Two colleagues walking and talking outside an office building

Exercise Snacks and Blood Sugar: How Often to Break Up Sitting

10 min read

After a meal, your blood glucose rises and then falls as your body clears it. In laboratory trials, the rise was larger when people sat still for hours than when short bouts of movement broke the sitting up. Those bouts are often called exercise snacks.

This guide covers what the research found, which kinds of break worked best in the trials, how often to take them, and where the evidence still has gaps. Researchers call these short bouts activity breaks, and this article uses the same term.

39%
Lower glucose load over 8 hours when 3-minute walking breaks interrupted sitting every 30 minutes
Dempsey et al. 2016, Diabetes Care, 24 adults with type 2 diabetes. Calculated from 24.2 versus 14.8 mmol.h/L
36%
Lower insulin load with the same walking breaks
Dempsey et al. 2016, Diabetes Care. Calculated from 3,293 versus 2,104 pmol.h/L
37
Studies pooled in a 2020 meta-analysis of activity breaks
Loh et al. 2020, Sports Medicine meta-analysis
30 min
Breaks at least this often lowered glucose more than less frequent breaks
Yin et al. 2024, 9 trials for glucose, low certainty
15
Crossover trials in a 2026 analysis of activity breaks in sedentary adults
Han and Li 2026, Public Health meta-analysis
Sources: Dempsey et al. 2016, Loh et al. 2020, Yin et al. 2024, Han and Li 2026. Most of these trials ran in a laboratory over a single day.

Why moving muscle lowers blood glucose

Glucose enters a muscle cell through a transporter called GLUT4. At rest, much of the GLUT4 sits inside the cell. In a 2013 review in Physiological Reviews, Erik Richter and Mark Hargreaves described how muscle contraction moves GLUT4 to the cell surface, where it lets glucose in from the blood. Contraction sets off a signalling chain inside the muscle to do this.

Standing up and walking, or a set of squats, puts the large muscles of the legs to work. A single activity break triggers this movement of GLUT4 for a short time. Regular training also raises the total amount of GLUT4 in muscle, which the same review suggests partly contributes to better insulin action after training.

The trial behind the numbers

Paddy Dempsey and colleagues tested activity breaks in 24 inactive adults with type 2 diabetes and overweight or obesity. Their mean age was 62. Each person completed three 8-hour days in the laboratory, with standardised meals, in random order. On one day they sat without a break. On another they walked for 3 minutes every 30 minutes at a slow 3.2 km/h. On the third they did 3 minutes of simple resistance moves every 30 minutes: half-squats, calf raises, buttock squeezes and knee raises.

Both kinds of break lowered the glucose load across the day by a similar amount. Insulin and C-peptide, a marker of the insulin your body makes, fell too. Triglycerides, a type of blood fat, dropped only with the resistance breaks.

Glucose load across an 8-hour day Incremental area under the glucose curve, mmol.h/L, lower is better 24.2 Sitting, no breaks 14.8 Walking breaks 14.7 Resistance breaks Breaks lasted 3 minutes, every 30 minutes
Source: Dempsey et al. 2016, Diabetes Care. Randomised crossover trial in 24 adults with type 2 diabetes. Both break conditions differed from sitting with P below 0.001.

What the wider research shows

One trial in 24 people is a starting point. Several meta-analyses have since combined the wider body of activity break research, and they point the same way on glucose.

In 2018, Travis Saunders and colleagues analysed 20 studies in Sports Medicine. Compared with continuous sitting, regular activity breaks lowered glucose and insulin after meals but not triglycerides. In their analysis, the size of the drop did not change with the intensity of the breaks, the type of meal, or the age or body mass index of the people tested.

In 2020, Roland Loh and colleagues pooled 37 studies in the same journal. Activity breaks lowered glucose, insulin and triglycerides after meals, an effect the authors described as moderate. Unlike the earlier review, they found a larger glucose drop in people with a higher body mass index. They also compared breaks with one continuous bout of exercise of the same total energy cost, and found a small advantage for breaks on glucose but no clear difference for insulin or triglycerides.

A 2026 analysis by Mubai Han and Xiaotao Li in Public Health combined 15 crossover trials in sedentary adults. Activity breaks lowered glucose, insulin and triglycerides, though the authors noted the prediction interval for triglycerides included no effect. For glucose, more total activity across the day gave a larger drop, and the effect was similar after breakfast, lunch and dinner.

Which kind of break works best

Standing up is the smallest possible break, and it helped a little. In a 2022 meta-analysis of seven trials, Aidan Buffey and colleagues found standing breaks lowered glucose compared with continuous sitting but had no clear effect on insulin. Light walking breaks lowered both, and compared head to head, they lowered glucose and insulin more than standing breaks.

Walking breaks lowered glucose more Effect size for lower post-meal glucose versus continuous sitting, larger means a bigger drop 0.31 Standing breaks 0.72 Light walking breaks
Source: Buffey et al. 2022, Sports Medicine. Meta-analysis of seven one-day crossover trials, mostly in adults with overweight or obesity. Values are the size of the glucose difference from continuous sitting, shown without the minus sign.

A 2021 network meta-analysis of 30 crossover trials compared break intensities. Minghui Quan and colleagues found light and moderate activity breaks both lowered glucose and insulin after meals, and moderate breaks ranked highest in their comparison. The Saunders review, by contrast, found intensity did not change the size of the effect, so the evidence on intensity is mixed. On type of break, the Dempsey trial found walking and simple resistance moves gave almost identical glucose results. Walking and bodyweight moves both count.

How often to break up sitting

A 2024 meta-analysis by Mingyue Yin and colleagues compared break schedules across 13 crossover trials with 211 participants, nine of which measured glucose. Breaks every 30 minutes or more often lowered glucose more than breaks spaced further apart. Frequency made no clear difference to insulin, triglycerides or blood pressure, and the authors rated the certainty of all this evidence as low. The authors concluded interrupting sitting at least every 30 minutes "may be an ideal strategy to improve glucose control".

Where the evidence has gaps

Most of these trials lasted a single day in a laboratory, with set meals. Buffey and colleagues called for studies in everyday settings to test whether people keep the habit and whether the benefit lasts.

A 2026 meta-analysis of exercise snacks, the term this review used, adds a caution for people at metabolic risk. Xuanming Hu and colleagues combined eight trials with 159 participants. The breaks lowered glucose after meals, but results varied widely between trials, the prediction interval for glucose included no effect, and the drop in insulin did not reach statistical significance. The authors described them as a promising add-on that "cannot yet be recommended as a specific, evidence-based clinical prescription".

Two ways to break up a desk day

Both options below follow the Dempsey protocol: 3 minutes of activity every 30 minutes of sitting. Pick whichever suits your setting, or alternate them through the day.

Walking break
Walk for 3 minutes. A corridor loop, a trip to the kitchen or a lap of the office all count. The trial used a slow 3.2 km/h pace, so this does not need to feel like a workout.
Resistance break
Stand at your desk for 3 minutes of half-squats, calf raises, buttock squeezes and knee raises. This version also lowered triglycerides in the Dempsey trial.

A phone or watch reminder every 30 minutes makes the routine easier to keep. The Han and Li analysis found similar glucose benefits after breakfast, lunch and dinner, so breaks help after any meal of the day.

If you take insulin or other glucose-lowering medicines, speak to your GP or diabetes team before you add regular activity breaks.

Common mistakes

  • Standing up and staying still. Standing breaks helped less than light walking in the Buffey analysis, so move your legs during the break.
  • Spacing breaks far apart. In the Yin analysis, breaks every 30 minutes or more often lowered glucose more than breaks spaced further apart.
  • Treating activity breaks as a full exercise plan. They target glucose after meals over a single day, and the trials did not test long-term fitness or strength.
  • Changing activity around diabetes medicine without advice. Talk to your diabetes team first if you take insulin or other glucose-lowering medicines.

Frequently asked questions

What is an exercise snack?
A short bout of activity, usually a few minutes long, to break up a long stretch of sitting. Researchers often call it an activity break. In the Dempsey trial, each break lasted 3 minutes.
How often should I break up sitting?
The Yin 2024 meta-analysis found breaks every 30 minutes or more often lowered glucose more than breaks spaced further apart. The authors rated the certainty of this evidence as low.
Is walking better than standing?
In the Buffey 2022 meta-analysis, light walking breaks lowered glucose and insulin more than standing breaks. Standing breaks lowered glucose a little but had no clear effect on insulin.
Do squats work as well as walking?
In the Dempsey trial, 3-minute bouts of half-squats, calf raises and similar moves lowered glucose by almost the same amount as 3-minute walks. The resistance version also lowered triglycerides.
Do I need diabetes for this to help?
No. The Quan 2021 network meta-analysis included only adults without chronic diseases, and light and moderate breaks lowered glucose after meals in that group. The Dempsey trial focused on adults with type 2 diabetes.

The bottom line

Across several meta-analyses, short activity breaks during long sitting lowered blood glucose after meals compared with sitting still. Light walking beat standing, the evidence on intensity is mixed, and breaks every 30 minutes or more often lowered glucose more than less frequent ones, though that evidence is low certainty. Most trials lasted a single day in a laboratory, so the long-term effect remains unproven. A 3-minute walk or set of bodyweight moves every half hour is a low-cost habit, and in these pooled laboratory trials it lowered glucose after meals over the short term.

Build movement into your working day
DT Fitness London designs training plans around desk jobs, including short daily movement and structured strength work.
Book a consultation at dtfitnesslondon.com

Related reading

Sources

  • Dempsey PC, Larsen RN, et al. Benefits for type 2 diabetes of interrupting prolonged sitting with brief bouts of light walking or simple resistance activities. Diabetes Care 2016. pubmed.ncbi.nlm.nih.gov/27208318
  • Richter EA, Hargreaves M. Exercise, GLUT4, and skeletal muscle glucose uptake. Physiological Reviews 2013. pubmed.ncbi.nlm.nih.gov/23899560
  • Saunders TJ, Atkinson HF, et al. The acute metabolic and vascular impact of interrupting prolonged sitting: a systematic review and meta-analysis. Sports Medicine 2018. pubmed.ncbi.nlm.nih.gov/30078066
  • Loh R, Stamatakis E, et al. Effects of interrupting prolonged sitting with physical activity breaks on blood glucose, insulin and triacylglycerol measures: a systematic review and meta-analysis. Sports Medicine 2020. pubmed.ncbi.nlm.nih.gov/31552570
  • Quan M, Xun P, et al. Effects of interrupting prolonged sitting on postprandial glycemia and insulin responses: a network meta-analysis. Journal of Sport and Health Science 2021. pubmed.ncbi.nlm.nih.gov/33359636
  • Buffey AJ, Herring MP, et al. The acute effects of interrupting prolonged sitting time in adults with standing and light-intensity walking on biomarkers of cardiometabolic health in adults: a systematic review and meta-analysis. Sports Medicine 2022. pubmed.ncbi.nlm.nih.gov/35147898
  • Yin M, Xu K, et al. Optimal frequency of interrupting prolonged sitting for cardiometabolic health: a systematic review and meta-analysis of randomized crossover trials. Scandinavian Journal of Medicine and Science in Sports 2024. pubmed.ncbi.nlm.nih.gov/39630056
  • Hu X, Sang X, et al. Acute effects of "exercise snacks" on postprandial glucose and insulin responses in populations at metabolic risk: a meta-analysis with exploratory meta-regression and vascular outcomes. Frontiers in Endocrinology 2026. pubmed.ncbi.nlm.nih.gov/42558375
  • Han M, Li X. Acute effects of interrupting prolonged sitting with short bouts of physical activity on postprandial metabolism in sedentary adults: a multi-level meta-analysis. Public Health 2026. pubmed.ncbi.nlm.nih.gov/42594545
Back to blog