[Mortality survey of treated mentally ill patients. Methodological reflections].

All the studies performed in many industrialized countries have shown that the mortality-rates of mentally-ill patients are higher than expected, despite of a recent trend indicating a diminishing difference by comparison with control-groups (review in Corten et al). In France, the works by Casadebaig and Quemada focused on the mortality-rates of psychiatric in-patients from 1968 to 1982 cannot unfortunately be performed nowadays because of an administrative change in the nation-wide death-registration. This is the main reason for the purpose of a longitudinal study on this topic for the next years in our country. The chief goals are concerned with public health and with analytical epidemiology: defining the Standardized Mortality Ratios (SMR) globally and by cause of death for the most typical clinical groups. Methodology and feasibility. The main methodological issues have to be examined first before implementation. Theoretically, one should wait that a multicentric longitudinal survey could be representative of the whole population of french mentally in and out-patients. Actually, this study will be more likely performed with collaborative centers agreeing for data-collection; and secondly, it will be necessary to control that these collaborative centers do not differ significantly from the other public psychiatry-departments for the main variables that may influence the mortality-rates. The definition of cases and their identification will be done according to the International Classification of Disorders (ICD) considering its ninth and tenth revision. This research also needs to include a large number of patients because of statistical constraints.(ABSTRACT TRUNCATED AT 250 WORDS)