Objectives: To explore associations between customary physical activity and three longitudinal outcomes: 12-year all-cause mortality, 12-year disease-specific mortality and 8-year change in general practitioner and personal social service use.

Design: Longitudinal study.

Subjects: 1042 people originally aged 65 and over randomly sampled from general practitioner lists in Nottingham, UK.

Methods: Cox regression survival and logistic regression analyses.

Main outcome measures: questionnaire-assessed levels of physical activity; 12-year mortality; reported health and personal social service contacts in month prior to interview.

Results: On the basis of factor scores derived from the interview questionnaire, activity levels were graded as high, intermediate or low, with respondents grouped accordingly. Relative to the high activity group, 12-year mortality was significantly increased in both the intermediate [adjusted hazard ratio (HR) = 1.53; 95% confidence interval (CI)= 1.10–2.14; P < 0.05] and low (HR= 1.75; 95% CI= 1.24–2.48; P <0.005) activity groups for men, and in the low activity group (HR= 1.73; 95% CI= 1.28–2.33; P<0.001) for women. Lower levels of activity were also associated with an increased likelihood of using health and personal social services 8 years after the initial interview, and an increased risk among men of having respiratory disease recorded as the primary cause of death. All models were adjusted for age, health and smoking status and weight category as measured at baseline.

Conclusions: The results are consistent with the conclusion that, among elderly people, health gain resulting from higher customary physical activity levels can promote a longer and more independent later life.