Open Access Research article

Estimating relative intensity using individualized accelerometer cutpoints: the importance of fitness level

Cemal Ozemek1*, Heather L Cochran2, Scott J Strath3, Wonwoo Byun1 and Leonard A Kaminsky1

Author Affiliations

1 Clinical Exercise Physiology Program, Human Performance Laboratory, Ball State University, Muncie, IN, 47306, USA

2 IU Health Ball Memorial Hospital, Cardiopulmonary Rehabilitation Program, 2401 University Ave, Muncie, IN, 47303, USA

3 College of Health Sciences, University of Wisconsin-Milwaukee, PO Box 413, Milwaukee, WI, 53201-04133, USA

For all author emails, please log on.

BMC Medical Research Methodology 2013, 13:53  doi:10.1186/1471-2288-13-53

Published: 1 April 2013



Accelerometer cutpoints based on absolute intensity may under or overestimate levels of physical activity due to the lack of consideration for an individual’s current fitness level. The purpose of this study was to illustrate the interindividual variability in accelerometer activity counts measured at relative intensities (40 and 60% heart rate reserve (HRR)) and demonstrate the differences between relative activity counts between low, moderate and high fitness groups.


Seventy-three subjects (38 men, 35 women) with a wide range of cardiorespiratory fitness (maximal oxygen consumption (VO2max): 27.9 to 58.5 ml · kg-1 · min-1), performed a submaximal exercise test with measures of heart rate (HR) and accelerometer activity counts. Linear regression equations were developed for each subject to determine accelerometer activity counts for moderate and vigorous intensity physical activity corresponding to 40% and 60% of HRR. Interindividual variability of activity counts between subjects at both 40% and 60% of HRR was demonstrated by plotting values using a box and whisker plot. To examine the difference between absolute and relative activity cutpoints, subjects were categorized into 3 fitness groups based on metabolic equivalents (MET) (<10 MET, 10–13 MET, >13 MET).


At 40 and 60% of HRR, activity counts ranged from 1455–7520, and 3459–10066 counts · min-1, respectively. Activity counts at 40% HRR (3385 ± 850, 4048 ± 1090, and 5037 ± 1019 counts · min-1) and 60% HRR (5159 ± 765, 5995 ± 1131 and 7367 ± 1374 counts · min-1) significantly increased across fitness groups (<10 MET, 10–13 MET, and >13 MET, respectively).


This study revealed interindividual variability in activity counts at relative moderate (40% HRR) and vigorous (60% HRR) intensities, while fitness level was shown to have a significant influence on relative activity counts measured at these intensities. Individualizing activity count cutpoints may be more representative of an individual’s PA level relative to their fitness capacity, compared to absolute activity count cutpoints.

Physical activity intensity; Physical activity assessment; Accelerometer