Self-rated health as a risk factor for prescribed drug use and future health and social service use in older people.

BACKGROUND Self-rated health is an independent predictor of mortality in older people. Recently, the need to explore other health outcomes that may be predicted by global self-ratings of health has been identified. The aim of this work was to explore the use of self-rated health as a predictor for future health and social service use and for use of prescribed medication. METHODS 1042 community-dwelling people aged 65 and over living in Nottingham, United Kingdom, were interviewed in 1985, and survivors were reinterviewed in 1989 and 1993. Cox regression and logistic regression models were developed to see whether a self-rating of health was a predictor of 12-year mortality and of baseline, 4-year, and 8-year health and social service use and use of prescribed medication. RESULTS Baseline self-rating of health was an independent risk factor for 12-year mortality, and for having seen the general practitioner, community nurse, home help support in the month before the interview, and for increased medication use. Baseline self-rating of health was also a risk factor for 4-year and 8-year use of these services and increased medication use, although it was not significant when baseline service/medication use was included in regression models. CONCLUSIONS Self-rating of health not only predicts mortality, but is also useful for predicting long-term service and medication use among older people who live for a number of years.