Calibration of Accelerometer Output for Elderly Men: 2012: Board #141 June 3 8:00 AM - 9:30 AM

Ribeiro, Jose; Bohn, Lucimere; Oliveira, Norton; Oliveira, Jose; Carvalho, Joana; Santos, Maria; Mota, Jorge

Medicine & Science in Sports & Exercise:
doi: 10.1249/01.MSS.0000385065.09749.24
C-26 Free Communication/Poster - Objective Measures: Accelerometry and Pedometry: JUNE 3, 2010 7:30 AM - 12:30 PM: ROOM: Hall C
Author Information

University of Porto, Sports Faculty, Porto, Portugal.


(No disclosure reported)

Accurate assessment of quantity and intensity of physical activity (PA) is necessary when the relationship between practice and health benefits is being investigated. Different instruments have this aim. Accelerometers (ACL) supply information about the body acceleration, providing data about the amount, frequency and intensity of movement for extend periods of time. Most ACL studies were performed in children, adolescents and adult samples. The available literature contains different cut points to classify the level (intensity) of PA on those populations. However, in elderly people only recently ACL have been used as a direct method to assess PA and, when necessary, young adults cut points have been used to classify PA. This might probably incur in incorrect classification of PA intensity.

PURPOSE: To establish accelerometer count ranges corresponding to METs category for elderly, for moderate, vigorous and very vigorous PA.

METHODS: Data was obtained from 23 health elderly men (mean age 66.4 years old) during a treadmill exercise at 5 different speeds (4-6.5 km/h). Expired respiratory gases were collected, and oxygen consumption was measured on a breath-by breath basis using the a portable metabolic system, and METs was used as direct method to asses the intensity of exercise, together with the accelerometer.

RESULTS: Activity counts and MET values were highly correlated (r= 0.84), and count ranges corresponding to light, moderate, hard, and very hard intensity levels were ≤2083 (≤3 METs), 2084-4842 (3-6 METs), 4843-7600 (6-9 METs), ≥ 7601 (≥ 9 METs) counts/min., respectively. A model to predict activity counts from MET values was developed (R Square = 0.71, SEE = 1158.4 counts./min.). Our results show at 4 km/h, in treadmill, an average and standard deviation of 2467,66±619,52 counts/min.

CONCLUSIONS: These data will provide a template on which patterns of PA will be classified into intensity levels using a uniaxial accelerometer, for elderly people. It will be important to compare the patterns of PA using these ranges and the specific for young adults, in order to search for an underestimation of the results. Further studies are needed in order to assure that the 3 MET cut-off points is appropriate for elderly.

Grant: MCTES/FCT: PTDC-DES-104518-2008

© 2010 American College of Sports Medicine