Beyond Cut-points: Accelerometer Metrics that Capture the Physical Activity Profile
Commonly used physical activity metrics tell us little about the intensity distribution across the activity profile. The purpose of this paper is to introduce a metric, the intensity gradient, which can be used in combination with average acceleration (overall activity level) to fully describe the activity profile.Methods
1669 adolescent girls (sample 1) and 295 adults with type 2 diabetes (sample 2) wore a GENEActiv accelerometer on their non-dominant wrist for up to 7-days. Body mass index and percent body fat were assessed in both samples and physical function (grip strength, Short Physical Performance Battery, sit-to-stand repetitions) in sample 2. Physical activity metrics were: average acceleration (AccelAV); the intensity gradient (IntensityGRAD from the log-log regression line: 25 mg intensity bins (x)/time accumulated in each bin (y)); total moderate-to-vigorous physical activity (MVPA); and bouted MVPA (sample 2 only).Results
Correlations between AccelAV and the IntensityGRAD (r=0.39-0.51) were similar to correlations between AccelAV and bouted MVPA (r=0.48), and substantially lower than between AccelAV and total MVPA (r>0.93). The IntensityGRAD was negatively associated with body fatness in sample 1 (p<0.05) and positively associated with physical function in sample 2 (p<0.05); associations were independent of AccelAV and potential co-variates. In contrast, MVPA was not independently associated with body fatness or physical function.Conclusion
AccelAV and the IntensityGRAD provide a complementary description of a person’s activity profile, each explaining unique variance, and independently associated with body fatness and/or physical function. Both metrics are appropriate for reporting as standardised measures and suitable for comparison across studies using raw acceleration accelerometers. Concurrent use will facilitate investigation of the relative importance of intensity and volume of activity for a given outcome.