Development of an accelerometer age- and sex-specific approach based on population-standardized values for physical activity surveillance: A proof of concept

Abstract

Background: A shift from self-reports to wearable sensors for global physical activity (PA) surveillance has been recommended. The conventional use of a generic cut-point to assess moderate-to-vigorous PA (MVPA) is problematic as these cut-points are often derived from non-representative samples under non-ecological laboratory conditions. This study aimed to develop age- and sex- (age-sex) specific cut-points for MVPA based on population-standardized values as a feasible approach to assess the adherence to PA guidelines and to investigate its associations with all-cause mortality. Methods: A total of 7601 participants (20–85+ years) were drawn from the 2003–2004 and 2005–2006 National Health and Nutrition Examination Surveys (NHANES). Minutes per week of MVPA were assessed with a hip-worn accelerometer. Counts per minute (CPM) were used to define an age-sex specific target intensity, representing the intensity each person should be able to reach based on their age and sex. Age-sex specific MVPA cut-points were defined as any activity above 40% of the target intensity. These population- and free-living-based age-sex specific cut-points overcome many of the limitations of the standard generic cut-point approach. For comparison, we also calculated MVPA with a generic cut-point of 1952 CPM. Both approaches were compared for assessing adherence to PA guidelines and association of MVPA with all-cause mortality (ascertained through December 2015). Results: Both approaches indicated that 37% of the sample met the 150+ min/week guideline. The generic cut-point approach showed a trend to inactivity with age, which was less pronounced using the age-sex specific cut-points. Overall mortality rates were comparable using generic cut-point (hazard ratio (HR) = 0.61, 95% confidence interval (95%CI): 0.50‒0.73) or age-sex specific cut-points (HR = 0.57, 95%CI: 0.50‒0.66) for the entire sample. The generic cut-point method revealed an age- and sex-related gap in the benefits of achieving 150+ min/week of MVPA, with older adults showing an 18% greater reduction in mortality rates than younger adults, and a larger difference in women than in men. This disparity disappeared when using age-sex specific cut-points. Conclusion: Our findings underscore the value of age-sex specific cut-points for global PA surveillance. MVPA defined with age-sex specific thresholds was associated with all-cause mortality and the dose‒response was similar for all ages and sexes. This aligns with the single recommendation of accumulating 150+ min/week MVPA for all adults, irrespective of age and sex. This study serves as a proof of concept to develop this methodology for PA surveillance over more advanced open-source acceleration metrics and other national and international cohorts.

Keywords

• Physical activity
• Exercise
• Mortality
• Devices
• Movement sensors