GLP-1 — What Your Prescriber Didn't Tell You

This is a composite of several clients I’ve seen, not one person. She had been on semaglutide (Ozempic) for eight months and had lost 18kg. Her GP was delighted. She was not. Her hair was falling out in handfuls, she could barely tolerate solid food, her energy had collapsed, and she had lost the muscle definition that years of regular training had built. Her blood results, which nobody had checked during the eight months of treatment, told a clear story: ferritin 8 µg/L, B12 214 pmol/L, zinc 9.4 µmol/L, albumin borderline low, HbA1c now well within range but DEXA scan showing significant lean mass reduction alongside the fat loss.

She had achieved what the drug promised. She had also paid a price that her prescriber had not mentioned — because nobody had tested her before she started, nobody had monitored her during treatment, and nobody had considered what would happen to her nutritional status when her appetite was suppressed to the point where she was eating less than 1,000 calories on most days.

This is not an argument against GLP-1 receptor agonists. The evidence that they work is real and compelling. This is an argument for using them with clinical intelligence — testing before, monitoring during, supporting the body through the intervention, and planning for after. The drug does one job very well. The work around it is what determines whether the outcome is genuinely beneficial or temporarily impressive.

What GLP-1 actually is — and what these drugs do to the system

GLP-1 (glucagon-like peptide-1) is an incretin hormone produced by L-cells in the small intestine and colon in response to nutrient arrival. Its natural role is brief and pulsatile: it is released for minutes after a meal, stimulates insulin secretion in a glucose-dependent manner, suppresses glucagon, slows gastric emptying, and sends satiety signals to the hypothalamus. It is then broken down rapidly by the enzyme DPP-4. The normal half-life of endogenous GLP-1 is approximately 2 minutes.

GLP-1 receptor agonists are synthetic molecules designed to resist DPP-4 breakdown, producing sustained GLP-1 receptor activation for hours or days. Semaglutide (Ozempic, Wegovy) has a half-life of approximately 7 days. Tirzepatide (Mounjaro, Zepbound) activates both GLP-1 and GIP receptors simultaneously. The body goes from receiving brief, meal-triggered pulses of GLP-1 to receiving continuous, supraphysiological stimulation 24 hours a day. The effects are powerful. So are the downstream consequences.

Drug	Brand names	Mechanism	Half-life	Average weight loss
Semaglutide	Ozempic (diabetes) · Wegovy (weight)	GLP-1 receptor agonist	~7 days · weekly injection	15–17% body weight over 68 weeks (STEP trials)
Tirzepatide	Mounjaro (diabetes) · Zepbound (weight)	Dual GLP-1 + GIP receptor agonist	~5 days · weekly injection	20–22% body weight (SURMOUNT trials) — currently highest efficacy available
Liraglutide	Victoza (diabetes) · Saxenda (weight)	GLP-1 receptor agonist	~13 hours · daily injection	8–10% body weight — lower than semaglutide; largely superseded
Dulaglutide	Trulicity	GLP-1 receptor agonist	~5 days · weekly injection	Primarily diabetes management · modest weight effect

The side effects your prescriber should have covered

The clinical trial data on GLP-1 agonists is genuinely impressive for weight loss and glycaemic control. It is less prominently discussed on the adverse effects side — not because the data is absent, but because the headline weight loss numbers dominate the conversation. The following is the clinical picture from the published evidence.

Serious · Muscle Loss

Part of the weight lost is lean mass

Across 22 randomised trials, about a quarter of the weight lost on GLP-1 drugs was lean mass rather than fat, and the most potent drugs (semaglutide 2.4 mg, tirzepatide 15 mg) preserved lean mass least well (Karakasis 2024, network meta-analysis). Two things usually get lost in the retelling: lean mass is not the same as muscle (it also includes water, organs and connective tissue), and substantial weight loss by any route, diet alone included, takes some lean mass with it. It still matters, because muscle is the main site of glucose disposal, and it is the reason to protect it with resistance training and adequate protein throughout.

Serious · GI Complications

Gallbladder problems, and slowed stomach emptying

A meta-analysis of 76 randomised trials found gallbladder and biliary disease about 37% more common on GLP-1 drugs overall, and more than twice as common in the weight-loss trials, with higher doses and longer use carrying more risk (He et al., JAMA Intern Med 2022, doi:10.1001/jamainternmed.2022.0338). Slower stomach emptying is part of how the drugs work; true gastroparesis is uncommon, and reports of it persisting after stopping are rare case reports rather than a trial finding.

Common · GI Symptoms

Nausea, vomiting, diarrhoea — common, mostly early

In the pooled STEP 1–3 trials, people on semaglutide 2.4 mg reported nausea (43.9%, against 16.1% on placebo), diarrhoea (29.7% vs 15.9%), vomiting (24.5% vs 6.3%) and constipation (24.2% vs 11.1%), mostly mild to moderate and during dose increases; 4.3% stopped because of them (Wharton 2022, a Novo Nordisk–funded analysis). In STEP 1 itself, 4.5% stopped for gastrointestinal effects against 0.8% on placebo (Wilding et al., NEJM 2021).

Important · Nutritional Depletion

Nutritional shortfalls when intake falls very low

For some people the appetite suppression is strong enough that intake falls well below what is needed to cover micronutrients, whatever the food quality. Protein, zinc, B12, iron, magnesium, and fat-soluble vitamins (A, D, E, K) are all at risk. In my view this is the most under-monitored consequence of GLP-1 therapy in current UK prescribing practice.

Monitor · Thyroid

Thyroid C-cell tumours — animal data, human uncertainty

Rodent studies showed dose-dependent thyroid C-cell hyperplasia and medullary thyroid carcinoma with GLP-1 agonists. Human data has not confirmed this risk at therapeutic doses, but all GLP-1 agonists carry a black box warning. Contraindicated in those with personal or family history of medullary thyroid carcinoma or MEN2. Baseline calcitonin is recommended by some guidelines before initiation.

Monitor · Rebound

Most of the weight returns after stopping

In the STEP 1 extension, one year after stopping semaglutide (and the trial’s lifestyle programme), participants had regained about two-thirds of the weight they had lost (Wilding et al., Diabetes Obes Metab 2022, doi:10.1111/dom.14725). The drug suppresses appetite and slows gastric emptying through continuous receptor activation. When it stops, the receptor activation stops. The underlying metabolic drivers — insulin resistance, dysbiosis, nutrient depletion, HPA dysfunction — that contributed to the original weight gain remain unaddressed.

The Lean Mass Question — Why the Weight Number Is Not the Whole Story

~25%

of weight lost across 22 trials was lean mass, more with the most potent drugs — not all of it muscle (Karakasis 2024)

Muscle is not cosmetic. It is the primary site of glucose disposal in the body — responsible for approximately 80% of insulin-stimulated glucose uptake. If part of a 10kg loss is muscle, the body has less glucose-clearing tissue even as its glycaemic markers improve. The HbA1c can look better while the muscle that supports it long term has shrunk, which is why body composition, not just weight, is worth measuring. Almost nobody gets a DEXA scan before or during GLP-1 therapy. This is a significant gap in standard prescribing practice.

What your pancreas was actually trying to tell you

It is often said that people with obesity or type 2 diabetes have a blunted GLP-1 response to meals, and that the drugs simply replace what the gut has stopped making. The evidence is more mixed than that. A meta-analysis of 22 studies found that people with type 2 diabetes generally released similar amounts of GLP-1 after a meal or glucose drink to people without it, although worse blood sugar control went with lower release (Calanna 2013). So a “GLP-1 deficiency” is not a given.

What the drugs do is different in kind: they give continuous, far-above-normal stimulation of the receptor, where the body’s own GLP-1 arrives in short pulses after meals. Appetite, fullness and blood sugar are still shaped by food, sleep, activity and the gut, and those are worth working on whether or not GLP-1 is the main route.

"A GLP-1 agonist gives you far more of a hormone your gut already makes in short bursts. The drug works. But it doesn't change why appetite and blood sugar were hard to manage in the first place. That is the conversation that almost never happens at the point of prescribing."

Supporting appetite and fullness without the drug

Before a prescription, and alongside one where it is clinically appropriate, these are reasonable things to do. Each has its evidence grade. The evidence for the habits is generally better than the evidence that they work specifically through GLP-1.

Appetite and fullness — what to do, with the evidence

Habits that support fullness, whether or not GLP-1 is the main route

Protein at every meal — Protein is the most filling of the macronutrients, and a protein-rich first meal is a sensible place to start (25–40 g a meal is a practical target, not a trial-tested threshold). For whey taken before a meal and its effect on insulin and GLP-1, see whey, insulin and GLP-1. Evidence: consistent for fullness in short trials; the GLP-1 share of that is less certain.

Whole foods over liquid and puréed ones — In a study of 58 adults, eating a whole apple before lunch cut total intake at the meal by 15% compared with no preload, and whole apple was more filling than apple sauce or apple juice matched for calories (Flood-Obbagy & Rolls 2009). Smoothies and juices leave the stomach faster than the same food eaten whole. Evidence: small controlled study, short term; it measured fullness and intake, not GLP-1.

Fermentable fibre and resistant starch — Gut bacteria ferment fibre to short-chain fatty acids such as butyrate and propionate, which stimulate the L-cells that release GLP-1 in laboratory and animal studies. Oats, legumes, cooked-and-cooled potato or rice, green banana. Evidence: mechanistic and untested in people for GLP-1; fibre is worth eating for other, better-supported reasons.

Akkermansia — In a pilot trial, 40 overweight or obese insulin-resistant adults were randomised and 32 finished; pasteurised Akkermansia muciniphila for three months improved insulin sensitivity and lowered insulin compared with placebo (Depommier 2019). GLP-1 was not one of its outcomes. Polyphenol-rich foods and resistant starch are linked to more Akkermansia, mostly in animal studies. Evidence: one small pilot trial.

Sleep — One night of fragmented sleep lowered GLP-1 and fullness in a 24-hour study of 12 healthy young men (Gonnissen 2012). Studies of sleep loss and GLP-1 have not all agreed. Evidence: small crossover studies, inconsistent.

Resistance training — The main protection for muscle while weight is coming off, and muscle is where most glucose goes after a meal. Evidence: well supported for protecting muscle during weight loss; trials of training alongside GLP-1 drugs are few.

Berberine — A meta-analysis of 46 trials, many small and many from Chinese databases, found berberine lowered HbA1c in type 2 diabetes, alone or with standard treatment (Guo 2021). Claims that it works by raising GLP-1 come mainly from animal studies. It interacts with several medicines: check with your prescriber or pharmacist first. Evidence: mixed (many small trials of variable quality).

Bitter foods before meals — Bitter-taste receptors in the gut wall can trigger GLP-1 release in laboratory studies. Whether bitters before a meal change GLP-1 or appetite in people has not been shown. Evidence: mechanistic and untested.

Gut symptoms — If you have gut symptoms, they are worth investigating. A stool test can add information, but its microbiome readings are not validated as a guide to GLP-1 or weight, and detections are leads to confirm rather than diagnoses.

What to test before, during, and after GLP-1 therapy

The following testing framework represents the minimum clinical standard for anyone initiating, currently on, or discontinuing GLP-1 receptor agonist therapy. In current UK prescribing practice, most of this is not done.

Timing	Tests	Why
Before starting	Full blood chemistry: HbA1c, fasting insulin, fasting glucose, full lipids, liver enzymes (GGT, ALT, AST), ferritin, B12, folate, zinc, vitamin D, albumin, thyroid panel (TSH + Free T3), calcitonin (if thyroid risk), DEXA if available.	Establishes the baseline metabolic and nutritional picture. Identifies contraindications. Documents starting lean mass. Reveals existing deficiencies that will worsen under appetite suppression.
3 months into treatment	Repeat: ferritin, B12, zinc, albumin, vitamin D. HbA1c and fasting insulin. Liver enzymes. Kidney function (eGFR, creatinine). Body weight AND body composition if possible.	Nutritional depletion is fastest in the first 3 months when appetite suppression is most severe during dose titration. Catching deficiencies early prevents the clinical picture my client presented with at 8 months.
Ongoing (6-monthly)	Full nutritional panel. Muscle mass monitoring — DEXA or at minimum bioimpedance. HbA1c. Liver and kidney function.	Sustained monitoring for cumulative depletion, lean mass loss trajectory, and organ function. Most prescribers do HbA1c only. The nutritional picture requires a separate, deliberate assessment.
Before discontinuing	DUTCH Plus (cortisol pattern, insulin-related markers). Full nutritional panel. GI-MAP (gut microbiome picture). Blood chemistry complete.	Understanding what underlying drivers remain active before stopping — otherwise rebound is almost certain. The gut, the HPA axis, the insulin resistance, the nutritional deficiencies all need to be addressed as part of a planned exit strategy, not discovered after the weight returns.
After stopping	Monthly weight and composition monitoring. Nutritional panel at 3 months. Full metabolic panel at 6 months.	The rebound window. Proactive monitoring allows early intervention if the metabolic picture is deteriorating before significant weight regain has occurred.

The exit strategy — why stopping without a plan produces the worst outcome

The two-thirds regain is not a failure of willpower. It is a predictable physiological consequence of removing a continuous GLP-1 receptor activation without having addressed any of the underlying drivers that led to the weight gain in the first place. The drug works as an override. When the override is removed, the underlying system reasserts itself.

A planned exit strategy addresses: the gut microbiome (restore the fibre and fermented food substrate for natural GLP-1 production), the HPA axis (chronic stress drives cortisol-mediated fat storage and appetite dysregulation), insulin resistance (not resolved by weight loss alone — requires dietary structure, exercise, and often targeted supplementation), nutritional repletion (ferritin, B12, zinc, vitamin D all need restoring before discontinuation), and the muscle loss (resistance training during the drug phase, not just after, is the single most important protective intervention).

None of this is complicated. All of it requires planning. Almost none of it happens in standard prescribing consultations.

The Functional Medicine Position on GLP-1 Agonists

These are powerful and effective drugs. The evidence for weight loss and cardiovascular risk reduction is real. The appropriate clinical response is not to dismiss them — it is to use them intelligently. That means testing before, monitoring nutritional status throughout, protecting muscle mass with resistance training, addressing the gut-microbiome substrate for natural incretin function, and planning the exit before initiating. A GP prescribing semaglutide in a ten-minute consultation without baseline bloods, DEXA, or a structured monitoring plan is not practising bad medicine by intention. The prescribing guidelines do not require most of this. The clinical intelligence does.

Why the scale moves fast at first, then stalls

Merged here from the former post “Before You Reach for Ozempic”. Most of the weight lost in the first week of any change in eating is glycogen, the stored form of glucose in liver and muscle, together with the water stored with it. When carbohydrate or total intake falls, glycogen is used first and that water goes with it; when eating returns to normal, or a new training programme starts, glycogen is restocked and the scale goes up without any fat being gained. Body composition or a consistent waist measurement tells you more than the scale over the first weeks, on a drug or off it.

Sources for this section

Calanna S et al. Secretion of GLP-1 in patients with type 2 diabetes: systematic review and meta-analyses. Diabetologia 2013;56:965–72. PMID 23377698
Wilding JPH et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). N Engl J Med 2021;384:989–1002. PMID 33567185
Wharton S et al. Gastrointestinal tolerability of once-weekly semaglutide 2.4 mg (pooled STEP 1–3). Diabetes Obes Metab 2022;24:94–105. PMID 34514682
Karakasis P et al. Effect of GLP-1 receptor agonists and co-agonists on body composition: systematic review and network meta-analysis. Metabolism 2025;164:156113. PMID 39719170
Flood-Obbagy JE, Rolls BJ. The effect of fruit in different forms on energy intake and satiety at a meal. Appetite 2009;52:416–22. PMID 19110020
Depommier C et al. Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study. Nat Med 2019;25:1096–1103. PMID 31263284
Gonnissen HKJ et al. Effects of sleep fragmentation on appetite and related hormone concentrations over 24 h in healthy men. Br J Nutr 2013;109:748–56. PMID 22682471
Guo J et al. The effect of berberine on metabolic profiles in type 2 diabetic patients: systematic review and meta-analysis of RCTs. Oxid Med Cell Longev 2021:2074610. PMID 34956436

Get the full metabolic picture before any weight intervention

Whether you are considering GLP-1 therapy, currently on it, or planning to stop — the blood chemistry, DUTCH cortisol pattern, and GI-MAP together give you the complete metabolic and nutritional picture that determines whether the intervention will produce durable results or temporary ones.

Book a Strategy Call →