Independence in instrumental activities of daily living and its effect on mortality.

Factors relating to six-year mortality in a representative sample of seventy-year-old Jerusalem residents (N = 605) were investigated using logistic regression techniques. Around 16.3% of the study population died during the six-year post-interview period. Bivariate analysis found elevated mortality related to being male, having more than one IADL dysfunction, more than two ADL dysfunctions, financial problems, no social support in times of emergencies, bad self-rated health status, cognitive impairment, confinement to bed during the fortnight prior to interview, and lack of regular exercise. Logistic regression controlling for gender, various clinical diagnoses, financial state, social support and smoking status showed IADL (ROR = 4.57, 95% CI 1.51, 13.90), cognitive impairment (ROR = 3.99, 95% CI 1.85, 8.59) and having been bed-sick a week or more during the preceding fortnight (ROR = 6.60, 95% CI 1.00, 43.86) to be independent predictors of mortality. All persons who had a cognitive problem and were dysfunctional in more than two IADL categories, and 93.8% of persons who had been bed-sick and had more than one IADL dysfunction died during the study period. Combined measures of these three easily obtainable variables could prove a cheap and efficient method of identifying at-risk elderly persons in order to provide them with specific programs aimed at decreasing functional decline, and hence mortality.