An examination of the temporal and geographical patterns of psychiatric emergency service use by multiple visit patients as a means for their early detection.

BACKGROUND: Frequent users of the psychiatric emergency service (PES) place a heavy burden upon the mental health care delivery system. The aim of this study was to identify distinct temporal or geographical patterns of PES use by these patients as potential markers for their early detection. METHODS: Diagnostic profiles were obtained for patients making an intermediate (4 to 10) or a high (11 or more) number of visits to a general hospital PES in Montreal (Canada) between 1985 and 2004. Between-group comparisons were made with regards to several parameters. These included the time intervals between consecutive visits, visit clustering (single, repeating, and the time interval to the first cluster) and visits made to three other services where data was similarly acquired from 2002 to 2004. RESULTS: The two multiple visit groups differed with regards to diagnostic profiles and actual time between consecutive visits (significantly shorter in patients with 11 or more visits). Patients with 11 or more visits were more likely to have a single cluster (3 or more visits/3 months) or repeating clusters (4 visits/3 months) in their patterns of use. Personality disorders were more prevalent in patients with single clusters as they were, along with schizophrenia, in those with repeating clusters. In addition, clusters were found to occur sufficiently early so as to be potentially useful as markers for early detection. Ten percent of those with 11 or more visits and 16% of those with an intermediate number of visits frequented at least one other PES. A small number of patients, primarily those with substance abuse, made over 50% of their visits to other services. CONCLUSION: Temporal and geographical patterns of use differed significantly between the multiple visit groups. These patterns, combined with distinct diagnostic profiles, could potentially lead to the more rapid identification and treatment of specific sub-groups of multiple visit patients.

Demographic and clinical profiles of patients who make multiple visits to psychiatric emergency services.

OBJECTIVE: The study identified clinical and sociodemographic characteristics of patients making multiple visits to a psychiatric emergency service. METHODS: Information was obtained for patients visiting a hospital psychiatric emergency service in Montreal from 1985 to 2000. Profiles were determined for four groups: one visit, two visits, three to ten visits, and 11 or more visits. To determine whether the profile for those with 11 or more visits was generalizable, data for patients visiting the main site and three other such services from 2002 to 2004 were similarly analyzed. RESULTS: At the main study site (1985 to 2000), patients with single visits accounted for 36% of the 29,569 visits. The 292 patients with 11 or more visits accounted for almost 21% of total visits. Timing of the visit-time of day and day of the week-did not differentiate between groups. However, time itself was important in identifying patients with 11 or more visits: use of 30-month observation periods resulted in identification of only 8% of this group. Patients with 11 or more visits were more likely to be diagnosed as having schizophrenia and as having a comorbid diagnosis and were generally younger at the index visit and more economically impaired than those in the other groups. Overall, and at two of the three other sites, schizophrenia was overrepresented in the highest user group. CONCLUSIONS: Most visits to the psychiatric emergency service were made by frequent users who had distinctive profiles, which are potentially useful for developing clinical strategies to reduce the impact of this patient group on this service.