Analysis No. 03 · Evidence

Which kind of movement helps reflux?

My first analysis found sedentary posture right behind stress as a predictor of a bad day. I explained the stress half in the last post. This is the other half: what "moving" needs to mean before it helps, and the ways it can just as easily make things worse.

01 — The lever nobody's explained yet

In the first analysis, two things predicted a bad reflux day: stress (OR 1.96) and how sedentary I'd been (OR 1.64, p = 0.051). The second analysis gave stress a mechanism: it lowers the threshold at which reflux is felt, rather than producing more of it. Sedentary posture has been sitting there since July, unexplained. This post is that half of the story.

02 — Does the wider evidence agree?

Before asking why, the same sanity check as last time: does a much bigger dataset than mine see the same thing? A 2024 meta-analysis pooling 33 studies and 242,850 people found physical activity consistently associated with lower GERD risk — a relative risk of 0.74 for symptomatic reflux and 0.80 for diagnosed GERD, with the protective effect growing at higher activity levels.¹ People hitting roughly 150 minutes of activity a week had about 70% lower risk than the inactive group.

UK Biobank, followed for over a decade across more than 100,000 people, found the same direction: high-intensity activity independently lowered GERD incidence (HR 0.86).² A smaller regional meta-analysis found the mirror image: low activity, 22% higher odds of GERD.³ Three different study designs, same sign.

03 — The mechanism: a valve, and what sits on top of it

The physical story is simpler than the stress one. The lower esophageal sphincter is a valve between stomach and esophagus, and whether it holds depends partly on the pressure balance across it: how hard the valve is squeezing shut versus how hard the stomach's contents are pushing up against it. Body position changes both sides of that balance directly. Measured with a pressure catheter, sitting upright lowers resting valve pressure and raises pressure inside the stomach, compared with lying flat, in both healthy people and people with reflux esophagitis.⁴

Posture changes the mechanics of reflux directly, not just how healthy or unhealthy a day was overall. Whatever raises pressure inside the abdomen makes reflux more likely. Whatever lets gravity and normal gut motility do their job makes it less likely. That's the idea behind everything in this post.

04 — What helps, in practice

Two different claims get bundled under "movement helps." They're not the same claim. The first is chronic: being a generally active person, across weeks and months, is protective (section 02). The second is acute: doing something specific right after a meal helps that meal. The acute evidence is much thinner. In a controlled trial, an hour of walking after a standard breakfast gave a mild, short-lived reduction in acid exposure, and only in people with reflux, not in healthy controls. Gum chewing, tested the same way, had a stronger and more consistent effect in both groups.⁵ Walking after a meal is a reasonable habit. It is not, on the evidence, a strong lever on its own.

05 — Where movement backfires

The clearest myth-buster in this literature is that "exercise" is not one thing. A study that fitted matched groups of conditioned weightlifters, runners, and cyclists with pH probes found weightlifters had by far the most reflux, both fasted and after a meal, with runners in the middle and cyclists least affected.⁶ That ordering tracks how much each sport loads the abdomen: heavy lifting compresses the trunk hard, running jostles it rhythmically, cycling barely does either.

My own main workout is climbing, mostly bouldering and sport climbing. It doesn't fit neatly into any of those three studied categories, but it isn't cardio either: hard moves mean sustained core tension, held breaths on cruxes, and pulling the trunk into compressed positions, closer to the weightlifting end of that spectrum than to running or cycling. Nobody has run this exact study on climbers. I'm flagging it as an open question about my own routine, not a result. I don't have the meal-linked data yet to test whether a hard session shortly after eating behaves differently from an easy one.

Chronic training status complicates the picture further. In a small pilot study, people who habitually trained at low intensity had faster gastric emptying than people who habitually trained at moderate intensity, the opposite of the simple "more fitness, faster digestion" story.⁷ Thirty people, one pilot study, not something to over-read. It's a reminder that exercise intensity and habitual training status aren't the same variable.

06 — Why that catch matters for this project

My own posture variable is exactly the kind of measurement that study is warning about: a subjective, once-a-day rating, not a step count or an accelerometer trace. I can't rule out that "how active I felt" and "how active I was" drifted apart on some of my days the same way they did in that dataset. It doesn't invalidate the pattern below, but it's the honest ceiling on how far I can trust it.

07 — Testing it on my own data

I log daily position on a four-point scale: mostly moving, balanced, mostly sitting, or sitting almost all day. Post 1 collapsed this into a binary sedentary/active split. Looked at as a full gradient instead, it's sharper than that binary version let on.

179 days. Ordinal logistic trend: OR 2.20 per step up the ladder, p < 0.001.

What those four labels mean in practice: I work a desk job, so "sitting almost all day" still includes a 20-minute bike commute each way. It means no deliberate movement beyond that. "Mostly sitting" is the same desk day plus a climbing session in the evening. "Balanced" is a day with more movement built into the work itself, or a weekend that wasn't spent at a desk at all: a family outing, errands on foot. "Mostly moving" is a full day climbing outdoors, a hike, or a climbing trip, with no long desk block at all.

One thing this chart can't show honestly: workout and position are almost the same variable in my data, not two independent signals. On "sitting almost all day" days I logged a workout on 1 of 63. On "mostly moving" days it was 18 of 19. In between, the relationship is looser, but the correlation across all 179 days is strong (r = −0.63). Post 1's note that workouts showed no independent effect, once stress and sedentary posture were in the same model, makes sense in light of this: there's barely enough independent variation left in whether I worked out to give it a separate vote.

08 — Good, hurts, unclear

Collecting the evidence above into one place, graded rather than presented as a flat list of rules:

09 — What I'm doing differently

I'm not adding a formal exercise prescription. The acute, post-meal-specific evidence is too thin for that, and my own data can't isolate it yet either. What changed is smaller: I'm treating a long, unbroken desk block as its own risk factor, independent of whether I trained that day, and I'm more suspicious of hard climbing sessions scheduled soon after a big meal, even though I can't yet prove that combination is worse than an easy session in the same slot.

Whether it's the meal itself or the gap before lying down that matters more is a question I already have the timestamps to test. It's the subject of the next analysis.

Frequently asked

Does walking after eating help acid reflux?

A little. In a controlled trial, an hour of walking after a meal gave reflux patients a mild, short-lived reduction in acid exposure. It didn't do anything measurable in people without reflux. Chewing gum for the same hour had a stronger, more consistent effect in both groups.

Can exercise make GERD worse?

Yes, depending on the type and timing. A study that measured acid exposure across matched groups of weightlifters, runners, and cyclists found weightlifters had by far the most reflux. The pressure spike is the problem, not the exercise itself, and it's worst on a full stomach.

Is sitting all day bad for acid reflux?

Mechanistically, yes: measured directly, sitting raises pressure inside the stomach compared with lying down, and lowers the resting pressure of the valve that's supposed to keep acid down. At a population level, more leisure sitting (specifically TV time) has been linked to higher GERD risk, while overall physical activity is consistently protective.

How much physical activity actually helps reflux?

The clearest number comes from a meta-analysis of 33 studies covering nearly 243,000 people: hitting roughly 150 minutes of activity a week was associated with about a 70% lower risk of GERD compared with inactivity. It's an association, not causation, but a large and consistent one.

References

1. Yu C, Wang T, Gao Y, et al. Association between physical activity and risk of gastroesophageal reflux disease: a systematic review and meta-analysis. J Sport Health Sci. 2024.
2. Wang Z, Lin S, Liu H, et al. Association of healthy lifestyle behaviors with incident gastroesophageal reflux disease: a large-scale population-based prospective cohort study using UK Biobank. Prev Med Rep. 2025.
3. Albeshry AM, Althobaiti SF, Almerri S, et al. Physical activity as a modifiable risk factor for gastroesophageal reflux disease in Saudi Arabia: a systematic review and meta-analysis. Cureus. 2025;17(6).
4. Iwakiri K, Sugiura T, Kotoyori M, et al. Effect of body position on lower esophageal sphincter pressure. J Gastroenterol. 1999;34(3):305–309.
5. Avidan B, Sonnenberg A, Schnell TG, Sontag SJ. Walking and chewing reduce postprandial acid reflux. Aliment Pharmacol Ther. 2001;15(2):151–155.
6. Collings KL, Pierce Pratt F, Rodriguez-Stanley S, Bemben M, Miner PB. Esophageal reflux in conditioned runners, cyclists, and weightlifters. Med Sci Sports Exerc. 2003;35(5):730–735.
7. Matsuzaki J, Suzuki H, Masaoka T, Tanaka K, Mori H, Kanai T. Influence of regular exercise on gastric emptying in healthy men: a pilot study. J Clin Biochem Nutr. 2016;59(2):130–133.
8. Lu S, Zhu Y, Cui M, Guo Z, Li X, Song Y. Sedentary leisure behaviour, physical activity, and gastroesophageal reflux disease: evidence from a Mendelian randomization analysis. Health Sci Rep. 2025;8(3).