Common and gender specific factors associated with one-year mortality in nursing home residents.

Objectives: To identify common and gender-specific factors associated with mortality in two distinct nursing home (NH) populations: newly admitted (NA), and long-stay (LS) residents.

Design: A retrospective cohort study.

Setting: NH facilities in the state of New York.

Participants: A total of 59,080 NA female and 28,080 NA male NH residents, and 24,260 LS female and 8,928 LS male NH residents evaluated between June 1994 and December 1997 who were at least 65 years of age.

Measurements: Minimum Data Set information including measures of health, functional, cognitive, psychological, and social status.

Results: Multivariate proportional hazards regression results indicate that in NA residents, use of feeding tubes, bowel incontinence, and refuses fluids were associated with mortality in women only, whereas fever was associated with mortality in men only. Cancer and congestive heart failure (CHF) were more strongly associated with mortality in women than men. In LS residents, deterioration in communication, refuses fluids, use of indwelling catheters, and deterioration in cognition were associated with mortality in women but not men. Bedfast most of the time, use of new medications, and a balance problem were associated with mortality in men but not women. Shortness-of-breath was more strongly associated with mortality in women than men.

Conclusions: In both NA and LS residents, although men and women share many common factors associated with mortality, each gender has some unique factors associated with mortality. Furthermore, the strength of some common factors is significantly different across genders. These readily available data could be useful in making medical decisions and advance directive planning, and in the development of quality improvement initiatives and mortality prediction models.