The use of a case register to evaluate the costs of psychiatric care.

The aim of this study was to show the usefulness of adding cost information to a routine data collection provided by a case register for analysing the association between patients' characteristics and the direct costs of the psychiatric care actually provided. All patients (n = 706) who in 1992 had at least one contact with services which report to the South Verona psychiatric case register and who received an ICD-10 diagnosis were included in the study. The costs of specialist psychiatric care provided during the 365 days following the first contact in 1992 were calculated using a unit cost list. A bespoke software, linked to the case register, was designed to produce the individual information on costs directly. For each patient, costs were categorized as in-patient costs, sheltered accommodation costs, day-care costs, out-patient costs and community costs. All costs, grouped by service type, were found to differ significantly (P < 0.01) between diagnostic groups (affective disorders, neurotic and somatoform disorders, schizophrenia and related disorders, and other diagnoses). The multivariate analyses showed that costs are significantly higher for patients with a diagnosis of schizophrenia and related disorders than for patients belonging to the other diagnostic groups. However, only 6% of the variation could be explained by diagnostic group alone. On the other hand, between 40 and 50% of the costs of mental health care was predicted by patients' personal characteristics and other measures recorded on the case register. The results of the present study show that, where service use is monitored regularly, the base from which decisions on community provision and placement are made can be informed by careful analysis of routinely or easily available data on direct costs.