RESUMO
Background: The average length of stay is often used to indicate health system efficiency; shorter stays are associated with reduced costs. In South Africa, mental healthcare expenditure is spent on inpatient care. Aim: To identify factors associated with a long stay in an acute psychiatric unit. Setting: A tertiary hospital. Methods: A case-control study review of inpatients diagnosed with psychotic symptoms was used. Sample was divided into two groups, length of stay (LOS) (LOS greater than 21 days, LOS less than 14 days). Total of 82 patients were divided into short stay group (SSG, n = 23) and long stay group (LSG) (n = 59). A comparison of demographic, clinical and system variables was conducted. Results: In demographics, LSG had fewer men compared to SSG (78.3%) and differed statistically from LSG with p = 0.05. Long stay groups were older in comparison to SSG with a p = 0.02. Illicit substance use in LSG was 44.1% and statistically less than SSG (73.91%; p = 0.02). A high proportion of LSG had medical or surgical and psychiatric comorbidities (67.8%) compared to SSG (43.5%) (p = 0.04). A total of 95% patients in SSG had family support. Conclusion: Longer stay was found to be associated with older females with primary psychotic disorders. Comorbidities with less availability of family support were associated with younger males presenting with psychotic symptoms that may be related to illicit substances that respond to rapid stabilisation. Contribution: Active surveillance of medical comorbidities amongst older female patients is necessary for early liaison services to reduce their length of stay.