Assessment of physical activity by telephone interview versus objective monitoring

Purpose: To compare different methods of quantifying time in physical activity (PA). Methods: Twenty-five participants (12 male, 13 female) volunteered to be monitored for seven consecutive days, during which different PA patterns were measured by the simultaneous heart-rate motion sensor technique (HR+M). At the end of the 7th day, participants completed questions taken from the 2001 Behavioral Risk Factor Surveillance System (BRFSS) PA module telephone survey, in which they recalled the amount of time spent walking, and in moderate and vigorous activities. The results of the BRFSS PA module were then compared with those of the HR+M. Results: No significant group differences were found in the amount of time spent in moderate and vigorous activities between methods. However, individual differences were greater for time spent in moderate activities (SE +/- 7.36 min(.)d(-1); range -70 to 77 min(.)d(-1)) than for time spent in vigorous activities (SE +/- 3.57 min(.)d(-1); range -39 to 33 min(.)d(-1)). Spearman correlation coefficients between the HR+M and the BRFSS were significant for vigorous activities (r = 0.54, P < 0.01). There was 80% agreement between the two methods of classifying individuals who either: (a) met the recommendations (through moderate and/or vigorous PA) or (b) did not meet the recommendations. Conclusion: The BRFSS and HR+M methods yielded similar group estimates of PA, but individual assessments of moderate activity differed more than those of vigorous activity. BRFSS estimations of group compliance with national PA recommendations were similar to those of the HR+M.