLUSIONS: This multicentric study revealed significant territorial differences in psychiatric emergency management between Novara and Alessandria. The Novara centre manages a younger population with complex, chronic psychiatric disorders, requiring an intensive, hospital-oriented approach. Alessandria serves an older population with higher conscious suicidal intent, managed primarily through territorial services. Standardisation of diagnostic protocols-particularly substance use screening-is essential for valid epidemiological comparison and uniform, evidence-based care.