Nausea gets the attention, but the complaint that arrives more quietly in the second week on a GLP-1 is a flat, heavy tiredness that makes a normal afternoon feel like the end of a long shift. People describe it as running on a low battery. Most of them have no idea how little they have eaten.
Fatigue is a listed side effect of semaglutide and tirzepatide, reported by about 11 percent of adults on semaglutide 2.4 mg in the pooled STEP trials and by 5 to 7 percent on tirzepatide in the SURMOUNT programme, against 3 to 5 percent on placebo. The drug does not slow the body down directly. What the drug does is cut energy intake by a fifth to a quarter, often within days, and the tiredness is the body's response to a sudden deficit in calories, carbohydrate, fluid, and sodium that the person has not registered because the appetite signal that would normally flag it has gone quiet.
This article sets out the trial rates, explains where the tiredness comes from, shows why it clusters in the weeks after each dose increase, introduces the concept of energy availability as a way to think about the floor below which fatigue becomes inevitable, and lists the unrelated causes that a GLP-1 can mask. A microdose produces a smaller intake reduction than any of the trial doses, so the figures here describe the upper end of what a low dose protocol should produce.
Key Takeaways
- Fatigue was reported by about 11 percent of adults on semaglutide 2.4 mg against 5 percent on placebo in the pooled STEP 1, 3, and 4 trials, and by about 5 percent at tirzepatide 5 and 10 mg and 7 percent at 15 mg against 3 percent on placebo in the pooled SURMOUNT trials, according to the US prescribing information.
- The primary cause is reduced energy intake. Semaglutide 1 mg lowered ad libitum daily energy intake by about 24 percent over 12 weeks in the crossover study by Blundell and colleagues in 2017, which for a 2,450 calorie eater is a drop of roughly 600 calories a day, most of it from carbohydrate.
- Fatigue clusters in the first two weeks after starting and after each dose increase, then fades over two to four weeks as intake stabilises and the body shifts toward burning more stored fat. A low dose protocol that skips repeated escalation steps avoids the repeated dips.
- Total body glycogen, about 400 to 500 grams in a fed adult, falls by roughly half within a few days when carbohydrate intake drops below 50 to 80 grams a day. Keeping 100 to 150 grams of carbohydrate in the day maintains a partial store while fat loss continues.
- Energy availability below about 30 kcal per kilogram of fat free mass per day, the threshold identified by Loucks and Thuma in 2003, is associated with hormonal suppression, poor recovery, and persistent tiredness. A person with 55 kg of fat free mass eating 1,400 calories and training 300 calories off sits at 20 kcal per kg.
- Iron deficiency, low B12, an underactive thyroid, sleep apnea, dehydration, and low mood all produce identical tiredness and are common in the same population. Fatigue that persists beyond six to eight weeks at a stable dose with adequate intake warrants blood work rather than a protocol change.
How often did fatigue occur in the semaglutide and tirzepatide trials?
Roughly double the placebo rate, rising with dose
Fatigue occurred in about 11 percent of adults without diabetes on semaglutide 2.4 mg across the pooled STEP 1, 3, and 4 trials, against about 5 percent on placebo, according to the Wegovy prescribing information. The Zepbound label reports the pooled SURMOUNT-1 and SURMOUNT-2 figures for tirzepatide: about 5 percent at 5 mg, 5 percent at 10 mg, and 7 percent at 15 mg, against 3 percent on placebo. In both programmes the events were overwhelmingly mild to moderate, few led to discontinuation, and the label groups fatigue with the other adverse reactions that peaked during dose escalation and declined at the maintenance dose.
The tirzepatide figures show a dose response, with the 15 mg arm about 40 percent more likely to report fatigue than the 5 mg arm. That pattern matches the intake data: the higher the dose, the larger the appetite suppression, and the larger the calorie gap the body has to absorb. The semaglutide figure is higher than any tirzepatide arm, which may reflect the more aggressive gastrointestinal profile of semaglutide at 2.4 mg, the different trial populations, or simply the way the two sponsors coded the symptom. Adverse event tables are not designed for head to head comparison.
These are full dose figures in trials that pushed every participant to the maximum tolerated dose on a fixed schedule. A microdose protocol runs at a quarter to a half of the labelled maintenance dose and typically holds there rather than climbing. There is no trial of fatigue at that exposure, but the dose response in the tirzepatide arms and the mechanism described below both suggest the rate would sit closer to the placebo line than to the 11 percent.
What Percent of Trial Participants Reported Fatigue?
Approximate percent reporting fatigue as an adverse event, pooled STEP 1, 3, and 4 (68 weeks) and pooled SURMOUNT-1 and 2 (72 weeks), from the Wegovy and Zepbound US prescribing information
The excess over placebo is about 6 points for semaglutide and 2 to 4 points for tirzepatide. Both programmes escalated every participant to the target dose on a fixed schedule, which is the pattern that produces the most fatigue reports. Adverse event tables from different trials are not directly comparable.
Where does the tiredness actually come from?
A deficit the appetite signal no longer reports
The tiredness comes from a sudden drop in energy intake that the body has not yet adapted to. GLP-1 receptor agonists act on appetite centres in the hypothalamus and brainstem, slow gastric emptying, and reduce the reward value of food, and the combined effect on how much people eat is large. Blundell and colleagues measured it directly in 2017: after 12 weeks of semaglutide 1 mg, ad libitum energy intake at a test lunch fell by about 35 percent and total daily intake across three meals and snacks fell by about 24 percent compared with placebo, with no compensatory increase later in the day. A later crossover study by Heise and colleagues in 2023 found a reduction of similar size on tirzepatide. For a person who was eating 2,450 calories a day, a 24 percent cut is roughly 600 calories, arriving without any sensation of restriction.
The composition of what disappears matters as much as the amount. Appetite suppression on a GLP-1 tends to fall hardest on the foods people stop wanting first, which in practice are bread, rice, pasta, sweets, and alcohol, so carbohydrate intake falls faster than protein or fat. Glycogen, the stored form of carbohydrate in liver and muscle, is the fuel the brain and working muscles reach for first, and a fed adult carries about 400 to 500 grams of it. At an intake below 50 to 80 grams of carbohydrate a day that store halves within three or four days. Until fat oxidation and ketone production rise to fill the gap, a process that takes one to three weeks, the person feels exactly the way people feel in the first week of a ketogenic diet, because metabolically that is what is happening.
Fluid and sodium fall alongside food. Roughly a fifth of daily water intake normally comes from food, glycogen is stored with about three grams of water per gram, and nausea in the first weeks reduces drinking further. The result is a mild volume depletion that shows up as heaviness, lightheadedness on standing, and a heart rate that runs a few beats faster than usual. The hydration article on this site covers the numbers. Sleep disruption from nausea or reflux, a cut in caffeine that often happens without the person noticing because coffee has lost its appeal, and lower protein intake that begins to cost muscle all layer on top. None of these is a direct action of the drug on energy production. All of them are consequences of eating and drinking less than the body is used to.
How Does the Daily Energy Budget Shift After Starting a GLP-1?
Modelled daily intake and expenditure in kcal for a 90 kg adult, using the roughly 24 percent intake reduction measured by Blundell et al 2017 and a gradual partial recovery of appetite, with expenditure falling as weight is lost
- Energy intake
- Total expenditure
The gap between the bars is the daily deficit. It is largest in the first month, near 800 kcal, which is where most fatigue reports fall, and narrows as appetite partly recovers and the body adapts. The values are a model built on the published intake reduction rather than measurements from a single person.
When does the fatigue start and how long does it last?
Two weeks after each dose step, then it fades
Fatigue starts within the first week of a new dose, peaks around days 7 to 14, and fades over the following two to four weeks. The pattern follows the drug's pharmacokinetics and the body's adaptation to the new intake. Semaglutide and tirzepatide have half lives near a week, so each dose step produces rising blood levels for about four weeks before a new plateau, and the appetite effect is strongest in the first half of that climb before partial tolerance to the gastric effects develops. The missed dose article on this site walks through the steady state arithmetic. Meanwhile, glycogen depletes over the first three or four days and fat oxidation rises over the following one to three weeks, so the metabolic adjustment lags the drug by about a fortnight.
The standard labelled schedules escalate every four weeks for semaglutide, from 0.25 mg through 0.5, 1.0, and 1.7 to 2.4 mg, and every four weeks for tirzepatide from 2.5 mg in 2.5 mg steps. Each step restarts the cycle: a new rise in drug level, a new drop in intake, a new dip in energy, and a new two week recovery. Someone on the full schedule experiences four or five of these cycles in the first five months, and the trial adverse event curves show fatigue and nausea reports spiking at each one. A protocol that reaches a low dose and stays there goes through the cycle once.
Fatigue that persists beyond six to eight weeks at a stable dose has a different character and usually a different cause. By that point the drug level has plateaued, gastric tolerance has developed, and metabolic adaptation is complete. Tiredness that has not lifted is telling the person that intake has stayed too low, that protein or carbohydrate has not been restored, that sleep is still disrupted, or that something unrelated to the drug is going on. That is the point to measure rather than wait.
How Does Fatigue Track the Dose Escalation Schedule?
Modelled percent of people reporting noticeable fatigue each week on a standard four step escalation (increases at weeks 5, 9, 13, and 17) versus a protocol held at a low dose from week 1
- Standard escalation
- Held at a low dose
The grey line spikes after each dose increase and recovers between them. The green line has a single early bump and then settles. The shape is modelled from the adverse event timing described in the STEP and SURMOUNT publications; the trials did not publish week by week fatigue rates, so the values are illustrative.
What is energy availability and where is the floor?
30 kcal per kilogram of fat free mass
Energy availability is the energy left for the body's basic functions after the cost of exercise is subtracted, expressed per kilogram of fat free mass, and the floor sits at about 30 kcal per kilogram per day. Loucks and Thuma established the threshold in 2003 by feeding exercising women controlled amounts of energy for five days and measuring luteinising hormone pulses, which track how the hypothalamus reads the energy supply. Below 30 kcal per kg of fat free mass the pulse pattern was disrupted; above it, it was normal. The same threshold has since been linked in the sports medicine literature to low thyroid hormone, low testosterone in men, poor recovery, disturbed sleep, feeling cold, and persistent fatigue, a cluster now called relative energy deficiency in sport, or RED-S.
The arithmetic is simple and worth doing. A person with 55 kg of fat free mass who eats 2,200 calories and trains off 300 has an energy availability of about 35 kcal per kg, which is comfortable. The same person eating 1,800 calories sits at 27, just under the line. At 1,400 calories with the same training, which is a common intake on a GLP-1 in the first month, the figure is 20, and at 1,100 calories it is 14. Those lower values are the territory in which endurance athletes develop the fatigue, mood, and hormonal problems that define RED-S, and a GLP-1 can put a sedentary office worker there without any deliberate restriction at all.
The point is not that a deficit is wrong. Fat loss requires one. The point is that the size of the deficit has a floor below which the body stops cooperating, and appetite suppression removes the signal that would normally prevent someone from going through it. A rough weekly check, total calories minus training calories divided by fat free mass, is the simplest way to see whether tiredness is a normal adaptation dip or a sign that intake has fallen too far. Fat free mass can be estimated from a body composition scale or from the DEXA figures discussed in the tracking article on this site.
Where Do Common Intake Patterns Fall Against the Energy Availability Threshold?
Energy availability in kcal per kg of fat free mass per day for a person with 55 kg of fat free mass, across five modelled intake and training combinations, against the 30 kcal/kg threshold from Loucks and Thuma 2003
Only the first scenario clears the line. The 1,400 calorie rows are typical of the first month on a GLP-1 at full dose and sit well inside the range where persistent fatigue, poor recovery, and hormonal suppression are expected. The threshold was defined in exercising women; the same figure is used as a working guide for men.
Does eating more carbohydrate help, or does it slow fat loss?
A partial store keeps the lights on
Eating 100 to 150 grams of carbohydrate a day helps with early fatigue and does not meaningfully slow fat loss, because fat loss is set by the total deficit rather than by carbohydrate restriction. A fed adult holds about 100 grams of glycogen in the liver and 300 to 400 grams in muscle. The liver store is what keeps blood glucose steady between meals and overnight, and it runs down within 24 hours of very low intake. Muscle glycogen depletes more slowly but is the fuel for anything above a walking pace. At 50 grams of carbohydrate a day both stores sit near their floor by day four; at 130 grams they stabilise at 80 to 85 percent of fed levels, which is enough for normal daily function and moderate training while the deficit does its work on fat.
The practical problem on a GLP-1 is that carbohydrate is the macronutrient that disappears by default. The protein article on this site sets out why protein has to be planned deliberately on these drugs; carbohydrate needs the same attention for a different reason. A person eating 1,500 calories made up of eggs, chicken, cheese, and salad, which is a very common pattern once bread and rice stop appealing, may be taking in 40 grams of carbohydrate without intending a ketogenic diet, and will feel the way people feel in the first week of one. Adding a serving of oats, potato, fruit, or rice to two meals a day usually resolves the flatness within a few days.
Anyone who has chosen a ketogenic approach alongside the drug should expect the same one to three week adaptation window that keto produces on its own, and should pay particular attention to sodium, which the kidneys excrete faster as insulin falls. The metabolic rate article covers the longer term question of adaptive thermogenesis, which is a separate and slower process from the early glycogen dip and is not fixed by carbohydrate.
How Fast Do Glycogen Stores Fall at Different Carbohydrate Intakes?
Modelled total body glycogen in grams over 14 days at about 50 g and about 130 g of carbohydrate a day, from a fed store of 500 g, with partial recovery as fat oxidation rises
- About 130 g carbohydrate a day
- About 50 g carbohydrate a day
The low carbohydrate line loses more than half its store by day four and only partly recovers. The moderate line dips slightly and holds. Each gram of glycogen is stored with about 3 grams of water, so the early weight drop on the low line includes roughly 700 g of water that returns when carbohydrate does. The curves are modelled from low carbohydrate feeding studies rather than measured in GLP-1 users.
What else causes fatigue that a GLP-1 can hide?
Six causes with the same symptom and a different fix
Several common conditions produce tiredness that is indistinguishable from the intake related fatigue described above, and starting a GLP-1 gives the person a ready explanation that can delay finding them. Each of the following is more common in people carrying excess weight than in the general population, and each is straightforward to check.
Ferritin below about 30 ng/mL produces fatigue before anemia appears, and reduced red meat intake on a GLP-1 lowers iron supply just as the deficit raises demand. The micronutrient article covers the numbers. A ferritin and full blood count settle it.
Metformin, acid suppressants, and very low animal protein intake all lower B12, and the fatigue it produces is slow and easy to attribute to the drug. A serum B12, with methylmalonic acid if the result is borderline, is the check.
Hypothyroidism is common, produces tiredness, weight gain, feeling cold, and low mood, and can be unmasked when weight loss changes levothyroxine requirements. A TSH with free T4 is the test, and the thyroid article on this site covers the dose interaction.
Untreated sleep apnea causes daytime fatigue regardless of diet, and it is present in a large share of people who start these drugs. It improves with weight loss but does not vanish in the first months. Loud snoring, witnessed pauses, or morning headaches warrant a sleep study.
Fatigue, poor concentration, and loss of interest overlap heavily with the flat feeling of a large deficit. A persistent low mood lasting more than two weeks, or any change in thoughts about self harm, should be raised with a prescriber directly rather than attributed to intake. The mental health articles on this site cover what the trials found.
Lightheadedness on standing, a faster resting heart rate, dark urine, and a headache alongside the tiredness point to volume depletion. Two to three litres of fluid with a gram or two of added sodium a day usually resolves it within 48 hours. A home blood pressure reading sitting and standing confirms the pattern.
What is worth doing in the first month, and what is worth raising with a prescriber?
Five points, none of which are a protocol
Everything below is context for a conversation rather than an instruction. The fatigue rates quoted apply to the approved products at labelled doses in trial populations, and any low dose or compounded regimen needs its own assessment from the prescriber who wrote it.
The single most useful piece of information when someone feels flat on a GLP-1 is how many calories, how much protein, and how much carbohydrate they actually ate. Most people are surprised. A phone app for 14 days is enough to see whether intake has fallen below 1,200 to 1,400 calories without anyone deciding it should.
The two macronutrients that fall hardest on a GLP-1 are the two that most affect how the person feels. A serving of starch at two meals and the protein target from the protein article prevent the glycogen dip and the muscle cost. Fat loss depends on the total deficit, not on removing carbohydrate.
Fluid intake falls with food intake, glycogen water is lost in the first week, and coffee often stops appealing. Replacing the fluid and salt, and keeping caffeine at whatever level was normal before, removes two of the most common contributors within a few days.
Calories eaten minus training calories, divided by fat free mass in kilograms. Below 30 for more than a couple of weeks is the range where fatigue becomes persistent and hormonal changes begin. A reading in the low 20s is a reason to raise intake regardless of how the scale looks.
A full blood count, ferritin, B12, TSH with free T4, and a basic metabolic panel cover the common alternative causes in one draw. Fatigue that has outlasted the adaptation window with adequate intake is more likely to be one of these than the drug.
What is genuinely unresolved here?
Three gaps worth naming
The trial fatigue rates are counts of spontaneous reports, not measurements. Nobody in STEP or SURMOUNT filled in a validated fatigue scale every week, and the difference between a participant who mentions tiredness to the study nurse and one who does not depends on culture, expectation, and how the question was asked. The 11 percent and the 5 to 7 percent are real signals against placebo, but the true proportion of people who feel meaningfully less energetic in the first month is probably higher than the adverse event tables suggest, and the week by week timing in this article is modelled from the published descriptions rather than taken from published curves.
The claim that the tiredness is almost entirely a consequence of reduced intake rather than a direct drug effect is well supported by the intake studies and by the fact that it resolves as intake normalises, but it has not been tested by holding intake constant. A study that fed GLP-1 and placebo participants identical diets and measured fatigue would separate the two, and it has not been done. GLP-1 receptors exist in the brain regions that regulate arousal and reward, and a small direct contribution cannot be excluded.
The energy availability threshold comes from exercising women studied over five days, and its application to men, to sedentary people, and to weeks rather than days is an extrapolation that the sports medicine field has adopted as a working guide rather than a proven boundary. It is a useful number for spotting an intake that has drifted too low. It is not a precise line, and the same person may feel fine at 27 one month and flat at 32 the next.
Frequently Asked Questions
Selected References
- Novo Nordisk. Wegovy (semaglutide) injection, US prescribing information. Section 6.1, adverse reactions in pooled STEP 1, 3, and 4, including fatigue.
- Eli Lilly and Company. Zepbound (tirzepatide) injection, US prescribing information. Section 6.1, adverse reactions in pooled SURMOUNT-1 and SURMOUNT-2 by dose, including fatigue.
- Wilding JPH, Batterham RL, Calanna S et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). New England Journal of Medicine, 2021.
- Jastreboff AM, Aronne LJ, Ahmad NN et al. Tirzepatide once weekly for the treatment of obesity (SURMOUNT-1). New England Journal of Medicine, 2022.
- Blundell J, Finlayson G, Axelsen M et al. Effects of once-weekly semaglutide on appetite, energy intake, control of eating, food preference and body weight in subjects with obesity. Diabetes, Obesity and Metabolism, 2017. Ad libitum energy intake reduced by about 24 percent.
- Heise T, DeVries JH, Urva S et al. Tirzepatide reduces appetite, energy intake, and fat mass in people with type 2 diabetes. Diabetes Care, 2023.
- Loucks AB, Thuma JR. Luteinizing hormone pulsatility is disrupted at a threshold of energy availability in regularly menstruating women. Journal of Clinical Endocrinology and Metabolism, 2003.
- Mountjoy M, Ackerman KE, Bailey DM et al. International Olympic Committee consensus statement on relative energy deficiency in sport (REDs). British Journal of Sports Medicine, 2023.
- Phinney SD, Bistrian BR, Evans WJ, Gervino E, Blackburn GL. The human metabolic response to chronic ketosis without caloric restriction: preservation of submaximal exercise capability with reduced carbohydrate oxidation. Metabolism, 1983. Glycogen depletion and the adaptation window.
- Bergström J, Hermansen L, Hultman E, Saltin B. Diet, muscle glycogen and physical performance. Acta Physiologica Scandinavica, 1967. Muscle glycogen stores and depletion at low carbohydrate intake.
- Olsson KE, Saltin B. Variation in total body water with muscle glycogen changes in man. Acta Physiologica Scandinavica, 1970. Water stored with glycogen.
- Wharton S, Davies M, Dicker D et al. Managing the gastrointestinal side effects of GLP-1 receptor agonists in obesity: recommendations for clinical practice. Postgraduate Medicine, 2022. Timing of adverse events around dose escalation.
- Drucker DJ. Mechanisms of action and therapeutic application of glucagon-like peptide-1. Cell Metabolism, 2018. Central appetite pathways and GLP-1 receptor distribution.
This article is for general educational purposes only and does not constitute medical advice, diagnosis, treatment, or a recommendation to start, stop, switch, or change the dose of any medication or diet. Semaglutide, tirzepatide, and all other GLP-1 receptor agonists are prescription medicines, and the choice of product, dose, and any low dose or microdosing approach should be decided and monitored by a qualified healthcare provider. The fatigue rates quoted here are taken from US prescribing information and published trials of the approved products at labelled doses in trial populations and do not apply to compounded products or to any regimen the labels do not describe. The intake, energy availability, and glycogen figures are illustrative models built on published studies in other populations and are not measurements from GLP-1 users. Nothing here is intended to diagnose, prevent, treat, or cure obesity, fatigue, iron deficiency, thyroid disease, depression, sleep apnea, or any other condition, and no low dose or microdosing approach described in this article has been shown in trials to be safe or effective outside the specific trial conditions cited. Anyone with persistent fatigue, fainting, chest pain, confusion, a persistently low mood, or thoughts of self harm should seek medical care. Individual results vary.
Smaller Doses. Sharper Protocols.
Metabolic health you can measure, built around a dose designed to work with your body rather than against it.
Find Your Protocol