No product label in the United States, the European Union or Singapore requires a blood test before starting GLP-1 medication for weight management. The only laboratory instructions in the US labels are conditional and narrow, and Singapore's public formulary does not publish a monitoring section at all. What tests a particular doctor orders is a clinical judgement about the person in front of them, which is a separate question from what a label requires. This article sets out what each document actually says — including the one test most people assume is mandatory, which the label argues against.
Monitoring instruction: a direction in a product label to measure something because of the medicine. It is a different thing from a screening recommendation, which is about assessing the person. Most of the confusion in this area comes from reading one as the other.
What the US labels actually instruct
Two things, and both are conditional.
The semaglutide label has a subsection headed "Important Monitoring and Administration Instructions". Its entire content is an instruction to monitor blood glucose before starting and during treatment in patients who already have diabetes mellitus. That is the label's only pre-treatment laboratory instruction, and it does not apply to someone without diabetes. Separately, the warnings section says to monitor renal function in patients reporting adverse reactions that could lead to volume depletion — a reactive instruction triggered by nausea, vomiting or diarrhoea, not a baseline. The same label asks for heart rate to be monitored at regular intervals, and for increased clinical or laboratory monitoring where the patient is also taking a medicine with a narrow therapeutic index (label).
The tirzepatide label carries two instructions and no more: renal function on the same volume-depletion trigger, and blood glucose in people who already have diabetes. It has no heart-rate warning section at all — a heart-rate increase of 1 to 3 beats per minute appears only in its adverse-reactions table (label). So the heart-rate instruction is confirmed for one of the two medicines, not for the class.
A direct full-text search of both labels returned zero matches for each of creatinine, liver function, HbA1c, thyroid function and TSH (label, label). There is no required liver panel, no required lipid panel, no required thyroid-function test and no required kidney test in either document.
One boundary before going further. Tirzepatide's current registrations carry a weight-management indication alongside the type 2 diabetes one; the 2023 injection registration SIN16718P carries type 2 diabetes only (Singapore formulary). Its label is quoted here as class-level safety information for a reader who may already be taking the medicine, not as something offered for weight management.
The test people assume is required is the one the label argues against
Calcitonin, and both US labels take the same position on it.
Verbatim, in both: "Routine monitoring of serum calcitonin or using thyroid ultrasound is of uncertain value for early detection of MTC" — medullary thyroid carcinoma — and such monitoring "may increase the risk of unnecessary procedures, due to the low-test specificity for serum calcitonin and a high background incidence of thyroid disease" (labels).
The same paragraph gives 50 ng/L as the level at which medullary thyroid carcinoma becomes a consideration. Read the sentence carefully and it is conditional: if serum calcitonin has been measured, for whatever reason, and is found to be elevated, the patient should be evaluated further. It is not a threshold anyone is being told to test against. The European product information for semaglutide goes further still — a full-text search of it returns no occurrence of calcitonin anywhere.
Which brings up the single easiest thing in this area to misread. The European product information for tirzepatide does list a rise in blood calcitonin. It lists it as a side effect — common, meaning up to 1 in 10, in people treated for weight management, and uncommon, up to 1 in 100, in adults treated for type 2 diabetes. Nowhere does that document instruct anyone to measure it (EU product information).
An observed adverse reaction and a monitoring instruction point in opposite directions, and converting one into the other is how a citation quietly becomes an invented protocol. If you have read that "Europe requires calcitonin monitoring", this is where that claim came from, and it is a misreading of a side-effect table.
The wider thyroid question — where the warning came from, and why the contraindication differs between countries — is covered in the thyroid warning on GLP-1 medication, explained.
What Singapore actually publishes
Four sections, and none of them is about monitoring.
The public formulary page for the registered semaglutide product publishes product information, indication, dosing and contraindications. A targeted search of that page for "monitoring", "blood test", "laboratory test", "calcitonin", "thyroid ultrasound", "renal function", "HbA1c", "blood glucose" and "lipids" returns none of those terms. Section 4.4 of the registered product information, which is where warnings and precautions live, and section 4.5, which covers interactions, are not published on the portal at all (Singapore formulary).
Two things follow, and it is worth separating them. There is no Singapore monitoring rule available to quote — the position is genuinely not public, rather than merely absent from a casual reading. And it does not follow from that silence either that Singapore has no requirement, or that the American or European position applies here by default. Singapore's registered product information differs from the US label in several confirmed places, so importing one jurisdiction's instructions into another is not safe.
So why do clinics take blood?
Because the recommendation that does exist is about screening the person, not monitoring the drug — and that distinction is the whole answer.
A US endocrinology guideline recommends that anyone with overweight or obesity be screened for the conditions that travel with it: fasting glucose and HbA1c for prediabetes and type 2 diabetes, a full lipid panel, blood pressure, and liver function tests for fatty liver, followed by imaging if transaminases are raised. The glucose and lipid recommendations carry the guideline's highest grade (Garvey 2016).
One absence in that guideline is worth pointing out, because it is the second test people assume is standard. Thyroid-function testing appears in it only inside a table of secondary causes of obesity — "TSH for suspected hypothyroidism" — and not as a screening recommendation for everyone (Garvey 2016). So a thyroid-function test is something a doctor may order because your history raises the question, not something either the guideline or the labels ask for as a matter of course.
That is a sensible reason to draw blood, and it has nothing to do with the medicine. Three caveats belong next to it. The guideline is from 2016, published before either of today's weight-management medicines existed. It is American, not Singaporean. And its single "before and during treatment" laboratory instruction belongs to a different drug — phentermine with topiramate — while its GLP-1 kidney caution names liraglutide in renal impairment rather than the medicines most readers are asking about (Garvey 2016).
Where GetLean stands on this
We do not do blood tests, and we do not order them.
GetLean is a video service. There is no phlebotomy in it and no laboratory attached to it, so the honest description of the consultation is the one the rules ask for: a registered doctor takes your history, goes through your medications, and assesses whether treatment is appropriate for you. That assessment is the requirement. A panel is not.
If you already have a recent health-screening report, it is worth having in front of you when you speak to the doctor — it is your document, and what is in it can change what the doctor asks about. If the doctor decides they want more information than a video consultation can give them, they will say so, and that is a normal step rather than a rejection.
What the labels report happening to blood results during treatment
Pancreatic enzymes rise, and both labels say a single raised reading means little on its own.
In weight-reduction trials, semaglutide produced a mean increase from baseline in amylase of 15% to 16% and in lipase of 39%, changes not seen in the placebo group. Tirzepatide produced mean increases of 20% to 25% in pancreatic amylase and 28% to 35% in lipase, against 2.1% and 5.8% in placebo-treated patients. Both labels then state, in almost identical words, that the clinical significance of these elevations is unknown in the absence of other signs and symptoms of pancreatitis (labels).
Neither label asks for routine amylase or lipase testing, and the reason is visible in the numbers: a raised enzyme is a common finding on this medicine, so a reading taken without symptoms is difficult to act on. What pancreatitis actually looks like, and when it needs urgent attention, is covered in pancreatitis and GLP-1 medication.
What to make of it if you are told a panel is required
Ask which document requires it.
None of the four labels checked in this article does, and Singapore's formulary publishes no monitoring section to require anything. A doctor may still have entirely good reasons for ordering blood tests: screening you for the conditions the guideline lists, establishing a baseline for something they intend to follow over time, or answering a specific question your history has raised. Those are clinical reasons, and they are better ones than a label requirement that does not exist.
The inverse is also worth saying. A clinic that orders no blood tests is not thereby cutting a corner that a label had specified, because no label specified one. What matters is whether a registered doctor has taken an adequate history and assessed you — which is a duty the rules do impose, and which we describe in what happens at a GLP-1 video consultation. If a doctor wants more information before prescribing, or wants a physical examination first, that is a normal step, not a rejection.
Who the medicine is not suitable for is a separate question again, and it is answered by the contraindications rather than by any test — see who should not take GLP-1 medication.
Individual circumstances vary. What should be checked, and when, is a decision for the doctor who has assessed you.
Common questions
Do you need a blood test before starting GLP-1 medication?
No product label in the United States, the European Union or Singapore requires one for weight management (semaglutide label, tirzepatide label). Whether a particular doctor orders blood tests is a clinical judgement about the person in front of them.
Is a thyroid test needed before GLP-1 medication?
Both US labels state that routine serum-calcitonin or thyroid-ultrasound monitoring is of uncertain value and may increase the risk of unnecessary procedures (labels). Neither label contains any thyroid-function or TSH testing instruction.
Why does the European leaflet mention calcitonin, then?
Because it lists a rise in blood calcitonin as a side effect of tirzepatide — common in people treated for weight management, uncommon in those treated for type 2 diabetes — rather than as something to measure (EU product information). A listed adverse reaction and a monitoring instruction are different things.
What does Singapore's product information say about monitoring?
Nothing that is publicly readable. The formulary publishes product information, indication, dosing and contraindications only, and a search of it for monitoring and laboratory terms returns none of them (Singapore formulary).
Should someone with overweight or obesity have blood tests anyway?
A US endocrinology guideline recommends screening anyone with overweight or obesity for related conditions — fasting glucose and HbA1c, a full lipid panel, blood pressure and liver function tests (Garvey 2016). That is screening the person rather than monitoring a medicine, and it is a 2016 American document.