Socio-demographic and Clinical Predictors of Relapses in Severe Mental Disorders: A Cross-sectional Study at El Razi Hospital, Tunisia
S. Ben Othmen, M. Sehli, M. A. Yaiche, H. Mami, A. AissaIntroduction
Relapses in schizophrenia and bipolar disorder are common and contribute to poor functional outcomes and frequent hospitalizations. Understanding the factors associated with relapse is essential to improve long-term management. In Tunisia, where psychiatric resources are limited, such data remain scarce and context-specific evidence is needed.
Objectives
This study aimed to identify socio-demographic and clinical predictors of relapse in a Tunisian cohort.
Methods
A descriptive cross-sectional study was conducted including 87 hospitalized patients (43 BD, 44 SCZ) aged 18–55 years, from March to June 2022. Socio-demographic data, personal and family psychiatric history, and validated scales (FACES IV, DAI-10, Birchwood Insight Scale, IEV) were collected. Analyses were performed using SPSS 22.0 (Chi-square, Kruskal–Wallis).
Results
Mean age was 37.3±9.9 years; male-to-female ratio 2:1. Most patients were single (70%), had primary education (45%), and were unemployed (64%); 50% lacked health coverage. Mean age at first hospitalization was 26.5±7.0 years; untreated psychosis in SCZ averaged 24 months. Annual relapse rate was 0.96. Significant predictors of relapse included age <35 (p<0.001), single/divorced status (p=0.012), substance use (78% smokers, 34% alcohol, 26% cannabis; p=0.007), DPNT >12 months (p=0.036), absence of consanguinity (protective; p=0.014), and poor inter-episodic interval quality (p=0.017).
Conclusions
Young age, single/divorced status, substance use, and prolonged untreated psychosis are major predictors of relapse. These findings highlight the need for early intervention, substance-use prevention, and targeted support for high-risk patients.
Disclosure of Interest
None Declared