• Absolute level of physical activity: [ Time Frame: Assessed at baseline, 6, 12 and 24-month follow-up as kcals/wk expenditure based on survey self-report, validated against activity monitor data in a random sub-sample. ] [ Designated as safety issue: No ]
  
• Maintenance of physical activity [ Time Frame: Physical activity level relative to baseline assessed at all follow-ups 6, 12, 24 months. ] [ Designated as safety issue: No ]
  

• Population penetration: Proportion of those eligible who agree to join the study, and proportion of those randomized to the intervention condition who complete the intervention protocol. [ Time Frame: Final endpoint - 24 months ] [ Designated as safety issue: No ]
  
• Mediating factors [ Time Frame: Level of physical activity self-efficacy, social support for physical activity, and perceived benefits and barriers to activity will also be assessed at baseline and the 6, 12 and 24-month follow-ups. ] [ Designated as safety issue: No ]
  

The benefits of physical activity for adults are well established, but less than one-third of older adults in the U.S. have achieved recommended levels of physical activity despite widespread clinical recommendations to increase physical activity in recent years. Clinic-based approaches to increasing physical activity are expensive, difficult to implement in busy practice settings, and have limited reach. Moreover, evidence of the efficacy of such approaches is equivocal. A population based approach may be a more effective and less costly strategy to increase levels of physical activity in older adults. Population studies of physical activity have demonstrated that each year, many sedentary older adults initiate physical activity, but a nearly equal number of those who were active become sedentary. Among older adults initiating physical activity, only half continue to be active 3 months later.

We hypothesize that a population based approach that emphasizes physical activity maintenance can substantially increase physical activity levels in a defined populations of older adults. This hypothesis is tested in a 24-month randomized trial evaluating an innovative, theory-based behavioral intervention to maintain physical activity in a random sample of 50-70 year old adults who have recently become at least moderately active. One thousand (1000) subjects will be randomized to one of two experimental groups: 1) a "usual care" control group, and 2) an interactive phone- and mail-based intervention program tailored to maintaining physical activity in older adults. The primary outcome measures are: 1) physical activity, assessed as kcals/wk expenditure; and 2) physical activity maintenance, assessed as follow-up kcals/wk expenditure relative to baseline. Careful measurement of the penetration of the intervention into a well characterized older adult population, and the costs of the intervention, will be assessed. Psychosocial and behavioral mediators of physical activity maintenance will also be examined.

Study results will be relevant to policy makers, health promotion practitioners and health plans, and will provide practical information on the effectiveness, population penetration, and costs of an intervention designed to maximize population levels of physical activity among older adults.