Skip to content
GLP Loss
← Research
Evidence

GLP-1 vs Orlistat: The 37% That Applied to a Fifth of the Trial

Orlistat's four-year trial cut diabetes incidence from 9.0% to 6.2%. Its own corrected conclusion says that difference was detectable only in the 21% with impaired glucose tolerance — while weight loss was the same in both subgroups.

Owen Castellanos9 min read
One headline, two different populationsXENDOS: 3,305 adults, orlistat or placebo, four yearsImpaired glucose tolerance21% of the trialDiabetes preventedYesWeight lost: 5.7 kgNormal glucose tolerance79% of the trialDiabetes preventedNot detectableWeight lost: 5.8 kgSame weight loss on both sides. The prevention is on one side.The quoted 37.3% risk reduction describes a fifth of the trial.Four years of orlistat: 5.8 kg. Placebo: 3.0 kg.Half the orlistat group and two thirds of the placebo group left.Figures from the published trial abstract, September 2026.

Orlistat is sold in a box labeled as an FDA-approved non-prescription weight loss aid, which no other drug in this comparison is. The prescription strength is 120 mg three times a day; a 60 mg version sits on the shelf beside the antacids. That accessibility is worth something real, and it is the reason the comparison with a weekly injection is asked constantly and answered badly — usually by quoting a diabetes-prevention figure that belongs to a fifth of the trial it came from.

The drug never leaves the gut

Every other drug in this comparison works by reaching the brain and the pancreas through the bloodstream. Orlistat does not enter the bloodstream in any meaningful quantity. It binds the active site of gastric and pancreatic lipases in the lumen of the stomach and small intestine, so dietary triglycerides are never broken down into absorbable form and pass through instead. At the recommended dose it blocks about 30% of ingested fat, with measured fecal fat excretion of 27% of dietary intake against 7% on placebo. Systemic exposure is described on the label as minimal, with plasma concentrations of intact drug near the limit of detection.[1]

That mechanism explains almost everything else about the drug: the size of the effect, the nature of the side effects, the vitamin instruction, and why a version of it is sold over a counter.

The effect size, and what it is being compared with

A meta-analysis of randomized, double-blind, placebo-controlled trials running a year or longer pooled 16 orlistat trials in 10,631 participants and found weight reduced by 2.9 kg against placebo (95% CI, 2.5 to 3.2). Attrition across the trials it reviewed averaged 30% to 40%.[2] The current over-the-counter Drug Facts label states the same thing in household terms: for every 5 pounds lost from diet alone, orlistat can help a person lose 2 to 3 pounds more, and most people in studies lost 5 to 10 pounds over six months.[3]

Against that, tirzepatide 15 mg produced −20.9% of body weight over 72 weeks in SURMOUNT-1 (95% CI, −21.8 to −19.9) from a mean baseline weight of 104.8 kg, with 57% of the top-dose group losing 20% or more.[4] Those are separate trials with separate participants, different durations and different eras of trial conduct, and no randomized comparison of orlistat against any incretin exists. The comparison is still not close: roughly three kilograms against roughly twenty percent of a hundred-kilogram body. The incretin figures are set out in the tirzepatide article.

The four-year trial, and the subgroup that carries it

Orlistat has something none of the newer drugs has: a four-year, double-blind, placebo-controlled trial. XENDOS randomized 3,305 adults with a body-mass index of 30 or above to orlistat 120 mg three times daily or placebo, on top of lifestyle changes. After four years, mean weight loss was 5.8 kg against 3.0 kg (P < 0.001), and the cumulative incidence of type 2 diabetes was 6.2% against 9.0%, a relative risk reduction of 37.3% (P = 0.0032).[5]

Then read the rest of the abstract. Of the participants, 79% had normal glucose tolerance at baseline and 21% had impaired glucose tolerance. The paper’s own corrected conclusion states that the difference in diabetes incidence was detectable only in the impaired glucose tolerance subgroup, while weight loss was similar whether a participant started with impaired tolerance (5.7 kg) or normal tolerance (5.8 kg).[5]

That single sentence changes who the headline is about. A reader with normal fasting glucose taking orlistat to avoid future diabetes is acting on a result the trial could not detect in people like them. A reader who has already been told their glucose tolerance is impaired is the person the 37.3% describes — and that is the population where every weight drug looks best, which is the point of the prediabetes article.

One more number keeps the four-year figure honest. Completion was 52% on orlistat and 34% on placebo, so the primary result describes a minority of those randomized. A second analysis carrying forward the baseline weight of everyone who dropped out gives 3.6 kg against 1.4 kg — the same direction, less than two thirds of the size.[5]

The tolerability trade is not the one people assume

Blocking fat absorption means the fat leaves by the only route available. The prescription label’s year-one rates, against placebo, are oily spotting 26.6% against 1.3%, flatus with discharge 23.9% against 1.4%, fecal urgency 22.1% against 6.7%, fatty or oily stool 20.0% against 2.9%, and fecal incontinence 7.7% against 0.9%.[1] These are not rare, and the last of them is the kind of event that ends a working day.

Two facts complicate the obvious conclusion. The first is that they fade sharply: the same table gives year-two rates of 4.4% for oily spotting, 2.1% for flatus with discharge and 1.8% for fecal incontinence, and the label notes that about half of all episodes lasted under a week, with first onset usually inside three months.[1] The second is that they do not drive people off the drug. Discontinuation for adverse events was 8.8% against 5.0% on placebo[1] — the same neighborhood as the 4.3%, 7.1% and 6.2% recorded in the three tirzepatide arms of SURMOUNT-1 against 2.6% on placebo.[4] The drug with the socially worst side effects and the drug with the largest effect size lose about the same share of their users to adverse events, in different trials decades apart.

There is also a control the incretins do not offer. Orlistat’s side effects are produced by dietary fat, so a low-fat meal largely prevents them, which the over-the-counter label states outright.[3] Nausea from a weekly injection does not work that way, as the side-effect article sets out.

The instructions that come with a lumen drug

Because fat-soluble vitamins travel with dietary fat, both labels require a daily multivitamin taken at bedtime, at least two hours apart from a dose.[1][3] Prescription orlistat is contraindicated in pregnancy, chronic malabsorption syndrome and cholestasis, and its warnings cover rare severe liver injury and oxalate nephropathy.[1] The over-the-counter box carries an organ transplant alert, because orlistat interferes with the drugs that prevent rejection, and tells anyone taking cyclosporine not to use it.[3]

The absorption interference extends to other daily tablets. The prescription label instructs that levothyroxine and orlistat be taken at least four hours apart and cyclosporine three hours after a dose, and a pharmacokinetic study found systemic exposure to amiodarone reduced by 23% to 27%.[1] The parallel problem with an injected drug is delayed stomach emptying rather than blocked absorption — a different mechanism with a similar consequence, covered in the levothyroxine article.

The most-studied weight drug is the weakest one

A systematic review published in 2025 identified 56 obesity pharmacotherapy trials enrolling 60,307 patients and counted them by drug: orlistat 22, semaglutide 14, liraglutide 11, tirzepatide 6, naltrexone-bupropion 5 and phentermine-topiramate 2. All of the drugs beat placebo significantly on total body weight loss, and only semaglutide and tirzepatide exceeded 10%.[6] Orlistat is not under-evidenced. It has been tested more than anything else in the field, and the answer has not moved.

The guidelines have drawn the obvious conclusion. The American College of Physicians living guideline of April 2026 names semaglutide and tirzepatide as first-line for adults with obesity, then phentermine-topiramate, liraglutide and naltrexone-bupropion; for adults with overweight and a weight-related condition it names semaglutide, tirzepatide and liraglutide.[7] Orlistat appears on neither list at any line, despite being the only agent on the shelf. Where the other non-incretin options sit is worked through in the phentermine comparison.

The prescription label is candid about the limit behind that omission: the long-term effects of orlistat on the morbidity and mortality associated with obesity have not been established.[1] A 2026 network meta-analysis of 262 trials found subcutaneous semaglutide to be the only drug in the field associated with reduced all-cause mortality (risk ratio 0.81; 95% CI, 0.72 to 0.93), an estimate drawn largely from cardiovascular outcome trials.[8]

Where orlistat wins

Price, and nothing else. An analysis of national drug price databases calculated an estimated minimum sustainable manufacturing price of $7 per 30-day course for orlistat, against estimated minimums of $40 for semaglutide and $50 for liraglutide — whose United States prices at the time were $804 and $1,418 for the same 30 days.[9] Two orders of magnitude separate the shelf product from the injection, and no access barrier stands in front of it: no prescriber, no prior authorization, no shipment.

What that buys is roughly a fifth to a sixth of the placebo-subtracted weight change an incretin produces, no established effect on long-term morbidity or mortality, and a diabetes-prevention result that the trial could only detect in people whose glucose tolerance was already impaired. For someone who cannot obtain anything else, that is not nothing. As a like-for-like alternative to an incretin it is not one, and the arithmetic of what a pound costs on each is in the cost-per-pound article and in the cost calculator.

Every incretin figure quoted here came from a trial of an FDA-approved product at a labeled dose. A compounded semaglutide or tirzepatide is not FDA-approved, and the FDA does not review it for safety, efficacy or quality before it is dispensed, so a compounded vial priced nearer to a box of orlistat does not thereby carry the trial result with it.

Frequently asked

How much weight does orlistat actually cause someone to lose?
About three kilograms more than placebo over a year or more. A meta-analysis of 16 randomized trials in 10,631 participants put the difference at 2.9 kg (95% CI, 2.5 to 3.2). The over-the-counter label says the same thing differently: for every 5 pounds lost from diet alone, orlistat helps a person lose 2 to 3 pounds more.
Does orlistat prevent diabetes?
In one population. XENDOS followed 3,305 adults for four years and found cumulative diabetes incidence of 6.2% on orlistat against 9.0% on placebo, a 37.3% risk reduction. The trial's own corrected conclusion states the difference was detectable only in the 21% of participants who began with impaired glucose tolerance, not in the 79% with normal glucose tolerance, even though both subgroups lost about the same weight.
Is orlistat's reputation for side effects deserved?
The rates are high and the discontinuations are not. In year one the prescription label records oily spotting in 26.6% against 1.3% on placebo and fecal incontinence in 7.7% against 0.9%, but the same rates fall to 4.4% and 1.8% in year two, and discontinuation for adverse events was 8.8% against 5.0%. Because the effects are produced by dietary fat, a low-fat meal largely prevents them.
Why can orlistat be sold without a prescription when GLP-1 drugs cannot?
Because it does not meaningfully enter the bloodstream. Orlistat inactivates lipases in the lumen of the stomach and small intestine so that roughly 30% of dietary fat is never absorbed, and the label describes systemic exposure as minimal, with plasma concentrations of intact drug near the limit of detection. Its risks are mostly local and its interactions are absorption interactions rather than systemic ones.
Is orlistat cheaper than a GLP-1 in a way that matters?
Yes, by roughly two orders of magnitude. An analysis of international price databases calculated an estimated minimum sustainable price of $7 per 30-day course for orlistat against $40 for semaglutide and $50 for liraglutide, at a time when United States prices for those two were $804 and $1,418 per 30 days. There is also no prescriber, prior authorization or shipment between the buyer and the 60 mg version.
Do any guidelines still recommend orlistat?
Not the current American College of Physicians guideline. Its April 2026 update names semaglutide and tirzepatide first-line for adults with obesity, followed by phentermine-topiramate, liraglutide and naltrexone-bupropion, and names semaglutide, tirzepatide and liraglutide for adults with overweight plus a weight-related condition. Orlistat appears at no line in either recommendation.

Sources

  1. [1] H2-Pharma LLC (2026). XENICAL (orlistat) capsules — prescribing information. DailyMed. Source
  2. [2] Rucker D, Padwal R, Li SK, et al. (2007). Long term pharmacotherapy for obesity and overweight: updated meta-analysis. BMJ. PMID 18006966
  3. [3] Haleon US Holdings LLC (2026). alli (orlistat) 60 mg capsules — over-the-counter Drug Facts label. DailyMed. Source
  4. [4] Jastreboff AM, Aronne LJ, Ahmad NN, et al. (2022). Tirzepatide Once Weekly for the Treatment of Obesity. N Engl J Med. PMID 35658024
  5. [5] Torgerson JS, Hauptman J, Boldrin MN, et al. (2004). XENical in the prevention of diabetes in obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes for the prevention of type 2 diabetes in obese patients. Diabetes Care. PMID 14693982
  6. [6] McGowan B, Ciudin A, Baker JL, et al. (2025). A systematic review and meta-analysis of the efficacy and safety of pharmacological treatments for obesity in adults. Nat Med. PMID 41039116
  7. [7] Qaseem A, Cross JT Jr, Harrod CS, et al. (2026). Pharmacologic Treatments With Lifestyle Modifications in Nonpregnant Adults With Overweight or Obesity in Outpatient Settings: A Living Clinical Guideline From the American College of Physicians (April 2026). Ann Intern Med. PMID 42296496
  8. [8] Nong K, Shi Q, Zeng L, et al. (2026). Comparative effects of drugs for adults with overweight or obesity: systematic review and network meta-analysis. BMJ. PMID 42419792
  9. [9] Levi J, Wang J, Venter F, et al. (2023). Estimated minimum prices and lowest available national prices for antiobesity medications: Improving affordability and access to treatment. Obesity (Silver Spring). PMID 36815242

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence