The mortality experience of people admitted to nursing homes

BACKGROUND: The objective of the study was to use routinely collected data to compare the mortality experience of people admitted to nursing homes from hospital and the community; and to describe changes in mortality over a 5 year period. METHODS: A retrospective cohort study and survival analysis were carried out. The subjects were 841 people admitted as Local Authority funded long-stay residents to Wakefield nursing homes between April 1993 and December 1997. Of these, 535 were admitted from hospital and 306 from the community. Subjects were divided into five admission year cohorts for analysis. The main outcome measures were comparison of survival curves for admissions from hospital and the community within each admission year cohort, comparison of survival curves for different admission year cohorts, and standardized mortality ratios (SMR) for all admissions (comparison with the population aged 65 years and over in England and Wales in 1995). RESULTS: Survival curves for admissions from hospital and the community were statistically significantly different for the 1993 admission cohort only. From 1993 to 1997, survival curves for all admissions demonstrate a steady improvement in mortality experience. SMR for all admissions from hospital was 606 per cent (95 per cent confidence interval (CI) 535.5-676.4 per cent). The difference between this and the SMR of 546.3 per cent (95 per cent CI: 457.7-634.9 per cent) for admissions from the community is not statistically significant. CONCLUSIONS: This study shows that the mortality experience of nursing home admissions from hospital was no worse than that of admissions from the community. Survival of people admitted to nursing homes in Wakefield has improved over the last 5 years. Possible reasons for these findings are discussed.