Walking After Meals: Ten Minutes and a Flatter Glucose Curve

“Take a walk after dinner” is old enough to be a proverb. A run of small randomised crossover trials has now actually tested it, and the proverb has mostly held up. Walking after a meal reliably shrinks the glucose rise that follows it; the same walk before the meal does much less. The effect is acute, modest, best documented in people who already have high post-meal glucose, and never followed for more than two weeks.

Timing: after the meal beats before it

The cleanest answer on timing is a 2023 meta-analysis in Sports Medicine by Engeroff and colleagues. It pooled eight randomised crossover trials — 116 participants, 47 of them with type 2 diabetes — in which the same people exercised before a meal, after a meal, or not at all. Exercise after eating lowered the post-meal glucose excursion compared with sitting (standardised mean difference 0.55) and compared with exercising before the meal (SMD 0.47). Exercise before the meal was not significantly different from doing nothing (SMD −0.13, confidence interval crossing zero). The gap between the meal and the exercise also mattered: the shorter the gap, the larger the reduction, which led the authors to conclude that walking does most when started “as soon as possible after a meal”, not after a delay and not before eating.

A second 2023 meta-analysis, by Kang and colleagues in Nutrients, pooled 31 studies in adults with overweight, obesity or type 2 diabetes and found post-meal exercise lowered glucose area under the curve (Hedges’ g −0.32) and 24-hour mean glucose (g −0.33) against time-matched no-exercise controls. Sessions longer than 30 minutes cut the area under the curve more than shorter ones. On start time it pointed the other way: the 24-hour benefit was larger when exercise began an hour or more after the meal, and larger in people with type 2 diabetes. The two disagree on when to start; they agree that after the meal works.

Ten minutes after each meal versus 30 minutes whenever

The largest trial here is Reynolds and colleagues, Diabetologia 2016. Forty-one adults with type 2 diabetes (mean age 60) followed two pieces of advice for two weeks each, in random order: walk 30 minutes a day whenever you like, or walk 10 minutes after each main meal. The three-hour post-meal glucose rise (incremental area under the curve) was 12 percent lower with the after-meal instruction (ratio of geometric means 0.88, 95% CI 0.78–0.99). After the evening meal — where carbohydrate intake peaked and sedentary time was highest — the reduction was 22 percent (0.78, CI 0.67–0.91).

Shambrook and colleagues (Journal of Science and Medicine in Sport 2020) compared three 10-minute walks, each starting 30 minutes after a meal, with one 30-minute walk after dinner, both at moderate intensity, in ten apparently healthy but insufficiently active adults. Both left post-dinner glucose up to 1.0 mmol/L lower than the control day, and the two were similar.

DiPietro and colleagues (Diabetes Care 2013) tested it in ten inactive adults aged 60 and over with fasting glucose of 105–125 mg/dL, a group the authors described as at risk of glucose intolerance. They did either three 15-minute treadmill walks after meals, or one 45-minute walk at 10:30 a.m. or 4:30 p.m., all at a fixed 3 METs, which the paper labels moderate. Post-meal walking lowered 24-hour mean glucose from 129 to 116 mg/dL; the sustained morning walk also helped (127 to 118). For the three hours after dinner, though, three short post-meal walks were significantly more effective than 45 minutes of sustained walking at either time of day, for the same total.

Does it work in people without diabetes?

The two trials with the largest reductions were in people with type 2 diabetes or raised fasting glucose, and Kang’s subgroup analysis found the effect larger in type 2 diabetes than without it. Three further small studies in healthy adults show the same direction.

Nygaard and colleagues (Applied Physiology, Nutrition, and Metabolism 2009) had 14 healthy women over 50 eat a carbohydrate-rich meal and then sit, walk slowly for 15 minutes, or walk slowly for 40 minutes. Fifteen minutes lowered glucose during the walk and delayed the peak; 40 minutes blunted the rise and reduced the two-hour incremental area under the curve. Longer walks gave larger reductions, and the women whose glucose rose most on the sitting day benefited most.

Bellini and colleagues (Nutrients 2022) had 21 healthy young volunteers walk briskly for 30 minutes after meals that varied in carbohydrate content and form. Walking cut the peak in both studies, and by a similar amount whatever the meal; across the full two hours after eating it did more after the lower-carbohydrate meal.

Hashimoto and colleagues (Scientific Reports 2025) gave 12 healthy young adults a 75-gram glucose drink followed by a 10-minute walk starting immediately, a 30-minute walk starting 30 minutes later, or sitting, at a comfortable pace. Both walks lowered the two-hour glucose area under the curve by roughly 5–6 percent, and the immediate 10-minute walk cut the peak from 182 to 164 mg/dL, while the later 30-minute walk’s peak of 176 was not significantly different from control. Ten minutes straight away matched 30 minutes started later on the area under the curve, and was the only condition that significantly lowered the peak.

How easy is “easy”?

Nobody in these trials was running. DiPietro’s walks were a fixed 3 METs; Nygaard’s were explicitly slow; Hashimoto’s were at whatever speed felt comfortable; Shambrook and Bellini used moderate or brisk walking. The effect appears across that range, which fits the literature on breaking up sitting.

Dunstan and colleagues (Diabetes Care 2012) gave 19 overweight or obese adults aged 45–65 a drink of 75 grams of glucose and 50 grams of fat, then tracked five hours of either uninterrupted sitting or sitting broken by two minutes of light or moderate walking every 20 minutes. Glucose area under the curve fell from 6.9 to 5.2 mmol/L·h with light breaks and 4.9 with moderate ones, and insulin fell by just under a quarter with either.

Loh and colleagues (Sports Medicine 2020) pooled 37 such trials: activity breaks versus continuous sitting lowered post-meal glucose (SMD −0.54) and insulin (SMD −0.56), with larger glucose effects at higher BMI. When total energy was matched, frequent breaks still edged out a single continuous bout on glucose (SMD −0.26). Buffey and colleagues (Sports Medicine 2022) split the breaks by type across seven studies: standing lowered post-meal glucose slightly versus sitting (Cohen’s d −0.31), light walking lowered it more (−0.72) and lowered insulin too (−0.83), and walking beat standing on both.

Put together, the trials used 10–15 minutes after each meal or 30–45 minutes as one bout, at anything from a slow to a brisk pace, started immediately or within about half an hour of eating, and in the one trial that compared meals the dinner walk produced the largest reduction. This is the easy end of the intensity scale; for what harder, steadier cardio does to fitness rather than to a single meal, see zone 2 training.

What muscle does with the glucose

Glucose enters muscle through GLUT4 transporters that move from storage depots inside the cell to the cell membrane. Muscle contraction itself sets that movement in motion, through signals including AMPK, calcium and nitric oxide synthase, as reviewed by Richter and Hargreaves in Physiological Reviews; glucose is a major fuel for working muscle. The plausible reading is that working muscle draws glucose out of the blood while the meal is still being absorbed, so less of it accumulates. That would explain the timing result — contraction-driven uptake can only flatten a spike it coincides with — and why the reduction is largest where the glucose response is largest: in type 2 diabetes (Kang), at higher BMI (Loh) and in the women with the biggest control-day rise (Nygaard).

The dinner finding sits alongside other evidence on meal timing, covered in gut circadian rhythm and meal timing. Unlike intermittent fasting, whose benefits mostly track eating less, this is a timing effect on its own: in the laboratory crossover trials the meals were identical across arms.

What the trials cannot tell you

Every study above is short — a day or two in a laboratory, or two weeks in Reynolds — and measures glucose, not outcomes. A lower post-meal curve is a surrogate; no trial shows that post-meal walking prevents anything. Sample sizes run from 10 to 41. Engeroff rated all eight of its trials at high risk of bias. The two meta-analyses disagree on start time. And the reduction is smaller in people without type 2 diabetes, even though every healthy-adult trial here found one.

Two practical notes. Anyone taking insulin or other glucose-lowering medication should talk to their prescriber before making meal-time walks a routine, because exercise around meals changes glucose in ways that may need a dose adjustment. And walking is not a substitute for anything a clinician has prescribed. For most people it is simply a low-cost habit with a consistent, replicated acute effect.

Bottom line

Walking after a meal lowers the glucose rise that follows it; in the pooled crossover trials, walking before the meal did not. The trials used 10–15 minutes at a slow-to-brisk pace after each meal, or 30–45 minutes as one bout, started immediately or within about half an hour of eating, and the one trial that compared meals found the dinner walk did most. The effect is well replicated in small crossover trials, largest in people whose glucose rises most, and unproven for long-term outcomes.