SUMMARYAll patients in continuing care geriatric beds in a London Health District were screened for psychiatric morbidity and physical dependency in 1990. This sample was followed up 1 year after screening. Forty-four per cent (32/72) were dead at follow-up. Patients in hospital-based continuing care beds had higher mortality than those in health authority beds in a private nursing home. Dementia, physical dependency and depression measured by Depressive Signs Scale (DSS) were associated with mortality. Using multivariate analysis, to partial out independent effects, depressive signs score measured by the DSS was the only significant predictor of mortality. at index admission have all been shown to be associated with morality at 8-week (Ramsay etal., 1991) and at l-year follow-up (Finch et al., 1991). These predictors were, however, significantly intercorrelated and multivariate analysis revealed delirium to be the only independent variable to predict mortality. This is the first study to examine the relationship between psychiatric morbidity and subsequent mortality in continuing care geriatric inpatients.
METHODThe patients studied were followed up 1 year after inclusion in an investigation of psychiatric morbidity in all (n = 74) continuing care geriatric inpatients in a London Health District (Shah et al., 1992). Fifty-two patients were in hospital-based continuing care wards and 22 patients were in health authority beds in a private nursing facility.