GLP-1 side effects: what the science actually shows

TL;DR

A February 2026 review in the Journal of Clinical Investigation found that GLP-1 medications cause nausea in about 19% of users versus 6.5% on placebo - but most side effects are manageable, and concerns about thyroid cancer, pancreatitis, and psychiatric harm are not supported by the best available evidence.

The internet has no shortage of GLP-1 horror stories. It also has no shortage of companies dismissing every side effect report as overblown. Neither extreme is helpful. A comprehensive review published in the Journal of Clinical Investigation (JCI) in February 2026 - covering data through the same month - is the most thorough synthesis of GLP-1 safety evidence produced so far, and it gives us actual numbers to work with instead of anecdotes.

Gastrointestinal side effects: real, common, usually temporary

The most common side effects of GLP-1 medications are gastrointestinal. That is not in dispute. The JCI review quantified them clearly:

• Nausea: 19.3% of GLP-1 users vs 6.5% on placebo
• Vomiting: 7.6% vs 2.0% on placebo
• GI events led to drug discontinuation in 6.5% of users vs 3.6% on placebo
• Nausea was the leading single cause of stopping treatment

Among newer agents, relative risk estimates for nausea were: semaglutide (RR 2.95), tirzepatide (RR 2.90), and orforglipron - the new oral GLP-1 pill - at RR 4.77. That last number stands out. Orforglipron appears to cause more nausea than injectable options, which tracks with what patients in early real-world use have reported.

Age matters for tolerability too. Older patients discontinue treatment at higher rates: 82.6% of those aged 75 and above had stopped by 24 months, versus 68.2% of those under 65. If you are older and finding the GI effects harder to manage, you are not alone - and you are not unusual.

Why GI symptoms happen - and why they are not always linked to benefit

The mechanisms behind nausea are better understood now. GLP-1 medications work through two pathways: peripheral effects that slow gastric emptying and reduce small intestinal motility, and central effects through direct brainstem interaction and vagal signalling. The critical insight from the JCI review is that reduced energy intake and delayed gastric emptying appear to be independent effects. In plain terms: nausea may be partly separable from the drug's appetite-suppressing benefit. This opens the door to approaches that preserve the weight loss while reducing the stomach discomfort - an active area of drug development.

Practical mitigation

The review noted that slower dose escalation improved adherence in pilot studies. If nausea is a barrier, this is worth discussing with your prescriber. Most protocols already include gradual titration, but some patients benefit from extending the dose-increase intervals further than the standard schedule allows.

The aspiration and surgery concern

GLP-1 medications slow gastric emptying, which raises a genuine concern for anyone undergoing procedures under sedation or general anaesthesia: retained gastric contents. The review documented this clearly - 56% of GLP-1 users had retained gastric contents compared to 19% of non-users before endoscopy, and odds ratios for premature endoscopy termination ranged from 4.5 to 13.9 across systematic reviews.

However - and this is the important qualifier - no increased risk of aspiration pneumonia was found across multiple studies. The retained contents are a risk factor, but they do not appear to be translating into the worst-case outcome at a population level. Multi-society consensus now recommends an individualised perioperative approach rather than a blanket rule to stop GLP-1 medications before surgery. Talk to both your GLP-1 prescriber and your anaesthetist before any planned procedure.

Thyroid cancer: what the data actually say

The thyroid cancer concern has circulated since GLP-1 drugs were first approved. Rat studies showed GLP-1 receptors expressed in thyroid C cells, leading to medullary thyroid carcinoma in rodents. But human biology differs in an important way: GLP-1 receptors are expressed minimally in healthy human thyroid C cells - though they are highly expressed in existing medullary thyroid carcinomas.

What the human data shows:

• A French health insurance database study found elevated medullary thyroid carcinoma risk (HR 1.78, 95% CI 1.04-3.05) in GLP-1 users
• Cardiovascular outcome trial meta-analyses found 6 cases in active treatment groups vs 2 on placebo - a small absolute number
• For other thyroid cancers, results conflict: the French study showed HR 1.58 (elevated), while a Scandinavian registry showed HR 0.93 (not elevated)

The current picture: there is a possible signal for medullary thyroid carcinoma that researchers are still investigating. The absolute risk remains very low. Both semaglutide and tirzepatide carry a black-box warning advising against use in people with a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia type 2 syndrome. Outside those contraindicated groups, initial results indicate the risk is small - but monitoring continues.

Pancreatitis: the concern that largely did not materialise

Early case reports raised alarm about GLP-1 medications and acute pancreatitis. The JCI review found that meta-analyses of multiple cardiovascular outcome trials have since excluded any causal link. GLP-1 medications do elevate serum amylase and lipase in most patients - enzymes associated with pancreatic stress - but without causing clinical pancreatitis in the studies examined.

The authors note that early concerns were partly traced to symptom overlap (GLP-1-induced nausea and abdominal discomfort can mimic pancreatitis symptoms) and diagnostic criteria misinterpretation. This is one area where the initial fear has been substantially, though not fully, resolved by accumulated data.

Retinopathy: a nuanced picture

The SUSTAIN-6 semaglutide trial found worsened diabetic retinopathy complications - but only in patients with pre-existing advanced retinopathy who experienced rapid HbA1c reduction of more than 16 weeks. The pattern mirrors what happens with intensified insulin therapy: rapid glucose normalisation can temporarily worsen existing retinopathy before long-term improvement sets in.

For NAION - a rare form of vision loss (non-arteritic anterior ischemic optic neuropathy) - evidence remains mixed. One cohort study found a hazard ratio of 4.28 for semaglutide users, but other studies show no increased incidence. Absolute incidence sits at approximately 14.5 per 100,000 person-years - low, but not trivial for an individual patient. The JCI review recommends retinal screening before starting GLP-1 therapy, particularly in people with pre-existing diabetic eye disease.

Psychiatric safety: the fear that did not hold up

Concerns about GLP-1 medications causing depression and suicidal ideation circulated widely in 2023-2024. The FDA investigated and subsequently removed the warning from the label in January 2026. The JCI review supports that decision. The best-designed analysis - a meta-analysis of 80 randomised controlled trials covering 107,860 participants - found no association between GLP-1 receptor agonist treatment and serious psychiatric adverse effects.

GLP-1 therapy was actually associated with improved mental health-related quality of life. Whether this is a direct neurological effect or a consequence of weight loss and improved physical health is still being investigated. Early findings suggest it may be both.

Muscle loss: a real concern, but graded by drug

The JCI review confirmed that GLP-1 medications affect body composition alongside fat loss. Semaglutide users lost approximately 39% of their total weight as fat-free mass (which includes muscle). Tirzepatide users lost approximately 25%. Both figures raise legitimate concern for sarcopenia - particularly in older adults who already have lower muscle reserves.

This is one area where nutrition during GLP-1 treatment matters directly. Adequate protein intake (current clinical guidance suggests a minimum of 1.2g per kg of body weight daily) and resistance exercise both help preserve lean mass. Vitamins that support muscle metabolism - including vitamin D and B vitamins - are also worth monitoring given evidence that GLP-1 users are at elevated risk of deficiency in both. GLP-1 Shield is built specifically around these gaps.

The gallstone risk nobody talks about

One side effect that gets less attention than nausea but is statistically well-established: GLP-1 medications increase the risk of gallstones (cholelithiasis). The JCI review found a relative risk of 1.46 (95% CI: 1.09-1.97) compared to placebo. Rapid weight loss itself is a known gallstone trigger - bile becomes more saturated during fast fat mobilisation. GLP-1 medications compound this through slowed gallbladder emptying.

If you develop upper right abdominal pain, particularly after meals, on GLP-1 medications, it is worth investigating. The good news: the review found no increased risk of cholecystitis (gallbladder inflammation), cholangitis, or gallstone-triggered pancreatitis - the more serious downstream complications.

Frequently asked questions

How common is nausea on Ozempic or Wegovy?

About 19.3% of GLP-1 users experience nausea in clinical trials, compared to 6.5% on placebo. That means roughly one in five users will experience it, with most cases concentrated in the dose-escalation phase. Slower titration reduces but does not eliminate this risk. Most people find it manageable or temporary.

Do GLP-1 medications cause thyroid cancer?

There is a possible signal for medullary thyroid carcinoma in some human studies, but the absolute risk is very low and evidence is not conclusive. Both semaglutide and tirzepatide are contraindicated in people with a personal or family history of medullary thyroid carcinoma or MEN 2 syndrome. Outside those groups, initial results indicate risk is small - ongoing monitoring will clarify this further.

Are GLP-1 medications safe for people with depression or anxiety?

A meta-analysis of 80 trials and over 107,000 participants found no association between GLP-1 medications and serious psychiatric side effects. GLP-1 therapy was actually associated with improved mental health-related quality of life. The FDA removed the earlier psychiatric warning from the label in January 2026 based on this accumulated evidence.

Does semaglutide cause muscle loss?

Semaglutide users lose approximately 39% of their total weight as fat-free mass, which includes muscle. Tirzepatide users lose less - about 25%. Both figures support the recommendation for adequate protein intake and resistance exercise during treatment. Monitoring vitamin D and B12, which affect muscle metabolism, is also advisable given known GLP-1 deficiency risks.