Abstract
BACKGROUND: Due to demographic ageing and increasing life expectancy, a growing demand for long-term nursing home care can be expected. Stays in nursing homes appear to be more socially determined than hospital stays. We therefore looked at the impact of socio-demographic and health care variables on the length of the last nursing home stay. METHODS: Nationwide individual data from nursing homes and hospitals in Switzerland were linked with census and mortality records. Gender-specific negative binomial regression models were used to analyze N = 35,739 individuals with an admission age of at least 65 years and deceased in 2007 or 2008 in a nursing home. RESULTS: Preceding death, men spent on average 790 days and women 1250 days in the respective nursing home. Adjusted for preceding hospitalizations, care level, cause of death and multimorbidity, a low educational level, living alone or being tenant as well as a low care level at the admission time increased the risk for longer terminal stays. Conversely, a high educational level, being homeowner, being married as well as a high care level at the admission time decreased the risk for longer stays. DISCUSSION: The length of the last nursing home stay before death was not only dependent on health-related factors alone, but also substantially depended on socio-demographic determinants such as educational level, homeownership or marital status. The support of elderly people at the admission time of a presumably following nursing home stay should be improved and better evaluated in order to reduce unnecessary and undesired long terminal nursing home stays. CONCLUSIONS: Health policy should aim at diminishing the role of situational, non-health-related factors in order to empower people to spend the last years before death according to individual needs and preferences.