Join us on March 2nd for an ActiGraph webinar:
Wearable Data Gone Awry: Cautionary Tales from the Clinical Research Trenches.Register Now
Combined Triaxial Accelerometry and Heart Rate Telemetry for the Physiological Characterisation of Latin Dance
- Presented on May 31, 2013
Although objectively measured assessments of Latin social dancing have been previously reported, dance-specific accelerometer value calibrations for the estimation of energy expenditure (EE), step count (SC), and the development of physical activity intensity cut-points have not yet been conducted.
Purpose To value calibrate, cross-validate, and determine the reliability of a combined triaxial accelerometry and heart rate telemetry technique for characterizing the physiological responses to Latin dance
Methods Twenty-two non-professional Latin dancers attended two laboratory based dance trials each. After familiarization and a standardized warm-up, a multi-stage (three x 5 min) incremental (based on song tempo) Afro-Cuban salsa choreography was performed while following a video displayed on a projection screen. Data were collected with a portable indirect calorimeter, a heart rate telemeter, and wrist-, hip-, and ankle-mounted Actigraph GT3X+ accelerometers. Prediction equations for EE and SC were value calibrated using forced entry multiple regression and cross-validated using a delete-one jackknife approach with additional Bland-Altman analysis
Results The average dance intensity reached 6.1±1.0 kcal·kg-1·h-1 and demanded 46±11% of the heart rate reserve. Predictive ability of the derived models was satisfactory, where R2=0.80; SEE=0.4 kcal·kg-1·h-1 and R2=0.74; SEE=4 step·min-1 for EE and SC, respectively. Dependent t-tests indicated no differences between predicted and measured scores for both EE (t65=-0.25, p=0.80) and SC (t65=-0.89, p=0.38). The 95% limits of agreement for EE and SC were -1.0 to 1.0 kcal·kg-1·h-1 and -7 to 7 step·min-1, respectively. Dance-specific vector magnitude cut-points for the wrist-worn accelerometer were established at ≤74, 75-128, 129-237, and ≥238 count·s-1 for sedentary (i.e., not dancing), light, moderate, and vigorous intensity activity, respectively.
Conclusion Latin dancing to salsa music elicits physiological responses representative of moderate to vigorous physical activity, and a wrist-worn triaxial accelerometer with simultaneous heart rate measurement constitutes a valid and reliable technique for the prediction of EE and SC during Latin dance.