The Effects of Breaking up Prolonged Sitting Time

A Review of Experimental Studies

Fabiana Braga Benatti; Mathias Ried-Larsen


Med Sci Sports Exerc. 2015;47(10):2053-2061. 

In This Article

Beneficial Effects of Breaking up Prolonged Sitting Time—Evidence From Prospective Experimental Studies

In the past few years, a number of prospective experimental studies primarily aimed at evaluating the short-term and, to a lesser extent, long-term effects of interrupting prolonged sitting with different types of physical activity on parameters of the metabolic profile have been published. These studies are discussed in succeeding subsections and summarized in Table 1 and Table 2.

Short-term Studies

Dunstan et al.[9] were the first to demonstrate in a laboratory setting that when compared to 7 h of uninterrupted sitting, breaking up prolonged sitting time with 2-min bouts of light-intensity walking (i.e., ~3.2 km·h−1) every 20 min for 5 h during the postprandial phase reduced both the glycemic and the insulinemic responses to a liquid meal test in 19 physically inactive (i.e., insufficient amount of habitual MVPA)[1] nondiabetic overweight and obese middle-age subjects. Interestingly, when the participants walked at a moderate intensity (i.e., ~6.0 km·h−1), the positive outcomes were similar. Subsequently, in a subsample of patients, Howard et al.[13] reported that breaking up sitting time with either low- or moderate-intensity physical activity attenuated the increase in hematocrit, hemoglobin, and red blood cell count and the decrease in plasma volume observed during uninterrupted sitting, whereas the offsetting of increased fibrinogen levels only reached statistical significance for the low-intensity physical activity breaks. These results suggest that breaking up prolonged sitting may be of importance not only to improving glucose metabolism but also to counteracting the increased risk of thrombosis associated with excessive sitting. Furthermore, if exercise frequency is the same, its intensity does not seem to play a relevant role on postprandial glucose clearance and blood viscosity parameters in overweight and obese physically inactive persons.

Peddie et al.[21] also evaluated physically inactive young healthy normal-weight subjects and showed that bouts of 1 min and 40 s of light-to-moderate exercise (45%–60% V̇O2max) every 15 min during 9 h of sitting time lowered the postprandial insulin and glucose responses when compared with 9 h of uninterrupted sitting, suggesting that breaking up prolonged sitting may in fact affect glucose clearance in physically inactive subjects independent of BMI. Remarkably, the same improvement in insulin and glucose responses to meal tolerance tests was not observed when the exercise (matched for intensity and total duration) was performed in a 1-bout fashion before the sitting hours.

Similarly, Duvivier et al.[10] observed the lipid profile and the glucose and insulin responses to an oral glucose tolerance test in young physically inactive participants who underwent three different conditions: prolonged sitting condition (14 h·d−1 of sitting + 1 h·d−1 of walking + 1 h·d−1 of standing); increased light physical activity with a concomitant significant reduction in sitting time (5 h·d−1 of walking + 3 h·d−1 of standing + 8 h·d−1 of sitting); and 1-h MVPA and subsequent prolonged sitting (13 h·d−1 + 1 h·d−1 of walking + 1 h·d−1 of standing). Participants underwent each condition for 4 d and were evaluated on the fifth day. The authors demonstrated that the increased light physical activity protocol was effective in improving the lipid profile and insulin sensitivity when compared with the prolonged sitting condition. Importantly, in the MVPA condition, despite the comparable energy expenditure to the light-activity protocol, no improvements were observed. However, as the monitor used for matching energy expenditure (ActivePal) has not been validated for measuring energy expenditure per se, this could have introduced a bias.

Notably, Newsom et al.[20] reported that either moderate-intensity (i.e., 50% V̇O2max) or vigorous-intensity (i.e., 65% V̇O2max) exercise bouts set to expend approximately 350 kcal performed after 7 h of prolonged sitting did not induce any changes in glucose and insulin responses to a meal immediately after the exercise but similarly increased insulin sensitivity (as assessed by a hyperinsulinemic euglycemic clamp) on the next day when compared to 8 h of uninterrupted sitting in obese physically inactive adult subjects.

Accordingly, Kim et al.[16] showed that breaking up 9 h of prolonged sitting with either 1-h moderate-intensity exercise (i.e., 65% V̇O2max) or energy-matched hourly light-intensity walking (25% V̇O2max) induced lower triglyceridemic and glycemic responses to a high-fat meal on the next day in nonobese healthy recreationally active young subjects. However, in contrast to the results of Newsom et al.,[20] the glycemic response was lower after the moderate-intensity exercise when compared with the light-intensity exercise.

Altenburg et al.[3] evaluated young healthy male and female adults who underwent 8 h of prolonged sitting and, on a different occasion, 8 h of sitting with hourly breaks of 8-min moderate-intensity cycling (50%–60% of the heart rate reserve). In contrast to Dunstan et al.,[9] they did not observe any differences in the postprandial glucose and insulin responses between trials, despite lower C-reactive protein levels during the breaking-up sitting condition.

Similarly, Saunders et al.[22] showed that breaking up sitting (8 h) with 2-min low-intensity walks every 20 min, did not impact postprandial responses of lipids, glucose, and insulin when compared with the prolonged sitting trial in healthy young boys and girls (10–14 yr). When participants repeated the same protocol but also performed two bouts of 20-min moderate-intensity exercise, the same results were observed. Moreover, in a similar cohort (i.e., healthy adolescents), Sisson et al.[23] reported no differences in postprandial responses of glucose, insulin, lipids, and endothelial function between 3 h of uninterrupted sitting and breaking up prolonged sitting with three 45-min light-intensity (i.e., 2 METs) walks. It is worth noting that the subjects in these studies had normal weight, and because the habitual physical activity levels of the participants were not provided, they may have been physically active (i.e., sufficient amount of habitual MVPA).[1]

Altogether, these data suggest that in contrast to physically inactive subjects, in physically active subjects, (a) breaking up prolonged sitting may in fact have positive although delayed effects on the metabolic profile, and (b) a higher physical activity intensity or duration, independent of frequency, seems to be more effective in counteracting the detrimental effects of prolonged sitting.

When studying T2D subjects, Van Dijk et al.[28] showed that when compared with a prolonged sitting condition, both a 45-min moderate-intensity continuous exercise (~350 kcal expended) and three 15-min bouts of light-intensity activity (~175 kcal expended) throughout the day were effective in improving the postprandial glucose and insulin responses. Moreover, although both strategies led to improvements in the 24-h glycemic control, the improvement was greater and only reached statistical significance in the MVPA trial. These results suggest that although both light-intensity and moderate-intensity exercise are capable of improving postprandial glucose handling, the long-lasting effects of exercise on glucose homeostasis may occur in a dose-response manner, at least in patients with T2D. Moreover, it seems that in T2D subjects, one bout of MVPA would be sufficient to improve glycemic control. It is possible that T2D subjects may respond differently to different exercise stimulus than nondiabetic subjects. In T2D subjects, data suggest that AMP-activated protein kinase (AMPK) activation is more pronounced at higher exercise intensity compared to healthy lean individuals.[24] It could thus be speculated that higher intensity during the study by Van Dijk et al.,[28] when compared to the study of Peddie et al,[21] could explain the discrepancies. However, more studies are needed to confirm this hypothesis.

Thorp et al.[26] reported that alternating sitting and standing (i.e., sitting for 30 min and standing for 30 min) over an 8-h period during the postprandial phase for five consecutive days modestly but significantly reduced the glycemic but not the insulinemic response to a liquid meal test in overweight and obese physically inactive subjects. It is worth noting that during the standing time, the subjects were allowed to ambulate, which may have influenced the results, as light walking has been reported to positively affect postprandial glucose.[9,21]

Accordingly, Bailey and Locke[4] did not observe any positive effects of 2-min bouts of standing every 20 min on postprandial glucose in 10 normal to overweight participants when compared with 5 h of prolonged sitting. Interestingly, when the subjects underwent 2-min bouts of light walking every 20 min, the glucose response was effectively reduced when compared with the prolonged sitting condition. Once again, since the participants' physical activity level was not provided, it is possible that they were at least fairly physically active. If so, these results indicate that breaking up sitting time with standing may not be a stimulus sufficient enough to improve the metabolic profile in these subjects.

When studying adult desk-based office workers, Buckley et al.[7] observed a 43% lower postprandial glucose excursion and higher energy expenditure (0.83 kcal·min−1) with subjects working on a sit-stand desk workstation during 4 h when compared to 4 h of seated desk work. Furthermore, a tendency toward decreased glucose levels overnight after the standing when compared with the sitting day was also reported. Although the authors did not clearly report the amount of time spent sitting and standing in the two conditions of the subjects' physical activity level, these results do suggest that standing may be a stimulus sufficient enough to counteract the hazards of prolonged sitting in office workers.

Chronic Studies

In a 9-month prospective uncontrolled trial, John et al.[14] investigated the effects of introducing treadmill desk workstations for 12 overweight and obese adult office workers. The authors reported significant increases in standing (~2 to 3 h·d−1) and stepping time (~1 to 1.5 h−1) in detriment of sitting, in addition to significant decreases in waist and hip circumferences, LDL and total cholesterol, and glycosylated hemoglobin. Notably, these positive changes were observed despite no changes in dietary intake.

In a quasiexperimental study, Alkhajah et al.[2] investigated the effects of introducing sit-stand workstations in adult nonobese healthy office workers. After 3 months, the authors reported a significantly reduced time sitting by more than 2 h·d−1, which was almost exclusively replaced by standing in the intervention group when compared with the control group (i.e., no intervention). Although no differences were observed with respect to anthropometrics and fasting glucose, a significant increase in high-density lipoprotein cholesterol was observed in the intervention group when compared with the control group. It is worth noting that food intake was not controlled in this study, which may have affected the results.