Dulaglutide

A laboratory-made copy of the gut hormone GLP-1, joined to a fragment of an antibody so that one injection lasts a week. It is sold as Trulicity and approved in the EU and the US for type 2 diabetes. Unlike semaglutide and liraglutide, neither the European nor the US regulator has approved it for weight loss, and its maker never ran a weight-management trial programme for it.

Sold as

Injection under the skin once a week, from a pre-filled pen.

Legal status in the EU

Authorised in the EU on 21 November 2014 (Eli Lilly Nederland B.V., EU/1/14/956), latest renewal 23 August 2019, prescription only. The approved use is type 2 diabetes in people aged 10 and above, as an addition to diet and exercise — alone where metformin is unsuitable, or on top of other diabetes medicines. There is no EU indication for weight management and no separate EU indication for the heart; the REWIND results sit in section 5.1 of the product information instead. The US approval, first granted in 2014, covers the same diabetes use plus a second one the EU does not have: reducing heart attacks, strokes and heart deaths in adults with type 2 diabetes who also have established heart disease or several heart risk factors. Dulaglutide appears nowhere in the 2026 WADA Prohibited List, and it is not on the 2026 WADA Monitoring Program either — that document lists markers of semaglutide and tirzepatide, in and out of competition, but not dulaglutide.

Dulaglutide switches on the GLP-1 receptor, which releases insulin only when blood sugar is high, turns down the hormone that raises blood sugar, slows how fast the stomach empties and reduces appetite. It is not a peptide chain. Both regulators describe the same structure: two identical chains held together by sulphur bridges, each made of a GLP-1 sequence about 90 per cent identical to the human one and joined by a short linker to a modified fragment of a human IgG4 antibody, grown in Chinese hamster ovary cells; the US label adds that the whole molecule weighs about 63 kilodaltons. That size is what stops the kidney clearing it quickly and gives it a half-life of 4.7 days. The sequence field is left empty because no one-letter string can describe a two-chain fusion protein of that size, and neither regulator prints a residue count that could be checked.

Part of the bodyWhat happensHow it is knownKind of evidence

Blood sugar and the pancreas

This is what the drug is approved for and what nearly all of its evidence is about. It tells the pancreas to release insulin — the hormone that pulls sugar out of the blood — but only while blood sugar is high, which is why on its own it does not cause the crashes that insulin injections can. It also turns down glucagon, the hormone that pushes sugar into the blood from the liver. The effect on fasting blood sugar arrives fast: most of it is there within two weeks.

Measured against dummy injections in adults already taking metformin. Over 26 weeks, long-term blood sugar fell by 1.22 percentage points on the 1.5 mg amount and by 1.01 on the 0.75 mg amount, against a rise of 0.03 on dummy injections, in 304, 302 and 177 people. In the 9,901-person heart trial the drop was much smaller because most participants started near target: 0.29 percentage points down at month 60 against 0.22 up on dummy injections.

Measured in people

Source [01]

Stomach and gut

The thing people notice and the reason most of those who stop, stop. Feeling sick, being sick and loose stools cluster in the first two weeks and then fade over the next four, after which the rate stays roughly flat. The stomach also empties more slowly, which is why it can slow down other tablets taken at the same time, and why anaesthetists are now told to consider that food may still be sitting there before an operation.

Counted against dummy injections in a pool of trials with 568 people on dummy injections, 836 on the 0.75 mg amount and 834 on 1.5 mg, average exposure 23.8 weeks. Any gut problem at all: 21% on dummy injections, 32% at 0.75 mg, 41% at 1.5 mg. Gut problems the investigators graded severe: 1.4%, 2.2% and 4.3%. Gut problems that made someone stop for good: 0.2%, 1.3% and 3.5%. In the 9,901-person heart trial over 5.4 years, 2,347 people (47.4%) reported a gut problem against 1,687 (34.1%) on dummy injections.

Measured in people

Source [02]Source [03]

Appetite and body weight

Appetite drops and some weight comes off, and this is the part of the record most likely to be misread. Dulaglutide is not approved for weight management anywhere, and the amount it takes off is small next to the drugs it shares a class with. The regulator's own summary of the entire trial programme is a range, not a number, and at the lower amount that range includes people who put weight on.

The European regulator summarises the whole programme like this: at 1.5 mg the change in body weight from start to final measurement ran from −0.35 kg to −2.90 kg across the studies, and at 0.75 mg it ran from +0.86 kg to −2.63 kg. Where there was a dummy-injection arm, the gap was of that size: −3.18 kg against −1.47 kg over 26 weeks on top of metformin, and −1.91 kg against +0.50 kg over 28 weeks on top of insulin. Added on top of an SGLT2 tablet, which itself takes weight off, the gap nearly vanished: −3.1 kg at 1.5 mg against −2.3 kg on dummy injections. Loss of appetite was reported by 8.6% at 1.5 mg against 1.6% on dummy injections.

Measured in people

Source [01]Source [02]

Heart and blood vessels

In a trial of nearly ten thousand people with type 2 diabetes, fewer of those on dulaglutide went on to have a heart attack, a stroke or to die of a heart cause. Most of the difference came from strokes rather than heart attacks. What makes this trial unusual in its class is who was in it: about two-thirds of the participants had never had a heart event, they were there because of risk factors. Blood pressure falls a little; the resting pulse goes the other way and speeds up slightly.

Measured in 9,901 adults with type 2 diabetes randomised to dulaglutide 1.5 mg or dummy injections and followed for a median of 5.4 years. 31.5% had established heart disease; 62.8% had risk factors only. The combined outcome happened to 594 people (12.0%) against 663 (13.4%). Blood pressure was measured separately with a 24-hour monitor in 755 adults: the top number fell 2.8 mmHg more than on dummy injections at 16 weeks, and the bottom number did not differ. Resting pulse rose by an average of 2 to 4 beats a minute; the product information gives no dummy-injection figure for that.

Measured in people

Source [01]Source [02]

Kidneys

Two opposite things, and they belong to different parts of the record. Over years, fewer people on dulaglutide started leaking large amounts of protein into their urine, which is the earliest sign that diabetes is damaging the kidney's filters. Over days, being sick and having diarrhoea dries a person out, and dried-out kidneys work badly — there are reports of kidney failure after exactly that, some of them needing dialysis.

The long-term finding comes from an exploratory analysis of the same 9,901-person trial, planned as part of a wider small-vessel outcome rather than as the trial's own question. A combined kidney outcome happened to 848 people (17.1%) on dulaglutide against 970 (19.6%) on dummy injections over a median of 5.4 years. The short-term harm is described in the product information as post-marketing reports, with no rate attached.

Measured in people

Source [01]Source [04]

Eyes

In people who already have damage to the small blood vessels at the back of the eye from diabetes, bringing blood sugar down can make that damage temporarily worse. The two regulators handle this differently, which is worth knowing before reading either one alone: the US label carries a whole warning section about it, and the European product information does not mention the eye anywhere.

Counted over 5.4 years in the 9,901-person heart trial, where it was set up in advance as part of a secondary outcome. Eye complications occurred in 1.9% on dulaglutide against 1.5% on dummy injections. Among people who already had eye damage at the start it was 8.5% against 6.2%; among people who did not, it was 1% against 1% — the whole difference sits in the group that already had the damage.

Measured in people

Source [02]

Ten randomised phase 3 trials in 8,035 adults with type 2 diabetes sit behind the approval, plus a cardiovascular outcome trial, REWIND (NCT01394952, Lancet 2019), in 9,901 adults followed for a median of 5.4 years. In REWIND the combined outcome of heart attack, stroke or death from a heart cause occurred in 594 of 4,949 on dulaglutide (12.0%) against 663 of 4,952 on placebo (13.4%); hazard ratio 0.88 (95% CI 0.79–0.99, p=0.026). Most of the difference was stroke, 158 (3.2%) against 205 (4.1%), hazard ratio 0.76. Deaths from any cause were 536 (10.8%) against 592 (12.0%), hazard ratio 0.90 (95% CI 0.80–1.01, p=0.067) — no difference. REWIND is the one trial in this class run mostly in people who had never had a heart event: 31.5% had established cardiovascular disease and 62.8% had risk factors only. An exploratory analysis of the same trial found a combined kidney outcome in 848 (17.1%) against 970 (19.6%), hazard ratio 0.85, almost all of it new large protein leak into the urine. For blood sugar, on top of metformin over 26 weeks, HbA1c fell 1.22 percentage points at 1.5 mg against a rise of 0.03 on placebo. Weight is the part the register must not let the GLP-1 halo inflate: the European regulator summarises the whole programme as a change from −0.35 kg to −2.90 kg at 1.5 mg and from +0.86 kg to −2.63 kg at 0.75 mg, and the cleanest placebo comparison is −3.18 kg against −1.47 kg over 26 weeks. In the head-to-head trial against semaglutide (SUSTAIN 7, Lancet Diabetes Endocrinol 2018, 1,201 participants, 40 weeks) semaglutide 1.0 mg gave −6.5 kg against −3.0 kg for dulaglutide 1.5 mg, and −1.8 against −1.4 percentage points of HbA1c. In 154 children and adolescents (AWARD-PEDS) HbA1c fell 0.8 percentage points against a rise of 0.6 on placebo while body mass index did not separate from placebo at all. The sequence field is empty because the administered molecule is a two-chain, roughly 63-kilodalton fusion protein made in hamster cells, not a synthesised chain, and writing any one-letter string would describe a different molecule; neither regulator's document states a residue count, so the chain length is left empty too.

Slow to arrive, slow to leave, and one injection a week is the whole design. The molecule is a copy of GLP-1 stuck to a fragment of an antibody, which makes it far too big for the kidney to clear quickly and shields it from the enzyme that destroys natural GLP-1 within about two minutes. The result is a half-life of 4.7 days. After an injection the level peaks at about 48 hours, and it takes two to four weeks of weekly injections before the level stops climbing and settles. Blood sugar starts moving before that: most of the effect on fasting blood sugar is there within two weeks, and the drug lowers glucose from the very first injection. The gut effects run on their own clock — they peak in the first two weeks, fall away over the next four, and then stay at a roughly constant low rate. Weight moves slowly and does not run away: in the trial that ran to 52 weeks at three different amounts, the curve was still drifting down at the end, having reached 3.5, 4.3 and 5.0 kg. Nobody has published what happens to weight after someone stops.

What was measuredWhat happenedIn whomHow it is known

Heart attacks, strokes and deaths from heart causes

594 of 4,949 people on dulaglutide had one of these (12.0%) against 663 of 4,952 on dummy injections (13.4%) — a hazard ratio of 0.88, meaning about an eighth fewer. Taken apart: deaths from heart causes 317 (6.4%) against 346 (7.0%); heart attacks that were not fatal 205 (4.1%) against 212 (4.3%); strokes that were not fatal 135 (2.7%) against 175 (3.5%). Only the stroke component had a range that excluded no difference.

9,901 adults with type 2 diabetes, average age 66, average body mass index 32.3, median long-term blood sugar 7.2%; 31.5% had established heart disease and 62.8% had risk factors only; median 5.4 years

Measured in people

Source [02]Source [03]

Deaths from any cause

536 of 4,949 people on dulaglutide died (10.8%) against 592 of 4,952 on dummy injections (12.0%). The hazard ratio was 0.90, and its range ran from 0.80 to 1.01 — it includes no difference at all. The authors state plainly that all-cause mortality did not differ between the groups.

The same 9,901 adults with type 2 diabetes, median 5.4 years; vital status was known for 99.7% of them

Measured in people

Source [03]

Strokes, looked at on their own

158 of 4,949 people on dulaglutide had a stroke (3.2%) against 205 of 4,952 on dummy injections (4.1%) — a hazard ratio of 0.76. The reduction was in the kind caused by a blocked vessel (hazard ratio 0.75); the kind caused by a bleed showed nothing (1.05, in a range from 0.55 to 1.99). Disabling strokes were also fewer, 0.74. How badly disabled people were afterwards did not differ between the groups.

The same 9,901 adults, median 5.4 years; strokes were classified and their severity scored by an independent panel

Measured in people

Source [05]

Kidney damage

A combined measure counted starting to leak large amounts of protein into the urine, losing 30% or more of kidney filtering ability, or needing long-term dialysis or a transplant. It happened to 848 of 4,949 people on dulaglutide (17.1%) against 970 of 4,952 on dummy injections (19.6%) — a hazard ratio of 0.85. Almost all of that came from the protein leak, 0.77. Losing filtering ability came out at 0.89 and dialysis or transplant at 0.75, and both of those ranges include no difference. This was an exploratory analysis, not the question the trial was built to answer.

The same 9,901 adults, of whom 791 (7.9%) were already leaking large amounts of protein at the start; median 5.4 years, 51,820 person-years in total

Measured in people

Source [04]

Long-term blood sugar, against dummy injections

Over 26 weeks on top of metformin, long-term blood sugar fell 1.22 percentage points at 1.5 mg and 1.01 at 0.75 mg, against a rise of 0.03 on dummy injections. The share reaching a blood sugar target under 7.0% was 60.9% and 55.2% against 21.0%. A daily sitagliptin tablet in the same trial gave 0.61 percentage points and 37.8%.

1,098 adults with type 2 diabetes on metformin — 304 on 1.5 mg, 302 on 0.75 mg, 177 on dummy injections, 315 on sitagliptin; the trial ran 104 weeks and the dummy group ended at 52

Measured in people

Source [01]

Body weight, against dummy injections

On top of metformin over 26 weeks, weight fell 3.18 kg at 1.5 mg and 2.63 kg at 0.75 mg, against 1.47 kg on dummy injections. On top of insulin over 28 weeks it fell 1.91 kg against a gain of 0.50 kg. On top of an SGLT2 tablet over 24 weeks the difference was much smaller: 3.1 kg at 1.5 mg against 2.3 kg on dummy injections. The regulator's summary of the whole programme is a range rather than a figure — from −0.35 kg to −2.90 kg at 1.5 mg, and from +0.86 kg to −2.63 kg at 0.75 mg.

Adults with type 2 diabetes across the phase 3 programme: 304 against 177 on metformin, 150 against 150 on insulin, 142 against 140 on an SGLT2 tablet

Measured in people

Source [01]

Against semaglutide, head to head

At the lower pair of amounts, long-term blood sugar fell 1.5 percentage points on semaglutide against 1.1 on dulaglutide, and weight fell 4.6 kg against 2.3 kg. At the higher pair it was 1.8 against 1.4 percentage points, and 6.5 kg against 3.0 kg — semaglutide took off more than twice as much weight. Gut problems were reported by 43% and 44% on the two semaglutide amounts and 33% and 48% on the two dulaglutide amounts.

1,201 adults with type 2 diabetes not controlled on metformin, at 194 sites in 16 countries, 40 weeks; 301 on semaglutide 0.5 mg, 299 on dulaglutide 0.75 mg, 300 on semaglutide 1.0 mg, 299 on dulaglutide 1.5 mg. The trial was open-label and funded by the maker of semaglutide.

Measured in people

Source [06]

Against liraglutide, the daily injection

Long-term blood sugar fell 1.42 percentage points on weekly dulaglutide against 1.36 on daily liraglutide — close enough that the trial's own conclusion was only that dulaglutide was not worse. Weight went the other way: 2.90 kg on dulaglutide against 3.61 kg on liraglutide, so the daily injection took off more.

599 adults with type 2 diabetes on metformin, 26 weeks; 299 on dulaglutide 1.5 mg weekly, 300 on liraglutide 1.8 mg daily

Measured in people

Source [01]

The higher amounts against the standard one

There was no dummy-injection group in this trial; every comparison is against the standard 1.5 mg amount. At 36 weeks, long-term blood sugar fell 1.87 percentage points at 4.5 mg and 1.71 at 3 mg against 1.53 at 1.5 mg. Weight fell 4.7 kg and 4.0 kg against 3.1 kg. The share losing at least 5% of their body weight was 49%, 40% and 31%. Stopping because of a side effect rose with the amount: 8.5%, 7.0% and 6.0% through 52 weeks.

1,842 adults with type 2 diabetes on metformin, split evenly between the three amounts (AWARD-11, NCT03495102), 52 weeks with the main comparison at 36

Measured in people

Source [01]Source [13]

Against a daily insulin injection

Long-term blood sugar fell 1.08 percentage points on dulaglutide 1.5 mg against 0.63 on insulin glargine at 52 weeks. Weight went in opposite directions: down 1.87 kg on dulaglutide, up 1.44 kg on insulin. Episodes of blood sugar dropping too low ran at 1.67 a person a year on dulaglutide against 3.02 on insulin.

807 adults with type 2 diabetes already on metformin and a sulphonylurea — 273 on dulaglutide 1.5 mg, 272 on 0.75 mg, 262 on insulin glargine, 78 weeks

Measured in people

Source [01]

In children and teenagers

Long-term blood sugar fell 0.8 percentage points on the two amounts pooled against a rise of 0.6 on dummy injections at 26 weeks, and 51.5% reached a target under 7.0% against 13.7%. Body mass index barely moved and did not separate from dummy injections: −0.1 kg/m2 pooled, −0.2 at 0.75 mg and −0.1 at 1.5 mg, against 0.0 on dummy injections. Whatever this drug does to weight, it did not do it here.

154 children and adolescents aged 10 and above with type 2 diabetes, with or without metformin and basal insulin (AWARD-PEDS, NCT02963766); 103 on dulaglutide and 51 on dummy injections for the 26-week blinded period

Measured in people

Source [01]Source [14]

In people whose kidneys already worked badly

Long-term blood sugar fell 1.10 percentage points at 1.5 mg and 1.10 at 0.75 mg against 1.00 on insulin glargine at 52 weeks — the trial's conclusion was only that dulaglutide was not worse. Weight fell 2.66 kg and 1.71 kg against a gain of 1.57 kg on insulin. Episodes of blood sugar dropping too low ran at 4.44 and 4.34 a person a year against 9.62 on insulin.

576 adults with type 2 diabetes and moderate to severe chronic kidney disease, average filtering rate 38 mL/min/1.73 m2, of whom 30% were under 30; 192, 190 and 194 in the three groups, 52 weeks

Measured in people

Source [01]

Gut effects dominate and cluster in the first two weeks. Against dummy injections: feeling sick 21.1% against 5.3%, being sick 12.7% against 2.3%, diarrhoea 12.6% against 6.7%, belly pain 9.4% against 4.9%, loss of appetite 8.6% against 1.6%; any gut problem at all 41% against 21%, and gut problems that made someone stop for good 3.5% against 0.2%. Over 5.4 years in REWIND, 47.4% reported a gut problem against 34.1% on placebo. Blood sugar dropping too low is not much of a risk from the drug alone but becomes one alongside insulin or a sulphonylurea. An inflamed pancreas is rare and did not exceed the placebo rate in the approval trials (0.07% against 0.14%), but both regulators keep it as a warning. Gallstones, a bowel that stops moving, kidney trouble caused by drying out from the vomiting, and severe allergic reactions are the other serious ones. The two regulators diverge in public: the US label carries a boxed warning about thyroid C-cell tumours seen in rats and forbids the drug to anyone with a personal or family history of medullary thyroid cancer or MEN 2, while the European product information states the same rat finding under preclinical data and lists no thyroid restriction; the US label also warns about worsening eye damage in people who already have it from diabetes (1.9% against 1.5% in REWIND, and 8.5% against 6.2% among those who arrived with it), which the European product information does not mention at all.

EffectWhat it feels likeHow often

Thyroid tumours — in rats, not shown in peopleSerious

This is the single biggest difference between the two regulators. The US label carries it as a boxed warning at the top of the document and forbids the drug outright to anyone with a personal or family history of medullary thyroid cancer or of the inherited condition MEN 2. The European product information states the same rat finding under preclinical data and says the clinical relevance is currently unknown, and lists no thyroid restriction at all. Nobody has shown the drug causes thyroid cancer in a person; nobody has shown it does not.

Not measured in people. In a two-year study in rats, at three times or more the exposure a person gets from the highest weekly amount, dulaglutide caused a dose-related rise in a particular kind of thyroid tumour. A six-month study in mice found no tumours at all.

Source [01]Source [02]

An inflamed pancreasSerious

Severe pain high in the belly that often bores through to the back, usually with vomiting, and the kind of thing people go to hospital for. Both regulators still carry it as a warning, because cases have come in after the drug went on sale and because it runs across the whole GLP-1 class. The instruction is that the drug is stopped if it is suspected and never restarted if it is confirmed. The drug also raises the pancreas enzymes in the blood by 11% to 21% on average in people with no symptoms at all, which the regulator says does not by itself predict anything.

Rare — between 1 and 10 people in every 10,000. In the trials behind the approval it happened to 0.07% on dulaglutide against 0.14% on dummy injections and 0.19% on other diabetes drugs, so the trials showed no excess.

Source [01]

A blocked bowelSerious

A swollen, painful belly with vomiting and no bowel movement. It is a surgical emergency. It sits in the product information as a reported event rather than a counted one, which means nobody knows how often it happens.

Frequency not known — the European regulator lists it with no rate that can be worked out from the available data. It is the non-mechanical kind: nothing is physically blocking the gut, it simply stops moving things along.

Source [01]

Gallstones and an inflamed gallbladderSerious

Gallstones can sit silently for years or block the bile duct, which is an emergency. Losing weight makes them more likely whatever causes the weight loss, and this drug causes less weight loss than most of its class — the difference in the heart trial was small.

Uncommon in the approval trials — between 1 and 10 people in every 1,000. Over 5.4 years in the 9,901-person heart trial, gallstones occurred at 0.62 per 100 patient-years on dulaglutide against 0.56 on dummy injections, and serious inflammation of the gallbladder in 0.5% against 0.3%.

Source [01]Source [02]

Kidney failure caused by drying outSerious

Being sick and having diarrhoea for days dries a person out, and dried-out kidneys work badly. Some of the reported cases needed dialysis, and some were in people with no known kidney problem beforehand. Both regulators tie it directly to the gut side effects rather than to any direct action on the kidney, and both warn about it most at the start of treatment and when the amount is increased.

Reported after the drug went on sale, with no rate published. Dehydration itself is listed as uncommon — between 1 and 10 people in every 1,000.

Source [01]Source [02]

Blood sugar dropping too lowSerious

Shakiness, sweating, confusion; the severe kind needs someone else's help. On its own or with metformin, between 5.9% and 10.9% of people had a documented episode and none of them were severe. With insulin taken at mealtimes the numbers become very different: 85.3% and 80.0% of people had an episode, and 2.4% and 3.4% had a severe one. The drug only releases insulin when blood sugar is already high, so the risk comes from what it is combined with.

It depends entirely on what else a person is taking, and the drug alone is not much of a risk. Alongside a sulphonylurea tablet the rate was 0.90 episodes a person a year against 0.04 on dummy injections with the same tablet. Alongside insulin it was 3.38 against 4.38 on dummy injections with the same insulin — lower, because people needed less insulin.

Source [01]

A whole-body allergic reactionSerious

The kind of reaction that drops blood pressure and closes the airway, along with swelling of the face, lips or throat. A previous allergic reaction to dulaglutide or to anything else in the injection is the only thing the European product information lists as an outright bar to taking it.

Rare — between 1 and 10 people in every 10,000, and reported after the drug went on sale rather than counted in the trials. Milder hypersensitivity such as hives or swelling was reported by 0.5% of people in the approval trials.

Source [01]

Stomach contents going into the lungs during an operationSerious

Because the drug slows the stomach, food can still be sitting there when someone is put to sleep for an operation or a scope, and it can then go down the wrong way into the lungs. It is a warning about a procedure, not about daily life. The instruction is to tell whoever is doing the procedure that you take it.

Reported cases, no rate published. Both regulators added it as a warning in 2024 rather than as a counted side effect.

Source [01]Source [02]

Worsening eye damage in people who already have it from diabetesSerious

The whole difference is in people who arrived with the damage already there. Bringing blood sugar down quickly is itself known to cause a temporary worsening, so the drug and the improvement it causes cannot be separated here. The US label instructs that people with a history of this should be watched; the European product information does not mention the eye at all, which is a real gap between the two documents rather than a difference of opinion anyone has published.

1.9% on dulaglutide against 1.5% on dummy injections over 5.4 years. Among people who already had eye damage at the start, 8.5% against 6.2%; among those who did not, 1% against 1%.

Source [02]

Feeling sick

The most common thing that happens, and the one most likely to make someone stop. It peaks in the first two weeks and falls away over the next four, after which the rate stays roughly flat. Across the whole class this is the price of admission; here it is somewhat lower than with the stronger GLP-1 drugs, because the drug itself is weaker.

About 2 in 10 at the 1.5 mg amount — 21.1% against 5.3% on dummy injections. At 0.75 mg it was 12.4%. Counted over an average of 23.8 weeks in 834, 836 and 568 people.

Source [01]Source [02]

Being sick

More than five times the dummy-injection rate at the higher amount. It is the effect most likely to dry a person out if it goes on for days, which is how the kidney warning arises.

12.7% at 1.5 mg and 6.0% at 0.75 mg, against 2.3% on dummy injections.

Source [02]

Diarrhoea

The smallest gap of the four common gut effects — about twice the dummy-injection rate rather than five times. Like the others it clusters in the first two weeks.

12.6% at 1.5 mg and 8.9% at 0.75 mg, against 6.7% on dummy injections.

Source [02]

Belly pain

Discomfort, tenderness, pain anywhere across the belly. It is the one gut symptom that overlaps with the warning sign for an inflamed pancreas, which is why the product information tells people what that pain feels like.

9.4% at 1.5 mg and 6.5% at 0.75 mg, against 4.9% on dummy injections.

Source [02]

Losing your appetite

Reported as a side effect rather than as a benefit, because in a diabetes trial that is what it is. It is more than five times the dummy-injection rate, and it is the mechanism behind whatever weight comes off.

8.6% at 1.5 mg and 4.9% at 0.75 mg, against 1.6% on dummy injections.

Source [02]

Indigestion and tiredness

Both sit just above the threshold at which the US label prints a side effect at all. Constipation, wind, a bloated belly, reflux and burping are all listed too, each in the low single digits and each a little above the dummy-injection rate.

Indigestion 5.8% at 1.5 mg against 2.3% on dummy injections; tiredness 5.6% against 2.6%.

Source [02]

A faster resting pulse and a slower electrical signal in the heart

Most people never notice either. The electrical delay is the mildest kind, picked up on a heart tracing rather than felt. Both are listed as common findings rather than as things that led anyone to stop.

An average rise of 2 to 4 beats a minute, with no dummy-injection figure printed for the average. A racing pulse with a rise of at least 15 beats was recorded in 1.3% at 0.75 mg and 1.4% at 1.5 mg in the European document, which prints no dummy-injection column; the US label, counting a different pool, gives 1.3% and 2.2% against 0.7% on dummy injections. A first-degree delay in the heart's electrical signal was recorded in 1.5% and 2.4% in the European document, and in the US label 1.7% and 2.3% against 0.9% on dummy injections.

Source [01]Source [02]

Reactions where the needle went in

Rash, redness, itching at the injection site, usually mild. The drug also produced antibodies against itself in 1.6% of people, generally at low levels and with no clear effect on how well it worked.

1.9% of people in the approval trials, of whom 0.7% had the kind the immune system drives; the European document prints no dummy-injection column for either. In the US label's placebo-controlled pool, which counts a narrower set of events, it was 0.5% against 0.0% on dummy injections. In the trial in children it was 3.9% and 3.8% on the two amounts against 2% on dummy injections.

Source [01]Source [02]

These are the amounts the studies below gave their participants, and they are here as facts about those studies. They are not a recommendation, not a starting point and not a range to pick from. What a person should take, if anything, is a question for a doctor who knows them.

REWIND, heart attacks and strokes in 9,901 adults with type 2 diabetes (NCT01394952) — the only amount tested in the heart trial

1.5 mg once a week, injected under the skin

Median 5.4 years

Source [02]Source [07]

The 104-week trial of blood sugar on top of metformin, against dummy injections and against a sitagliptin tablet, in 1,098 adults

0.75 mg or 1.5 mg once a week, injected under the skin

104 weeks, with the dummy group ending at 52

Source [01]

AWARD-11, the trial comparing the higher amounts with the standard one in 1,842 adults on metformin (NCT03495102), with no dummy group

1.5 mg, 3 mg or 4.5 mg once a week, injected under the skin

52 weeks, with the main comparison at 36

Source [01]Source [13]

SUSTAIN 7, the head-to-head trial against semaglutide in 1,201 adults on metformin (NCT02648204)

Dulaglutide 0.75 mg or 1.5 mg once a week against semaglutide 0.5 mg or 1.0 mg once a week, all injected under the skin

40 weeks

Source [06]

The 26-week trial against daily liraglutide, both on top of metformin, in 599 adults

Dulaglutide 1.5 mg once a week against liraglutide 1.8 mg once a day, both injected under the skin

26 weeks

Source [01]

AWARD-PEDS, the trial in 154 children and adolescents aged 10 and above with type 2 diabetes (NCT02963766)

0.75 mg or 1.5 mg once a week, injected under the skin

26 weeks blinded against dummy injections, then 26 weeks open

Source [01]Source [14]

The 52-week trial in 576 adults with type 2 diabetes and moderate to severe chronic kidney disease, against insulin glargine

0.75 mg or 1.5 mg once a week, injected under the skin

52 weeks

Source [01]

What people report

• What dulaglutide does to weight in someone who does not have type 2 diabetes — no trial has ever been run to find out, and every weight figure in this record comes from a diabetes trial
• Whether any of the weight comes back after stopping — no withdrawal trial has been published, so the one thing every reader wants to know about a weight effect is unmeasured here
• Whether the weight that comes off is fat or muscle — no body composition scan appears in either regulator's document
• Why the heart benefit happened — blood sugar, weight, blood pressure and kidney protein leak all improved together and the trial cannot separate them
• Whether the higher 3 mg and 4.5 mg amounts protect the heart — the heart trial used only 1.5 mg, and the higher amounts were approved on blood sugar and weight alone
• Whether the kidney finding would survive a trial designed to test it — it comes from an exploratory analysis of the heart trial, and only the protein-leak part of it was clear
• What happens past about five and a half years — that is the longest randomised follow-up there is, and type 2 diabetes is lifelong
• Whether the thyroid tumours seen in rats mean anything for a person — the two regulators read the same animal data and reached publicly different conclusions about what to do with it
• Whether the sudden optic-nerve injury the European regulator confirmed for semaglutide medicines in June 2025 applies to dulaglutide — no such review has been published for it, and its European product information does not mention the eye at all
• Why the European product information carries no warning about worsening eye damage in diabetes when the US label carries a whole section on it, from the same trial
• What it does in children under 10 — the youngest people studied were 10, and even in those aged 10 and up body weight did not separate from dummy injections
• Whether it is safe in pregnancy — the human data are described as none or limited, and the animal studies found reduced growth of the young and, in one, memory deficits in female offspring
• How it compares with tirzepatide for heart outcomes, or with semaglutide — the only head-to-head trial measured blood sugar and weight over 40 weeks, not events over years
• What happens to someone who switches from dulaglutide to a stronger GLP-1 medicine, or back — nobody has randomised that sequence
• How often the blocked bowel listed in the European product information actually happens — the regulator states no rate can be worked out from the data available
• What people who take it say about it over time — the only published experience source found for this record asked people which product profile they would prefer, not what happened to them on one

Does Trulicity make you lose weight?

A little, and much less than people expect from a GLP-1 drug. The European regulator's summary of the whole trial programme puts the change at the usual 1.5 mg amount somewhere between 0.35 kg and 2.90 kg, and at the lower 0.75 mg amount somewhere between a gain of 0.86 kg and a loss of 2.63 kg. Where a trial had a dummy-injection group, the difference was of that size: 3.18 kg against 1.47 kg over 26 weeks. At the highest 4.5 mg amount, 49% of people lost at least 5% of their body weight over 36 weeks, against 31% on the standard amount — but that trial had no dummy group at all.

Source [01]

Is Trulicity approved for weight loss?

No. The European approval is for type 2 diabetes in people aged 10 and over, as an addition to diet and exercise. The US approval is the same, plus a second use — reducing heart attacks, strokes and heart deaths in adults with type 2 diabetes who also have established heart disease or several heart risk factors. Neither regulator has approved dulaglutide for weight management, and its maker never ran a weight-management trial programme for it. Weight loss appears in the record only as a secondary measurement inside diabetes trials.

Source [01]Source [02]

Trulicity vs Ozempic — which one works better?

They were compared directly in 1,201 people over 40 weeks, and semaglutide — the substance in Ozempic and Wegovy — won on both measures. Long-term blood sugar fell 1.8 percentage points on the higher semaglutide amount against 1.4 on the higher dulaglutide amount. Weight fell 6.5 kg against 3.0 kg, so semaglutide took off more than twice as much. Gut problems were reported by 44% on that semaglutide amount and 48% on that dulaglutide amount. The trial was run open-label and paid for by the maker of semaglutide, which is worth knowing, though the size of the weight gap is hard to explain away.

Source [06]

What are the side effects of Trulicity?

Mostly the gut, mostly early. Against dummy injections, feeling sick was reported by 21.1% against 5.3%, being sick by 12.7% against 2.3%, diarrhoea by 12.6% against 6.7%, belly pain by 9.4% against 4.9% and loss of appetite by 8.6% against 1.6%. Any gut problem at all: 41% against 21%. They peak in the first two weeks and settle over the next four. The serious but uncommon ones are an inflamed pancreas, gallstones, a blocked bowel, kidney trouble caused by drying out from the vomiting, and severe allergic reactions. In the US the label also carries a boxed warning about thyroid tumours seen in rats.

Source [01]Source [02]

Does Trulicity protect your heart?

In people who already have type 2 diabetes, the evidence says yes, by a modest amount. In a trial of 9,901 adults followed for a median of 5.4 years, a heart attack, a stroke or a death from a heart cause happened to 594 people on dulaglutide (12.0%) against 663 on dummy injections (13.4%). Most of the difference was strokes: 158 against 205. Deaths from any cause were 536 against 592, and that difference was small enough that it could have been chance. What makes this trial unusual is that about two-thirds of the people in it had never had a heart event — they were there because of risk factors.

Source [02]Source [03]

How long does Trulicity take to work?

Blood sugar starts moving straight away — it falls after the very first injection, and most of the effect on fasting blood sugar is there within two weeks. The drug itself takes longer to reach a steady level in the blood: it peaks about 48 hours after an injection and stops climbing after two to four weekly injections. Weight moves far more slowly and keeps drifting down for a year; in the trial that ran to 52 weeks, the loss at that point was 3.5 kg, 4.3 kg and 5.0 kg on the three amounts tested.

Source [01]

Can you buy dulaglutide online without a prescription?

Not in the way people buy other peptides online, because it is not a peptide. Dulaglutide is a fusion protein of about 63 kilodaltons made of two identical chains held together by sulphur bridges, and both regulators state it is grown in Chinese hamster ovary cells rather than built by chemical synthesis. That puts it out of reach of the workshops that turn out grey-market peptides. It also does not appear anywhere on the US regulator's list of bulk substances nominated for pharmacy compounding, updated 14 May 2026, which does list several of the peptides sold online. It is a prescription-only medicine in the EU and the US.

Source [01]Source [02]Source [12]

Is dulaglutide banned in sport?

No. Dulaglutide does not appear anywhere in the 2026 World Anti-Doping Agency Prohibited List, which came into force on 1 January 2026. It is not caught by the catch-all clause for unapproved substances either, because it is an approved medicine. It is also not on the separate 2026 Monitoring Program — that document does list markers of semaglutide and tirzepatide, in and out of competition, but not dulaglutide.

Source [09]Source [10]

Does Trulicity cause depression or suicidal thoughts?

The European regulator investigated exactly that question, and dulaglutide was one of the five substances in the review. At its meeting of 8 to 11 April 2024 the safety committee concluded that the available evidence does not support a causal link between these medicines and suicidal or self-harming thoughts and actions. The review covered laboratory studies, the trials, reports collected after the medicines went on sale, and two large studies of electronic health records. No change was made to the product information, and manufacturers were told to keep watching for it.

Source [11]

1. [01]EMA – Trulicity (dulaglutide) summary of product characteristics, sections 4.1, 4.3, 4.4, 4.6, 4.8, 5.1 (Tables 3–13, cardiovascular outcome study), 5.2 and 5.3
2. [02]FDA – TRULICITY (dulaglutide) injection, US prescribing information, revised May 2025: boxed warning, section 1 indications, sections 5.5–5.9, section 6.1 Table 1, section 11 Description and section 14.5 with Table 14 (REWIND) (2025)
3. [03]Gerstein HC et al. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet 2019;394:121–130 (2019)
4. [04]Gerstein HC et al. Dulaglutide and renal outcomes in type 2 diabetes: an exploratory analysis of the REWIND randomised, placebo-controlled trial. Lancet 2019;394:131–138 (2019)
5. [05]Gerstein HC et al. The effect of dulaglutide on stroke: an exploratory analysis of the REWIND trial. Lancet Diabetes Endocrinol 2020 (2020)
6. [06]Pratley RE et al. Semaglutide versus dulaglutide once weekly in patients with type 2 diabetes (SUSTAIN 7): a randomised, open-label, phase 3b trial. Lancet Diabetes Endocrinol 2018;6:275–286 (2018)
7. [07]ClinicalTrials.gov API v2: NCT01394952 (REWIND), 9,901 participants, phase 3, started 22 July 2011, primary completion 21 August 2018, completed, results posted
8. [08]EMA – Trulicity, EPAR (authorised 21 November 2014, Eli Lilly Nederland B.V.)
9. [09]WADA Prohibited List 2026, in force 1 January 2026 (official publication, Austrian BGBl. III no. 219/2025) – dulaglutide appears in no class (2025)
10. [10]WADA – The 2026 Monitoring Program: item 6 lists markers of semaglutide and tirzepatide, in and out of competition; dulaglutide is not listed (2026)
11. [11]EMA – Meeting highlights from the Pharmacovigilance Risk Assessment Committee (PRAC), 8–11 April 2024: no causal link between GLP-1 receptor agonists and suicidal or self-injurious thoughts; review covered dulaglutide (Trulicity) among five substances (2024)
12. [12]FDA – Bulk Drug Substances Nominated for Use in Compounding Under Section 503A, updated 14 May 2026: dulaglutide appears in no category (2026)
13. [13]ClinicalTrials.gov API v2: NCT03495102 (AWARD-11), 1,842 participants, completed, results posted
14. [14]ClinicalTrials.gov API v2: NCT02963766 (AWARD-PEDS), 154 participants, completed, results posted