Physical Activity Patterns and Mortality: The Weekend Warrior and Activity Bouts

PURPOSE

This study aimed to examine how accelerometer-assessed physical activity accumulation patterns (e.g., is activity performed daily or only 1 or 2 d·wk or is activity accrued in bouts) may affect the association with mortality.

METHODS

Adults (N = 3438), age 40 yr and older, who wore an accelerometer (ActiGraph 7164), were drawn from the longitudinal follow-up of the National Health and Nutritional Examination Survey (2003-2006), a population-based survey of the United States. Accelerometer-assessed moderate-to-vigorous physical activity (MVPA) was described by activity patterns. Participants engaging in the majority of their activity on only 1 or 2 d·wk· were classified as "weekend warriors." Activity bouts were defined as a period of at least moderate intensity lasting at least 10 min. Bout characteristics included bout frequency and length. Mortality was assessed through National Death Index matching through 2013. Mortality rates were compared among groups with different activity patterns.

RESULTS

Over an average follow-up of 77.4 months, 394 deaths occurred. Compared to participants with <37.5 min of MVPA per week, those with greater amounts of activity had a 60% to 69% mortality rate reduction after adjusting for relevant covariables. Similar risk reductions were found when contrasting weekend warriors with those who were more frequently active. An increase of one MVPA bout per week was associated with a 13% increased mortality rate. Bout duration was not associated with mortality.

CONCLUSION

Physical activity is associated with decreased mortality rate, even among those who are active 1 or 2 d·wk.

Day length is associated with physical activity and sedentary behavior among older women

Physical activity may be influenced by one’s physical environment, including day length and weather. Studies of physical activity, day length, and weather have primarily used self-reported activity, broad meteorological categorization, and limited geographic regions. We aim to examine the association of day length and physical activity in a large cohort of older women, covering a wide geographic range. Participants (N = 16,741; mean (SD) age = 72.0 (SD = 5.7) years) were drawn from the Women’s Health Study and lived throughout the United States. Physical activity was assessed by accelerometer (ActiGraph GT3X+) between 2011 and 2015. Day length and weather information were obtained by matching weather stations to the participants’ location using National Oceanic and Atmospheric Administration databases. Women who experienced day lengths greater than 14 hours had 5.5% more steps, 9.4% more moderate-to-vigorous physical activity, and 1.6% less sedentary behavior, compared to women who experienced day lengths less than 10 hours, after adjusting for age, accelerometer wear, temperature, and precipitation. Day length is associated with physical activity and sedentary behavior in older women, and needs to be considered in programs promoting physical activity as well as in the analyses of accelerometer data covering wide geographic regions.

Daily physical activity patterns from hip- and wrist-worn accelerometers

Accelerometer wear location may influence physical activity estimates. This study investigates this relationship through the examination of activity patterns throughout the day. Participants from the aging research evaluating accelerometry (AREA) study (n men  =  37, n women  =  47, mean age (SD)  =  78.9 (5.5) years) were asked to wear accelerometers in a free-living environment for 7 d at three different wear locations; one on each wrist and one on the right hip. During waking hours, wrist-worn accelerometers consistently produced higher median activity counts, about 5  ×  higher, as well as wider variability compared to hip-worn monitors. However, the shape of the accrual pattern curve over the course of the day for the hip and wrist are similar; there is a spike in activity in the morning, with a prolonged tapering of activity level as the day progresses. The similar patterns of hip and wrist activity accrual provide support that each location is capable of estimating total physical activity volume. The examination of activity patterns over time may provide a more detailed way to examine differences in wear location and different subpopulations.