Predictors of houseboundedness among elderly persons living autonomously in a rural community

Abstract

Aim: The housebound state is a risk factor for disability. This prospective study aimed to determine factors predictive of houseboundedness in the elderly, with an ultimate goal of preventing this condition. Methods: A self-report questionnaire pertaining to mental, physical and social status was administered to 732 communitydwelling elderly persons (313 men, 419 women;age range, 65-85 years) in October 2000. All subjects independently performed both basic and instrumental activities of daily living, went out alone for long distances, and did not use long-term care insurance. They were followed up until March 2003. "Housebound" was defined as leaving the house once a week or less. A stepwise multiple logistic regression model, adjusted for age, was used to identity factors predictive of houseboundedness. Data were analyzed on the basis of gender. Results: By the end of the follow-up period, 14.4% of men and 26.0% of women had become housebound. Stepwise multiple logistic regression analysis showed that predictive factors for men were lack of frequent contact with friends, neighbors and relatives;symptoms of lower limb pain; and self-assessed weight or muscle loss; and predictive factors for women were lack of frequent contact with friends, neighbors and relatives; lower limb pain;and self-assessed deterioration in health. Limited social contact and the presence of lower limb pain were common predictive factors for houseboundedness in both men and women. Conclusion: The findings from this study show that, among autonomous elderly persons, those who are socially isolated or who have physical pain are more likely to become housebound.