If you’ve started on a GLP-1 medication and the conversation went well, your clinician probably warned you about nausea. Maybe vomiting. Maybe that the first few weeks would be the hardest, that it would settle, that taking it slowly with food would help.
What they probably didn’t talk much about is the side effect that, for a meaningful minority of patients, becomes the harder one to live with, and the one most likely to make people quit the medication weeks or months later. Constipation. Slow, dragging, uncomfortable, often worse at every dose escalation, and rarely the subject of the pre-prescription conversation it deserves.
This piece is about why that conversation gets skipped, why the standard advice often fails even for people already eating well, and what the evidence actually supports. The short version is that the strategy needs to be in place before the symptoms arrive, not after, and the most effective interventions are not the obvious ones.
The numbers, briefly, and why they’re undercounts
The published prevalence of constipation on GLP-1 medications looks moderate. In pooled clinical trial data across recent meta-analyses, around 8 percent of patients report it. Real-world data, including a 2026 analysis of 67,000 self-reporting users of semaglutide and tirzepatide on Reddit, puts the figure at approximately 15 percent of those who reported any side effect (around 14.9 percent for semaglutide users specifically). The PLOS Medicine review of GLP-1 safety published in early 2026 summarised the broader picture: gastrointestinal side effects “occur in up to half of patients” at peak, mostly during dose escalation.
The honest reading of these numbers is that they are floors, not ceilings. There are three reasons to think the real prevalence of constipation on GLP-1s is meaningfully higher than the published figures suggest.
First, the trial reporting infrastructure relies on patient-initiated disclosure. Patients underreport bowel function symptoms in clinical research the same way they underreport them in clinics. A patient going from a daily bowel movement to one every three days may not flag it on the symptom diary. The change has to feel like a problem before it gets reported, and for many people the gradual slowing doesn’t cross that threshold until well after the trial endpoint.
Second, real-world data like the Reddit analysis captures only the patients motivated enough to post about it. The majority of people taking these medications, including most of the older patients for whom GLP-1s are increasingly prescribed, never engage in online weight-loss communities. Their experience is invisible to that data.
Third, and the reason most worth being honest about, most people don’t know what good bowel function actually looks like. If you’re already going every two days, dropping to once every four or five days on a GLP-1 doesn’t feel like a step change; it feels like “my digestion has always been a bit slow.” It doesn’t get reported, because in the patient’s frame of reference it isn’t new. They may have spent years with a baseline that’s already poor by clinical measures and never recognised it as such.
By any honest reading, the figure is higher than 15 percent. How much higher is hard to say. In clinical practice, constipation appears to be among the more significant reasons patients eventually stop the medication, even when they’ve successfully lost weight. The numbers in the trial literature don’t reflect that.
Why the consult conversation skips it
There are three reasons constipation gets less time in the pre-prescription conversation than it should, and a fourth that has only become clear with the emerging neuroscience of how GLP-1s actually work.
The first is clinical. The fifteen-minute consult, in either GP or telehealth setting, has a finite agenda. The clinician needs to assess eligibility, walk through the major risks, talk about nausea management, and answer the patient’s questions, which are usually about whether the medication will work and what it costs, not about what it will do to bowel function in month four. Constipation gets a sentence. The patient nods and forgets.
The second is structural. Asynchronous prescribing models (the script-mill platforms) don’t have a real consult to skip the conversation in. Even where there’s a video appointment, the time pressure is severe. The standard intake form may include “any digestive issues” as a checkbox. The actual conversation about what to do if you stop having a bowel movement for four days at the higher dose almost never happens unless the patient asks.
The third is cultural, and worth being honest about. Bowel function is not a topic most adults discuss easily, even with a doctor. The gastroenterology literature has documented this for decades: patients underreport constipation, clinicians underprobe for it, and both sides find the conversation slightly uncomfortable. The pattern is most pronounced in older patients and women, where chronic constipation is significantly more prevalent and significantly more under-treated. By the time the patient mentions it, they’ve usually been suffering for weeks.
The fourth is more recent, and worth understanding because it changes how the standard advice should be framed. GLP-1 medications work in part by reducing the salience of food cues. The neuroscience now shows the same pathway dampens motivational signalling more broadly. The 2025 work on dopamine signalling in the ventral tegmental area found that semaglutide reduces motivation for rewards, not just food. Other research has shown the medications reduce cravings for alcohol, nicotine, and other addictive substances through the same mechanism.
The corollary, less discussed, is that the medication also seems to reduce the patient’s attention to background bodily signals. Patients on GLP-1s commonly describe a kind of motivational flattening: fewer cravings, less restlessness, less urgency about ordinary things. That’s largely the desired therapeutic effect. But it also means the low-level signals that would normally prompt someone to think “I haven’t been to the bathroom in a few days, I should do something” are themselves dampened. The constipation isn’t being noticed at the same rate it would otherwise be. Several patients in our reviewer conversations have described having to consciously remember to monitor bowel function, when in normal life it would have been automatic.
This is not laziness on the patient’s part. It’s the medication doing its job and inadvertently muting a category of self-monitoring most people take for granted. Once you understand the mechanism, the response changes. You don’t trust yourself to notice the problem in time. You build the strategy in advance.
The diet baseline problem
There’s a quieter problem with the standard advice that affects exactly the patients who’d be expected to do best.
The conventional recommendation (drink more water, eat more fibre) assumes the patient is moving from a poor diet to a better one. For someone who was eating fast food and bran cereal, there is real headroom. Adding leafy greens, switching to wholegrain bread, drinking water instead of soft drink, taking some psyllium: all of those interventions increase fibre and water intake significantly above where they were.
For someone already eating a plant-rich, low-processed diet, that headroom doesn’t exist. The fibre is already there. The water is already there. The advice has nowhere to go.
One Anova reader, vegetarian for years and almost no processed food, reported going from twice a day to once a week within weeks of starting a GLP-1. The dietary baseline was as good as the standard advice can produce. The medication slowed transit anyway, by an order of magnitude. This is one anecdote and not evidence on its own, but it points at a structural argument the trial literature supports: GLP-1 constipation is not primarily a diet problem. It’s a mechanism problem that diet can only partially offset. The medication slows gastric emptying. It slows colonic transit. It reduces total food and fluid volume by reducing appetite. None of those are addressable by improving an already-good diet.
The implication is uncomfortable but worth stating plainly. For people whose diets were already evidence-aligned, the standard advice is misleading because it suggests they can solve the problem through nutrition alone. They probably can’t. They can mitigate it. They can stop it from becoming severe. They can’t make it go away while remaining on the medication at therapeutic dose.
Why the standard advice often fails: the mechanism
GLP-1s work by slowing gastric emptying. That is one of their primary therapeutic mechanisms: food stays in the stomach longer, satiety lasts longer, the patient eats less. The same mechanism slows transit further down the gut. The colon, which depends on regular volume of stool moving through it to function well, gets less. This is not a side effect. It’s how the medication works.
At the same time, appetite drops. Total food intake falls. For most patients, that’s the point. The food intake that falls includes the fibre and fluid the colon needs to function. Even a “perfect” diet, one that hits the same proportional targets as before, produces less absolute fibre and water, because the patient is eating and drinking less of everything. The body is shedding the volume that used to keep things moving.
This is why “eat more fibre” can fail as advice for someone whose appetite has dropped 30 percent. They can’t eat more fibre; they can barely finish a normal meal. The replacement advice has to be density: denser fibre per gram of food, deliberately chosen highest-leverage sources, and fluid built into the routine rather than waited for.
There’s a second problem. Generic recommendations to increase fibre often produce the wrong kind of fibre. Bran cereal and high-fibre crackers can make GLP-1 constipation worse, not better, because insoluble fibre without sufficient fluid creates harder stools and more bloating. The fibre that helps with reduced-transit constipation is the soluble, gel-forming kind that retains water in the colon. The body of evidence on which sources work best is more specific than the standard advice acknowledges.
The plateau that isn’t a plateau
There’s a related effect that almost no consult covers, and that’s worth flagging because it changes how patients interpret what’s happening on the scale.
Healthy adults carry between half a kilogram and two kilograms of stool in the colon at steady state. When transit slows, that figure rises. A patient whose daily output was 200 grams and is now passing stool every five days is, mechanically, accumulating around a kilogram of additional gut content over a typical fortnight. That weight is real, and it sits on the scale.
The patient sees no movement on the scale during a dose escalation. They conclude the medication is plateauing. Some of them lose confidence and reduce adherence. Some increase the dose at the next escalation expecting more, and run into worse side effects.
What’s actually happening, in many of these cases, is that the fat loss is continuing as expected, but the retained gut content is masking it on the scale. Patients who address the constipation often see what looks like a “drop” in weight over a few days. It isn’t a sudden loss; it’s the clearing of a backlog that had been hiding the underlying loss for weeks. The medication was working all along.
This is one more reason the strategy matters. Constipation isn’t just uncomfortable. It distorts the signal the patient is using to evaluate the medication.
What the evidence actually supports
The dietary management of chronic constipation has been the subject of more controlled trials than most people would guess. Three interventions stand out in the recent literature.
Psyllium husk is the longest-studied and one of the most effective. Sourced from the seeds of Plantago ovata, it forms a gel in the gut, retains water, increases stool bulk, and accelerates transit. The therapeutic dose for constipation is roughly 7 to 12 grams per day, taken with adequate fluid. Well-tolerated by most people, though some experience bloating in the first week as the gut adjusts. Available in any pharmacy, cheap, and the easiest single intervention to add to a routine.
Prunes are the unsung classic. A 2014 randomised trial established that 50 grams per day of dried plums performed at least as well as, and on some measures better than, psyllium on stool frequency and consistency, and a 2023 meta-analysis confirmed the result holds across populations. The mechanism is a combination of soluble fibre, sorbitol, and phenolic compounds. The dose is achievable: about ten prunes a day. The taste is the only barrier.
The most recent and most surprising finding, and one that has changed how some gastroenterologists think about first-line dietary advice, is on kiwi fruit. A 2021 American College of Gastroenterology randomised trial compared two green kiwi fruit per day against psyllium and prunes in patients with chronic constipation. All three improved symptoms. Kiwi fruit produced the highest treatment satisfaction and the lowest adverse event rate. A 2022 New Zealand trial of two gold kiwi fruit per day, fibre-matched against psyllium, found kiwi at least as effective for stool consistency and easier for patients to stick with. Kiwi fruit also contains actinidin, a proteolytic enzyme that may aid digestion through mechanisms beyond its fibre content. The practical implication is that two kiwi fruit per day is a serious, evidence-based intervention, not a folk remedy.
Chia seeds and ground flaxseed work on similar principles to psyllium: gel-forming soluble fibre with substantial water-retention capacity. The evidence base is smaller but consistent. Two tablespoons a day of either, hydrated and stirred into a smoothie or yoghurt, contributes around 5 to 7 grams of fibre with very little eating volume required.
Water alone is not the answer, but water alongside soluble fibre is necessary. Dehydrated psyllium or chia produces worse outcomes than either taken with adequate fluid. The “drink more water” half of the standard advice is correct in its place.
A concrete morning strategy
The most practical implementation, for someone on a GLP-1 medication who wants to put the strategy in place before symptoms arrive, is a deliberate, fibre-dense morning intake.
A morning smoothie that combines a tablespoon of psyllium husk, two tablespoons of chia or ground flax, half a cup of frozen berries, a handful of leafy greens, a kiwi fruit, and a glass of water or unsweetened plant milk delivers around 12 to 15 grams of soluble, gel-forming fibre in a glass that requires almost no chewing, useful when appetite is low and food volume is hard. Followed by another glass of plain water within the hour.
If the smoothie isn’t appealing, the alternative path is two kiwi fruit at breakfast and a small bowl of stewed prunes, or oats with chia and berries. The principle is the same: front-load fibre when the stomach is empty and the appetite-suppression hasn’t kicked in for the day.
For people with already-established constipation, this is the foundation, not the cure. Add 10 to 12 grams of psyllium daily as a separate intake, ensure 2 litres of water spread through the day, and give the regimen at least two weeks. If symptoms haven’t shifted by then, the conversation moves to short-term laxative use under clinical supervision, and that conversation should be with a clinician who knows you’re on the GLP-1, not a pharmacist deciding from a packet.
A few things genuinely worth flagging to a clinician without delay: blood in stool, no bowel movement for more than five days despite the regimen, severe abdominal pain, vomiting in combination with constipation, or any new symptom that feels different from the gradual slowing the medication causes. These are not the standard pattern. They warrant a conversation, not a longer wait.
What this means in practice
The point of this piece is not to be alarmist about a side effect that, for most patients on GLP-1 medications, is manageable. It is to argue that the management is much easier when the strategy is in place from week one, before the dose escalations, before the appetite drop has reduced fibre intake, before the gradual slowing of transit has had time to compound, and before the medication’s own dampening of background bodily signals makes the patient slow to notice.
Three things follow from the evidence the public conversation has missed.
The published prevalence figures are floors. The real number is higher, particularly among patients whose baseline diet was poor enough that they don’t recognise the change as new.
The standard “drink water, eat fibre” advice fails predictably for people whose diets were already evidence-aligned, because the medication overrides the dietary lever. The strategy has to be density, not volume.
The constipation distorts the signal. Apparent plateaus on the scale during dose escalation are often just retained gut content masking ongoing fat loss. Addressing the constipation often clears what looked like a stall.
If your pre-prescription consult didn’t cover this, your consult was incomplete. That is not a criticism of the clinician; it is a criticism of a system that gives them fifteen minutes to discuss a medication that will change the patient’s life for the next two years. The gap is real. Filling it is, fortunately, mostly a matter of buying psyllium and kiwi fruit, and remembering, deliberately at first, to use them.
That is the conversation worth having early.