Large-Scale Applications of Accelerometers

The study by Hagstromer et al. (2) in this issue of Medicine & Science in Sports & Exercise® is a recent example of the growing number of large-scale applications of accelerometers to measure physical activity. Since the late 1990s, publications of studies with accelerometer measurements have transitioned from describing data collected on tens of subjects to hundreds and now describe results from thousands of participants. The earliest large applications focused on youths. Examples include reports from the European Youth Heart Study (1), the Trial of Activity for Adolescent Girls (6), and the Avon Longitudinal Study of Parents and Children (4). Objective assessments of physical activity on large samples of adults and population-based samples are rarer, but more are coming soon. Objective data on physical activity have been collected on thousands of individuals from nationally representative samples in the United States (3), and are currently being collected in Canada (9). Objective data from population samples provide a new perspective and raise new questions. Hagstromer et al. (2) note that the amount of physical activity determined with an accelerometer was dramatically lower than that based on self-reports. Concern about the accuracy of self-reports is not new (7). However, the validity studies summarized by Sallis and Saelens (7) were relatively small. Accumulating information from large studies with accelerometers allows comparison with population estimates based on self-reports. Whereas self-report data tend to estimate approximately 23-44% European population adherence (8) to the current physical activity recommendations (5), accelerometer data suggest the figure may be less than 5% if sustained bouts of activity are necessary to meet the criterion (2). If self-report data dramatically overestimate levels of activity and accelerometer data reflect something closer to behavioral reality, physical activity researchers may need to revisit some foundations of current recommendations. The multiple benefits of physical activity are well established, but the epidemiological relationships and the resulting physical activity recommendations (5) rely heavily on self-reports. The observed differences between self-report and objective data raise questions about recommended duration, intensity, and need for bouts of physical activity. If overreporting is due to inflated estimates of activity duration, might it be possible that less than 30 min of moderate activity, as measured by accelerometer, conveys health effects associated with a self-report of 30 or more minutes? If intensity is misclassified by reporting lower intensity activity as moderate or greater intensity, might activity of less than 3.0 METs convey benefits that have been associated with reports of moderate intensity activity? Although focusing on bouts may aid recall of activity, are objectively measured bouts of 8-10 min required for health benefits? Insights into these questions may be possible from the cross-sectional population data now available (3), but real progress will depend upon integration of objective measures in prospective studies. Recent applications of accelerometers in population studies suggest that this progress is feasible. Richard P. Troiano National Cancer Institute Bethesda, MD

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.