[Frequent visitors to psychiatric emergency service: Demographical and clinical analysis].

INTRODUCTION: Frequent visitors of psychiatric emergency wards are an important health care problem. Previous studies underlined that 2 % to 9 % of patients induce 15 % to 33 % of total clinical activity. Those patients have chronic and severe mental illness such as schizophrenia, associated with social and financial difficulties.

OBJECTIVE: The aim of this study was to describe demographic and clinical characteristics of frequent visitors to a psychiatric emergency ward in a French Academic hospital over 6years in comparison to non-frequent visitors.

METHODS: The study is based on a retrospective review of the psychiatric emergency wards' administrative and medical computer databases; data that included demographic, financial, clinical, and management information. During this 6-year study, the psychiatric ward recorded 16,754 care episodes for 8800 different patients. We compared frequent visitors with other visitors using univariate and multivariate analyses. Frequent visitors were defined by a number of visits greater than 2 of the mean standard deviation.

RESULTS: Two percent of patients (n=192) had nine or more visits during the period. These patients caused 21 % of the total number of the visits. In the univariate analysis, the most significant reasons for referral in frequent visitors versus others (P<0.001) were: more frequent anxiety (37.6 % vs. 32.1 %), less frequent disruptive behavior (8.4 % vs. 12.9 %), depression (7.8 % vs. 17.2 %) and suicide attempt (4.5 % vs. 11.1 %). Factors associated with frequent visitors (P<0.001), after including all significant or confounding variables (multivariate analysis), were: schizophrenia and schizophrenia spectrum disorders (OR=29.5, IC: 11.4-76), DSM-IV cluster B personality disorders (OR=5.5, IC: 3.6-8.4), mental and behavioral disorders due to psychoactive substance use (OR=4.6, IC: 3.1-7), financial assistance through social government programs (OR range: 9.1-2.4, all significant) and being homeless (OR=2.7, IC: 1.8-4). Factors associated with non-frequent visitors were mood disorders (OR=0.07, IC: 0.03-0.19) and neurotic, stress-related, and somatoform disorders (OR=0.14, IC: 0.05-0.4). Sex and age were not significant in multivariate analysis.

DISCUSSION: This study identifies significant demographic and clinical factors associated with frequent visits in psychiatric emergency ward in accordance with the large majority of previous studies. We found that psychotic disorders or schizophrenia were the main diagnosis of these patients. Moreover, precariousness (homeless, financial assistance) is an important demographic factor associated with recurrence. However, contrary to numerous studies, we found no effect of sex or age. Due to this important economical and clinical burden, more specific care and alternative solutions to emergency care have to be proposed to this population of patients.