The physical activity health paradox and risk factors for cardiovascular disease: A cross-sectional compositional data analysis in the Copenhagen City Heart Study

Abstract

Background: Studies indicate that physical activity during leisure and work have opposite associations with cardiovascular disease (CVD) risk factors, referred to as the physical activity health paradox. We investigated how sedentary behaviour and physical activity types during leisure and work are associated with systolic blood pressure (SBP), waist circumference (WC), and low-density lipoprotein cholesterol (LDL-C) in an adult general population sample using compositional data analysis.

Methods: Participants wore accelerometers for 7 days (right thigh and iliac crest; 24 h/day) and had their SBP, WC, and LDL-C measured. Accelerometer data was analysed using the software Acti4 to derive daily time spent in sedentary behaviour and physical activity types. The measure of association was quantified by reallocating time between sedentary behaviour and 1) walking, and 2) high-intensity physical activity (HIPA; sum of climbing stairs, running, cycling, and rowing), during both domains.

Results: In total, 652 participants were included in the analyses (median wear time: 6 days, 23.8 h/day). During leisure, the results indicated that less sedentary behaviour and more walking or more HIPA was associated with lower SBP, while during work, the findings indicated an association with higher SBP. During both domains, the findings indicated that less sedentary behaviour and more HIPA was associated with a smaller WC and lower LDL-C. However, the findings indicated less sedentary behaviour and more walking to be associated with a larger WC and higher LDL-C, regardless of domain.

Conclusions: During leisure, less sedentary behaviour and more walking or HIPA seems to be associated with a lower SBP, but, during work, it seems to be associated with a higher SBP. No consistent differences between domains were observed for WC and LDL-C. These findings highlight the importance of considering the physical activity health paradox, at least for some risk factors for CVD.

Fig 1. Formation of the final study population of eligible participants in the fifth examination of the Copenhagen City Heart Study.

N/n indicates number of participants. The sum across reasons for exclusion of non-eligible study participants exceeds 1367 since some participants fulfilled more than one exclusion criterion.

Fig 2

Adjusted estimated differences in systolic blood pressure (mm Hg, y-axis) given the reallocation of time between sedentary behaviour and A) walking, and B) HIPA among 652 adults. A negative value on the x-axis reflects the pairwise reallocation of time from sedentary behaviour to walking or HIPA, while a positive value reflects the reallocation of time from walking or HIPA to sedentary behaviour. The origin represents the reference composition (i.e., 372.7 and 234.2 min sedentary behaviour, 127.0 and 74.6 min standing, 49.1 and 22.5 min moving, 56.8 and 32.5 min walking, and 10.9 and 2.5 min HIPA, during leisure and work, respectively, and 457.3 min in bed). The difference in outcome was calculated by subtracting the estimated outcome of the reference composition from the estimated outcome for each reallocated composition. SBP is systolic blood pressure. HIPA is high-intensity physical activity (i.e., sum of climbing stairs, running, cycling, and rowing). Vertical lines correspond to the 95% confidence intervals.

Fig 3

Adjusted estimated differences in waist circumference (cm, y-axis) given the reallocation of time between sedentary behaviour and A) walking, and B) HIPA among 652 adults. A negative value on the x-axis reflects the pairwise reallocation of time from sedentary behaviour to walking or HIPA, while a positive value reflects the reallocation of time from walking or HIPA to sedentary behaviour. The origin represents the reference composition (i.e., 372.7 and 234.2 min sedentary behaviour, 127.0 and 74.6 min standing, 49.1 and 22.5 min moving, 56.8 and 32.5 min walking, and 10.9 and 2.5 min HIPA, during leisure and work, respectively, and 457.3 min in bed). The difference in outcome was calculated by subtracting the estimated outcome of the reference composition from the estimated outcome for each reallocated composition. WC is waist circumference. HIPA is high-intensity physical activity (i.e., sum of climbing stairs, running, cycling, and rowing). Vertical lines correspond to the 95% confidence intervals.

Fig 4

Adjusted estimated differences in low-density lipoprotein cholesterol (mmol/L, y-axis) given the reallocation of time between sedentary behaviour and A) walking, and B) HIPA among 652 adults. A negative value on the x-axis reflects the pairwise reallocation of time from sedentary behaviour to walking or HIPA, while a positive value reflects the reallocation of time from walking or HIPA to sedentary behaviour. The origin represents the reference composition (i.e., 372.7 and 234.2 min sedentary behaviour, 127.0 and 74.6 min standing, 49.1 and 22.5 min moving, 56.8 and 32.5 min walking, and 10.9 and 2.5 min HIPA, during leisure and work, respectively, and 457.3 min in bed). The difference in outcome was calculated by subtracting the estimated outcome of the reference composition from the estimated outcome for each reallocated composition. LDL-C is low-density lipoprotein cholesterol. HIPA is high-intensity physical activity (i.e., sum of climbing stairs, running, cycling, and rowing). Vertical lines correspond to the 95% confidence intervals.