To compare the degree to which four accelerometer metrics—total activity counts per day (TAC per day), steps per day (steps per day), physical activity energy expenditure (PAEE) (kcal·kg−1·d−1), and moderate- to vigorous-intensity physical activity (MVPA) (min·d−1)—were correlated with PAEE measured by doubly labeled water (DLW). Additionally, accelerometer metrics based on vertical axis counts and triaxial counts were compared.
This analysis included 684 women and 611 men age 43 to 83 yr. Participants wore the Actigraph GT3X on the hip for 7 d twice during the study and the average of the two measurements was used. Each participant also completed one DLW measurement, with a subset having a repeat. PAEE was estimated by subtracting resting metabolic rate and the thermic effect of food from total daily energy expenditure estimated by DLW. Partial Spearman correlations were used to estimate associations between PAEE and each accelerometer metric.
Correlations between the accelerometer metrics and DLW-determined PAEE were higher for triaxial counts than vertical axis counts. After adjusting for weight, age, accelerometer wear time, and fat free mass, the correlation between TAC per day based on triaxial counts and DLW-determined PAEE was 0.44 in women and 0.41 in men. Correlations for steps per day and accelerometer-estimated PAEE with DLW-determined PAEE were similar. After adjustment for within-person variation in DLW-determined PAEE, the correlations for TAC per day increased to 0.61 and 0.49, respectively. Correlations between MVPA and DLW-determined PAEE were lower, particularly for modified bouts of ≥10 min.
Accelerometer measures that represent total activity volume, including TAC per day, steps per day, and PAEE, were more highly correlated with DLW-determined PAEE than MVPA using traditional thresholds and should be considered by researchers seeking to reduce accelerometer data to a single metric.
1Department of Epidemiology and Biostatistics, School of Public Health, Indiana University, Bloomington, IN; 2Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA; 3Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA; 4Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD; 5Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD; 6Division of Cancer Prevention, National Cancer Institute, Bethesda MD; 7Pennington Biomedical Research Center, Baton Rouge, LA; 8Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University, Boston, MA; 9Friedman School of Nutrition Science & Policy, Tufts University, Boston, MA; 10Channing Division of Network Medicine, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA; and 11Department of Kinesiology, Recreation, and Sport Studies, The University of Tennessee, Knoxville, TN
Address for correspondence: Andrea K. Chomistek, Sc.D., Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington 1025 E. 7th Street, Room C101, Bloomington, IN 47405; E-mail: email@example.com.
Submitted for publication October 2016.
Accepted for publication March 2017.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.acsm-msse.org).