This study has analyzed data on physical activity pattern in participants in the Baltimore Longitudinal Study of Aging to determine how physical activity levels change as people get older, whether and how they differ between men and women, and whether regular physical activity has increased over the last 15 years. Answers to these questions will help to understand how best to evaluate physical activity levels in older persons compared to younger persons, and in women compared to men. The most valid approach is necessary to determine the association of physical activity to health as people age. In addition, understanding of whether physical activity levels have increased over time is important in understanding the impact of public health campaigns in the U.S. This study has determined that physical activity levels decrease as people get older, and that older adults, both men and women, primarily obtain their physical activity from walking and housework. This information will help to change the questionnaires currently used for assessing physical activity. In the BLSA population, there has been a 21% increase in energy expenditure over a 15 year period. This significant increase indicates a major change in this important health habit. Recent analyses suggest that the secular trend of increasing physical activity in this cohort is found at all ages of BLSA men, and alters the decline in physical activity expected with aging.