Abstract
BACKGROUND: The risk factors for mortality after hospitalization in older persons are not fully understood. The aim of the present study was to examine the three-year (1,096 days) mortality in previously hospitalized older patients from rural areas, and to explore how objectively and self-reported health indicators at baseline were associated with mortality. METHODS: The study included 484 (241 men) medical inpatients with age range 65-101 (mean 80.7, SD 7.4) years. Baseline information included the following health measures: the Charlson Index, the Mini-Mental-State Examination, Lawton and Brody's scales for physical self-maintenance and the instrumental activities of daily living, the Hospital Anxiety and Depression scale, self-reported health (one item), and perceived social functioning (one item) and assistance in living at discharge. RESULTS: In all, 172 (35.5%) of those patients included had died within the three years of the follow-up period. Three-year mortality was associated with a high score at baseline on the Charlson Index (HR 1.73, 95%CI 1.09-2.74) and poor self-reported health (HR 1.52, 95%CI 1.03-2.25) in a Cox regression analysis adjusted for age, gender, other objectively measured health indicators, and perceived impaired social functioning. CONCLUSION: In a study of older adults admitted to a general hospital for a wide variety of disorders, we found co-morbidity (as measured with the Charlson Index) and poor self-reported health associated with three-year mortality in analysis adjusting for age, gender, and other health-related indicators. The results suggest that self-reported health is a measure that should be included in future studies.