Measuring physical activity with the behavioral risk factor surveillance system

Because regular physical activity reduces the risk of premature death and disability, accurate methods of population-based measurement are important for public health surveillance efforts such as those based on the Behavioral Risk Factor Surveillance System (BRFSS). The present study: 1) briefly reviews and compares currently available methods to measure physical activity using BRFSS data, 2) describes physical activity patterns in the United States using these state-aggregated measures, and 3) provides suggestions on future directions for practitioners and researchers. Using a random-digit dialing, telephone survey, we collected data for noninstitutionalized adults aged 18 yr and older. We analyzed BRFSS data for 1996 from 50 states and the District of Columbia and Puerto Rico (N = 124,085). Based on recent literature and public health priorities, we developed eight different physical activity indices (one vigorous and seven moderate). These varied in their threshold for duration, kcal expenditure, and in frequency and intensity of activity.

Using different algorithms, the population prevalence of moderate physical activity ranged from about 20% to 38%. Only 20% of adults met the Healthy People 2000 definition for regular, sustained activity (> or =30 min of moderate activity per day for at least 5 d x wk(-1)).

Considerable progress is needed if the United States is to reach the current public health goal for regular physical activity. Standardized approaches to analyzing and collecting physical activity data are essential for public health surveillance, policy making, and communication to the public.