Is the GLP-1 effect just eating less?

Why ask at all?

The first piece, Same weight lost, different liver, covered a trial where three diets produced the same 10% weight loss. In people with prediabetes and fatty liver, the very-low-carbohydrate diet still did more for the liver. Same amount lost, different result. What people ate mattered, not only how much.

GLP-1 drugs keep turning up benefits no one set out to measure. Liver fat. Heart attacks and strokes. Kidneys. Maybe drinking. Two questions follow. Is the drug doing all of that, or doing one thing, taking away the urge to eat, with everything else following from fewer calories? And when appetite drops, does what people eat change too?

What does the drug actually change?

Appetite, first. People on semaglutide ate 24% to 35% less at lab meals. They reported fewer cravings and chose high-fat foods less often. The drugs work on the brain's appetite and reward signals.

Whether they also change what people eat day to day, for example less carbohydrate, has barely been studied.

What happens to the liver?

A lot, and the evidence is randomized. In a phase 3 trial of semaglutide, fatty liver inflammation resolved in 62.9% of people against 34.3% on placebo. Tirzepatide resolved it in 44% to 62% against 10%. Liraglutide, an older drug in the same class, resolved it in 39% against 9%.

How much of that is the weight?

Most of it. Researchers took the semaglutide trial data and asked how much of each improvement ran through weight lost. Weight loss accounted for about 83% of the fall in liver fat and about 69% of the inflammation resolving. For scarring the estimate was 25%, but with an interval so wide it means little.

Two small trials asked the question head on. People took liraglutide or changed their diet and exercise, and both groups lost about the same weight. Liver fat fell by about the same amount in both. Once the drug stopped, that group regained weight and liver fat. The diet group did not.

Do people move more on them?

No. A 2026 review pooled seven studies that measured movement with a device, 924 people in all. Activity did not change overall, and most of the studies leaned slightly lower. Every one of them tested the older drugs, exenatide and liraglutide. No trial of semaglutide or tirzepatide had published device-measured activity as of March 2026.

People do report that moving gets easier. In heart failure, semaglutide added about 20 meters to a six-minute walk. Being able to do more is not the same as doing more.

Fitness came from exercise. After a diet, people kept off 4.1 kg more than placebo with exercise alone, 6.8 kg with liraglutide alone, and 9.5 kg with both. A year after everything stopped, the drug-only group had regained 6.0 kg more than the exercise-only group.

Do low-carb and fasting do something extra for the liver?

Low-carb may get there faster. In a two-week controlled study, liver fat fell 55% on carbohydrate restriction against 28% on calorie restriction, at similar weight loss. A randomized trial saw the same early gap at 48 hours, 29.6% against 8.9%. By the time both groups had lost 7% of their weight, it was gone.

Fasting has matched ordinary calorie cutting, not beaten it. A 12-month trial of time-restricted eating found liver fat fell the same on both. So did a trial of the 5:2 diet. When a 5:2 diet and a low-carb diet were tested side by side at the same weight loss, 7.4 kg and 7.3 kg, the liver result was the same too.

What about the heart?

This is where the drug and eating less part ways.

The largest test of the drug's effect on the heart is a trial called SELECT. It followed 17,604 people without diabetes, on semaglutide or placebo, for about three years. Major heart events fell from 8.0% to 6.5%. A planned analysis found that how much weight people had lost by week 20 did not predict their later risk. Shrinking waistlines explained about a third of the benefit.

A trial called LOOK AHEAD tested the other route: eating less and moving more. Its 5,145 adults with type 2 diabetes cut calories, added activity, and lost 8.6% of their weight in the first year. Heart events did not fall, and the trial was stopped for futility after a median of 9.6 years. People who lost 10% or more did better, but that comparison was not randomized.

The diet trials that did cut heart events changed what people ate, not how much. In PREDIMED, a Spanish trial, a Mediterranean diet with extra olive oil or nuts lowered heart events by roughly 30% while weight barely moved. The trial was retracted in 2013 over how people were randomized, then republished in 2018. The result held.

No trial of fasting, time-restricted eating, low-carb or keto has ever counted heart attacks and strokes. What exists is risk factors. Low-carb lowers triglycerides and raises HDL, with a small rise in LDL. Across 99 trials, fasting schedules did about the same as cutting calories every day.

The widely shared claim that an eight-hour eating window raises heart death by 91% does not appear in any primary source. The study behind it is observational. It counted eating hours from two days of food recall, not from people practicing time-restricted eating.

Is it what people stop eating?

Here is a hypothesis worth testing. People who start these drugs most likely ate the standard American diet, and that diet is mostly processed food: processed meat, fast food, snacks. If the calories they cut came mostly from those foods, the drug would be changing what they eat, not only how much. And cutting the wrong kind of food can help the heart on its own.

Each step, and how far the evidence goes:

• What did they eat before? No one has measured it directly, so this step is an inference. In one small study of 28 people starting semaglutide or tirzepatide, diet quality scored about 59 out of 100 before treatment. The national average is 58. Starters, on that evidence, ate a typical American diet. That diet now gets 57% of its calories from ultra-processed food, up from 53.5% in 2001. Not more processed than everyone else's, then, but the standard diet, which is mostly processed.
• Do the processed foods go first? Partly, and part of it is arithmetic. If most of what someone eats is ultra-processed, an even cut across everything still removes more ultra-processed calories than anything else. That follows from the inference above, not from a measurement. Beyond the arithmetic, the data lean the same way. In a US household panel, spending on savory snacks fell 10.1% and on fast food 8.0%, against 5.3% for groceries overall. In a small Japanese study of 25 people with no comparison group, sweets and snacks roughly halved while total intake fell by about a fifth. Meat did not change significantly. Danish supermarket receipts barely changed per 100 g, which points to eating less of everything with only a slight tilt away from ultra-processed goods. No one measured processed meat.
• Does cutting the wrong kind help the heart? For saturated fat, in trials, yes. Across 15 randomized trials and 56,675 people, cutting it lowered combined heart events by about 17%. It made little difference whether the calories went to polyunsaturated fat or to carbohydrate. For processed meat and ultra-processed food, the links come only from observational studies, and the evidence is weak.
• Did it happen in the SELECT heart trial? Unknown. SELECT never recorded what people ate.

The diets that cut heart events in trials also point somewhere the drugs do not. They added olive oil and nuts against low-fat advice. The drugs lower the appetite for fatty food, and no study has measured whether people on them eat more olive oil, nuts or fish.

So the hypothesis is reasonable and mostly untested. A food diary inside a trial like SELECT could have answered it.

What about the kidneys?

The drug's kidney trial is called FLOW. In people with type 2 diabetes and kidney disease, semaglutide cut a combined measure of serious kidney decline by 24%. No one has yet separated that from the weight lost.

On the diet side, low-carb and higher-protein diets did not harm kidney function in people without kidney disease, across 28 trials and 1,358 people. Human trials of fasting and the kidney are small and short.

LOOK AHEAD cut a measure of very-high-risk kidney disease, though that measure was put together after the trial ended. Over 15.6 years, serious kidney decline was not significantly reduced overall. No diet trial has tested kidney outcomes the way FLOW did.

So is it just eating less?

Less, and possibly different. The drug takes away much of the urge to eat. For the weight, the liver and the inflammation markers, that is most of the story. They follow how much people stop eating, whatever the route. The drug does not add movement. That still takes exercise.

The heart is where "how much" stops explaining it. In SELECT, the benefit did not track the weight people lost. In LOOK AHEAD, eating less and moving more did not reproduce it. The diet trials that did cut heart events changed what people ate.

So the better question may be what goes when appetite drops. If people start from the standard American diet, most of what they stop eating is processed food, and in purchase data the snacks and fast food go faster than the rest. Whether that is part of the heart benefit has never been measured. SELECT never asked what people ate.

The heart finding itself came from an analysis inside SELECT, funded by the drug's maker. It is a strong signal, not a closed case.

This piece reports what the trials found. It is not a case for or against taking one, and it is not medical advice.

Sources

1. Blundell J, et al. Effects of once‐weekly semaglutide on appetite, energy intake, control of eating, food preference and body weight in subjects with obesity. Diabetes Obes Metab. 2017. https://doi.org/10.1111/dom.12932
2. Friedrichsen M, et al. The effect of semaglutide 2.4 mg once weekly on energy intake, appetite, control of eating, and gastric emptying in adults with obesity. Diabetes Obes Metab. 2021. https://doi.org/10.1111/dom.14280
3. Christensen S, et al. Dietary intake by patients taking GLP-1 and dual GIP/GLP-1 receptor agonists: A narrative review and discussion of research needs. Obes Pillars. 2024. https://doi.org/10.1016/j.obpill.2024.100121
4. Sanyal AJ, et al. Phase 3 Trial of Semaglutide in Metabolic Dysfunction–Associated Steatohepatitis. N Engl J Med. 2025. https://doi.org/10.1056/NEJMoa2413258
5. Loomba R, et al. Tirzepatide for Metabolic Dysfunction–Associated Steatohepatitis with Liver Fibrosis. N Engl J Med. 2024. https://doi.org/10.1056/NEJMoa2401943
6. Armstrong MJ, et al. Liraglutide safety and efficacy in patients with non-alcoholic steatohepatitis (LEAN): a multicentre, double-blind, randomised, placebo-controlled phase 2 study. Lancet. 2016. https://doi.org/10.1016/S0140-6736(15)00803-X
7. Jara M, et al. Modulation of metabolic, inflammatory and fibrotic pathways by semaglutide in metabolic dysfunction-associated steatohepatitis. Nat Med. 2025. (sponsor-affiliated authors) https://doi.org/10.1038/s41591-025-03799-0
8. Khoo J, et al. Randomized trial comparing effects of weight loss by liraglutide with lifestyle modification in non‐alcoholic fatty liver disease. Liver Int. 2019. https://doi.org/10.1111/liv.14065
9. Khoo J, et al. Comparative effects of liraglutide 3 mg vs structured lifestyle modification on body weight, liver fat and liver function in obese patients with non‐alcoholic fatty liver disease: A pilot randomized trial. Diabetes Obes Metab. 2017. https://doi.org/10.1111/dom.13007
10. Kim H, Chae JH, Moon HY. The impact of glucagon-like peptide-1 receptor agonists on objective physical activity in adults: a systematic review and exploratory meta-analysis. Int J Obes. 2026. https://doi.org/10.1038/s41366-026-02141-z
11. Kosiborod MN, et al. Semaglutide in Patients with Heart Failure with Preserved Ejection Fraction and Obesity. N Engl J Med. 2023. https://doi.org/10.1056/NEJMoa2306963
12. Lundgren JR, et al. Healthy Weight Loss Maintenance with Exercise, Liraglutide, or Both Combined. N Engl J Med. 2021. https://doi.org/10.1056/NEJMoa2028198
13. Jensen SBK, et al. Healthy weight loss maintenance with exercise, GLP-1 receptor agonist, or both combined followed by one year without treatment: a post-treatment analysis of a randomised placebo-controlled trial. eClinicalMedicine. 2024. https://doi.org/10.1016/j.eclinm.2024.102475
14. Browning JD, et al. Short-term weight loss and hepatic triglyceride reduction: evidence of a metabolic advantage with dietary carbohydrate restriction. Am J Clin Nutr. 2011. (semirandom assignment) https://doi.org/10.3945/ajcn.110.007674
15. Kirk E, et al. Dietary Fat and Carbohydrates Differentially Alter Insulin Sensitivity During Caloric Restriction. Gastroenterology. 2009. https://doi.org/10.1053/j.gastro.2009.01.048
16. Wei X, et al. Effects of Time-Restricted Eating on Nonalcoholic Fatty Liver Disease: The TREATY-FLD Randomized Clinical Trial. JAMA Netw Open. 2023. https://doi.org/10.1001/jamanetworkopen.2023.3513
17. Schübel R, et al. Effects of intermittent and continuous calorie restriction on body weight and metabolism over 50 wk: a randomized controlled trial. Am J Clin Nutr. 2018. https://doi.org/10.1093/ajcn/nqy196
18. Holmer M, et al. Treatment of NAFLD with intermittent calorie restriction or low-carb high-fat diet – a randomised controlled trial. JHEP Rep. 2021. https://doi.org/10.1016/j.jhepr.2021.100256
19. Lincoff AM, et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. N Engl J Med. 2023. https://doi.org/10.1056/NEJMoa2307563
20. Deanfield J, et al. Semaglutide and cardiovascular outcomes by baseline and changes in adiposity measurements: a prespecified analysis of the SELECT trial. Lancet. 2025. (sponsor-funded) https://doi.org/10.1016/S0140-6736(25)01375-3
21. Look AHEAD Research Group (Wing RR, et al). Cardiovascular Effects of Intensive Lifestyle Intervention in Type 2 Diabetes. N Engl J Med. 2013. https://doi.org/10.1056/NEJMoa1212914
22. Gregg EW, et al. Association of the magnitude of weight loss and changes in physical fitness with long-term cardiovascular disease outcomes in overweight or obese people with type 2 diabetes: a post-hoc analysis of the Look AHEAD randomised clinical trial. Lancet Diabetes Endocrinol. 2016. https://doi.org/10.1016/S2213-8587(16)30162-0
23. Estruch R, et al. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts. N Engl J Med. 2018. (republished after retraction) https://doi.org/10.1056/NEJMoa1800389
24. Mansoor N, et al. Effects of low-carbohydrate dietsv. low-fat diets on body weight and cardiovascular risk factors: a meta-analysis of randomised controlled trials. Br J Nutr. 2016. https://doi.org/10.1017/S0007114515004699
25. Semnani-Azad Z, et al. Intermittent fasting strategies and their effects on body weight and other cardiometabolic risk factors: systematic review and network meta-analysis of randomised clinical trials. BMJ. 2025. https://doi.org/10.1136/bmj-2024-082007
26. Chen M, et al. Association of eating duration less than 8 h with all-cause, cardiovascular, and cancer mortality. Diabetes Metab Syndr. 2025. https://doi.org/10.1016/j.dsx.2025.103278
27. Perkovic V, et al. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes. N Engl J Med. 2024. https://doi.org/10.1056/NEJMoa2403347
28. Devries MC, et al. 'Changes in Kidney Function Do Not Differ between Healthy Adults Consuming Higher- Compared with Lower- or Normal-Protein Diets: A Systematic Review and Meta-Analysis. J Nutr. 2018. https://doi.org/10.1093/jn/nxy197
29. Look AHEAD Research Group. Effect of a long-term behavioural weight loss intervention on nephropathy in overweight or obese adults with type 2 diabetes: a secondary analysis of the Look AHEAD randomised clinical trial. Lancet Diabetes Endocrinol. 2014. https://doi.org/10.1016/S2213-8587(14)70156-1
30. Knowler WC, et al. Within and post-trial effects of an intensive lifestyle intervention on kidney disease in adults with overweight or obesity and type 2 diabetes mellitus: a secondary analysis of the Look AHEAD clinical trial. BMJ Open Diabetes Res Care. 2024. https://doi.org/10.1136/bmjdrc-2024-004079
31. Petersen MC, et al. Effect of diet macronutrient content on the cardiometabolic response to weight loss: A randomized clinical trial. Cell Metab. 2026. https://doi.org/10.1016/j.cmet.2026.07.020
32. Hristakeva S, Liaukonytė J, Feler L. The No-Hunger Games: How GLP-1 Medication Adoption Is Changing Consumer Food Demand. J Mark Res. 2026. https://doi.org/10.1177/00222437251412834
33. Sawada M, et al. Effects of Oral Semaglutide on Dietary Intake and Body Composition in Japanese People With Type 2 Diabetes: A Prospective Observational Study in Clinical Practice. Endocrinol Diabetes Metab. 2026. https://doi.org/10.1002/edm2.70295
34. Sørensen KK, et al. Consumer Food Purchases After Glucagon-Like Peptide-1 Receptor Agonist Initiation. JAMA Netw Open. 2026. https://doi.org/10.1001/jamanetworkopen.2025.55449
35. Hooper L, et al. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2020. https://doi.org/10.1002/14651858.CD011737.pub3
36. Micha R, Wallace SK, Mozaffarian D. Red and Processed Meat Consumption and Risk of Incident Coronary Heart Disease, Stroke, and Diabetes Mellitus: A Systematic Review and Meta-Analysis. Circulation. 2010. https://doi.org/10.1161/CIRCULATIONAHA.109.924977
37. Zeraatkar D, et al. Red and Processed Meat Consumption and Risk for All-Cause Mortality and Cardiometabolic Outcomes: A Systematic Review and Meta-analysis of Cohort Studies. Ann Intern Med. 2019. https://doi.org/10.7326/M19-0655
38. Babazadeh D, et al. Changes in food cravings, dietary quality, body composition, and dietary intake during GLP-1 receptor agonist therapy: The CRAVE study. Obes Pillars. 2026. https://doi.org/10.1016/j.obpill.2026.100292
39. Juul F, et al. Ultra-processed food consumption among US adults from 2001 to 2018. Am J Clin Nutr. 2022. https://doi.org/10.1093/ajcn/nqab305
40. USDA Food and Nutrition Service. HEI Scores for Americans (average HEI-2020 score of 58 for ages 2 and older, citing Shams-White et al. 2023). https://www.fns.usda.gov/cnpp/hei-scores-americans

General fitness and wellness information, not medical advice. Nothing here substitutes for assessment by a qualified clinician.

← Back to the library