Response to: Does the Dissociated Effect of Physical Activity and Sedentary Behaviors on Overall Health Depend on Their Occupational or Leisure Setting?

Abstract

Reply: We thank the authors for their letter and review of our article. We are pleased that our paper has generated some thoughtful discussion on what we agree are important questions. At its basic level, our study begs the question of whether all physical activity is health-promoting. Based on the findings of our observational study, it seems occupational physical activity (PA) and sedentary behavior (SB) may not predict health as well as leisure PA and SB. However, we recognize the limitations of our study and the need for additional work to confirm these findings and inform targeted health promotion recommendations. First, it is necessary to understand the long-term effects of occupations that require high levels of SB and high levels of PA. In order to advance our understanding in this area, there is a need for prospective cohort studies that examine the temporal sequence between domain-specific SB and PA on a range of indicators of health over time. These studies should examine a range of health indicators including disease burden, health care expenditures, and health-related quality of life. Second, there is a need for more research to better understand the mechanisms explaining the relations between domain-specific SB and PA and these indicators of health. Experimental studies that examine a range of potential mechanisms are needed. For example, Holtermann et al1 proposed PA intensity level and insufficient rest/recovery might explain the differential effects of occupational and leisure PA. The rationale behind this suggestion is occupational PA is often performed at lower intensities but for long durations and with fewer opportunities for rest/recovery while leisure PA is often performed at higher intensities but for shorter durations and longer opportunities for rest/recovery. Experiments that control for intensity and volume of PA as well as duration of recovery are needed to confirm this proposal. It should be noted that the limited evidence in this area is due in part to the difficulty of measuring domain-specific PA and SB. For example, in our study, we used an ActivPAL monitor to measure steps per hour as a primary outcome for PA. While the ActivPAL is a good measure of SB, light PA and moderate PA which make up most occupational time, it is limited in its ability to capture upper body movements that are common is some occupations (eg, manufacturing).1,2 Additionally, best practice for quantify intensity (eg, heart rate, subjective intensity, objective measures [counts], etc.) of occupational PA remains unclear. Moreover, psychological mechanisms, such as low worker control,1 may contribute to the differential effects. Additional studies are needed that account for these methodological limitations in order to advance our understanding of appropriate interventions for individuals working in specific occupations. Health promotion experts in this area will also need to consider carefully the objectives of health advancing interventions. For example, individuals completing high levels of occupational PA and low levels of leisure PA may not benefit from increases in leisure PA. Rather an intervention that allows the individual to take more recovery breaks during work might be more useful for advancing their overall health. The rationale guiding these decisions should be based on a holistic model of health that considers health outcomes not often measured including stress, anxiety, mood, fatigue, burnout, pain, and quality of life. Additionally, we recommend these decisions be made collectively by stakeholders at numerous levels including employers, employees, and health promotion experts. We, again, thank the authors for their letter and hope this discussion contributes to growth in this field.