Age, period, and cohort effects on physical activity among elderly men during 10 years of follow-up: The Zutphen elderly study

Background. Data regarding the nature of change in physical activity as elderly people become progressively older are scarce. The present study describes changes in the physical activity pattern of a cohort of elderly Dutch men between 1985 and 1995. Methods. Self-reported physical activity was assessed with a reliable and valid questionnaire designed for retired men. In 1985, 863 men (aged 65-84 years) were examined, in 1990, 520 surviving men, and in 1995, 343 men. Three analytical perspectives (cross-sectional, longitudinal, and time-series) were used concurrently to untangle effects of aging, period, and birth cohort on the 10-year change in physical activity. Results. Mean total time spent on physical activity decreased by 33% (28 min/day) during 10 years of follow-up. Time spent on bicycling, gardening, and total activity decreased with aging. A period effect was observed for time spent on bicycling and total activity in 1990 (increase) and gardening in 1995 (decrease). No differences in physical activity between birth cohorts were observed. Time spent on walking remained stable during follow-up, but its relative contribution to total time spent on physical activity increased with aging. The pattern of change in total activity was not affected by functional status. Conclusions. Mean total time spent on physical activity by elderly men clearly decreased during follow-up. This could not be fully explained by declining functional status, but was partly explained by aging. In contrast to other physical activity parameters, time spent on walking was not affected by aging. These results suggest an increasingly restrictive physical activity pattern with aging.