Factors associated with mortality in national health service nursing homes for elderly people and longstay geriatric wards in hospital

This article presents the analyses of mortality rates of elderly people in a randomized controlled trial of nursing home and long‐stay ward care in hospital. No statistically significant differences between settings were found in relation to mortality rates. Age, accident rate and existence of cognitive impairment were associated with mortality with bivariate analyses, although they were found to explain very little of the variation between survival and mortality when logistic regression analysis was employed. In both settings, those who had some cognitive impairment had just over twice the relative risk of dying than others, although this difference was not statistically significant.