Association Between Heart Rate And Accelerometry Measures Of Physical Activity In Older Adults

Abstract

Efforts to assess physical activity (PA) in older adults have primarily focused on using survey type instruments. There is a paucity of research on the effectiveness of objective measures such as accelerometry and heart rate in this population, particularly in field settings. PURPOSE To examine the efficacy of accelerometry as a PA measure in older adults in both lab and field settings, using heart rate as the criterion measure. METHODS Five women and 3 men ages 66±9 years wore uniaxial accelerometers at 3 sites (waist, leg, arm) and a heart rate monitor, while performing both lab and field activities. Lab activities included active sit (moving 4 weighted bottles in a predescribed manner, while seated), active stand (moving the same bottles in the same manner, while standing), and walking on a treadmill at 1.9 and 3.0 mph. These were selected based on the existing literature and to represent a range of activities typical to this population. Each activity was performed for 5 min, preceded by a 5 min rest. Measurement of field activities occurred during the subjects' regular, individualized, structured exercise routine. Frequency, duration, and type of activity were recorded via direct observation. Accelerometer counts and heart rate were recorded in 15 second epochs and averaged according to anatomical placement and activity in each setting. Pearson correlations were calculated to examine the association between accelerometry and heart rate, and significance was set at P<0.05. RESULTS Field activities were similar between subjects and categorized as warm-up (stretching), strength training, and cardiovascular. There were strong correlations among accelerometers at each anatomical placement in the lab (arm v leg, r = 0.66, P<0.05; arm v waist, r = 0.65, P<0.05; waist v leg, 0.90, P<0.05), but only the leg and waist sites were correlated in the field (r = 0.60, P<0.01). Significant associations occurred between heart rate and leg (r = 0.63, P<0.01), arm (r = 0.47, P<0.05), and waist (r = 0.58, P<0.05) during the lab activities; however, only leg and heart rate were weakly related in the field (r = 0.32, P<0.05). Correlations according to specific activities were poor, regardless of anatomical site or setting. A significant relationship occurred between arm and heart rate during lab active sit (r = 0.74, P<0.05), and there was a negative association between these measures during field warm-up activities (r = −0.97, P<0.01). CONCLUSIONS The results indicate that placement of the accelerometer on the wrist provides a reasonable assessment of low intensity PA. However, accelerometry may not be sufficient to capture overall PA during structured exercise in older adults.