Thyroid Medication Explained: T4, T3, Combination Therapy & Desiccated Thyroid

Thyroid Medication Should Not Be a Loyalty Test

A woman is diagnosed with hypothyroidism and leaves with a prescription. Sometimes that’s the end of the story — she starts the medication and within a couple of months feels like herself again.

Sometimes it isn’t. Her TSH comes back normal and she’s told the thyroid is handled, but she’s still tired in a way sleep doesn’t fix, still cold, still constipated, still losing more hair than seems reasonable. She is now in the position of being told her thyroid is fine while not feeling fine.

So she looks online, where she finds two camps that despise each other. One says levothyroxine is the only legitimate thyroid medication and interest in anything else means she’s read the wrong websites. The other says synthetic hormone is a pharmaceutical compromise, desiccated thyroid is what nature intended, and anyone still on Synthroid hasn’t found the right doctor.

Neither helps the person trying to feel better, because both start from a conclusion about the medication rather than a question about the patient. The more useful question is which form and amount of thyroid hormone replacement works best for this particular person — and how we would know.

First, What T4 and T3 Actually Do

T4 (thyroxine) is what your thyroid releases in the largest quantity. It’s long-acting — a half-life of about a week — which is why once-daily dosing produces steady levels. Think of it as a circulating reservoir that tissues draw on and convert into T3 where and when they need it.

One detail reframes the entire debate: most of the T3 in your body was never secreted by your thyroid. The gland makes roughly 85 micrograms of T4 daily and only about 6.5 micrograms of T3 directly; about 80% of the T3 you use comes from converting T4 in tissues (American Thyroid Association, 2014). That’s the physiological case for T4-based therapy, and it’s a good one — supply the reservoir, and let each tissue make what it needs.

Synthetic T4 medications include generic levothyroxine, Synthroid, Levoxyl, Unithroid, and Tirosint and Tirosint-SOL. Worth noting early: “levothyroxine” and “T4 therapy” are not synonyms. Someone can do poorly on one T4 product and considerably better on another.

T3 (triiodothyronine) is the more biologically active hormone — the one that engages receptors and drives metabolic effects. Its pharmacology differs in ways that matter: a half-life of roughly a day rather than a week, and a peak within one to two hours of a dose. It acts fast, and levels rise and fall through the day rather than sitting flat. Liothyronine (Cytomel) supplies it directly; desiccated thyroid contains it alongside T4. That doesn’t make T3 a problem hormone — it makes it a different tool with different dosing and monitoring requirements.

Where Free T3 Fits — and What It Is Not

We pay attention to Free T3, because it says something about the active hormone actually circulating that TSH and Free T4 don’t, and that matters when someone isn’t responding as expected.

But a higher Free T3 is not the goal of treatment. It’s one measurement in a larger picture, read alongside TSH, Free T4, symptoms and medication type. A low-normal Free T3 does not by itself prove poor conversion, and it is not a reason on its own to add T3.

Timing also changes the number, and this trips people up constantly. If your medication contains T3, the level in your blood peaks about one to two hours after you take it, then drifts down over the rest of the day. So a Free T3 drawn mid-morning, two hours after your pill, can come back looking high — and the same person, tested right before their next dose, can come back looking low. Neither result is wrong. They’re snapshots of different moments.

Which is why we want to know what time you took your medication before we read a Free T3, and why “my Free T3 was low” doesn’t tell us much on its own.

More on how the markers fit together in our article on thyroid testing.

The Options, and What Each One Is For

Levothyroxine monotherapy is the recommended first-line treatment across every major guideline, including the American Thyroid Association’s, for defensible reasons: it resolves symptoms in most people, has a long track record, doses predictably, and costs little. Many people take it, feel well, and never think about any of this again. When someone is doing well, there’s no reason to change anything — a significant part of good thyroid care is recognizing when to leave a working plan alone.

Our position isn’t that this is wrong. It’s that first-line for most and correct for everyone are different statements, and the interesting clinical work happens in the gap between them.

The available approaches are T4 alone, T3 added to T4, T3 alone (rarely), and desiccated thyroid, which contains both.

Levothyroxine vs Synthroid and Tirosint: Why Formulation Can Matter

Consistency Comes First

Generic levothyroxine, Synthroid, Levoxyl and Unithroid are all T4. That doesn’t mean patients experience them identically, and it definitely doesn’t mean the product in your bottle stays the same over time.

Which manufacturer’s levothyroxine you receive can change at the pharmacy counter based on supply contracts, usually without anyone telling you. In a large analysis of real-world U.S. data, roughly a quarter of patients had at least one formulation change within twelve months — and only about 81% of treated patients had a TSH in range at any given point, meaning roughly one in five people on thyroid medication is not currently well-replaced in one direction or the other (Bianco et al., Thyroid, 2023).

The practical response: know which product you’re taking, keep it consistent, and recheck labs about six weeks after any change in dose, brand, or manufacturer.

This isn’t a fringe worry. Three major endocrinology organizations — the American Association of Clinical Endocrinologists, the Endocrine Society and the American Thyroid Association — have jointly recommended the same thing, and their reasoning is worth understanding.

When the FDA decides two levothyroxine products can be swapped for each other, it tests them in healthy volunteers whose own thyroids still work. Those people are making their own thyroid hormone all the way through the test, which makes small differences between the two products harder to detect. And the testing doesn’t look at TSH at all — the one number clinicians actually use to judge whether a person’s thyroid is well replaced. Two products can clear that bar and still, in these organizations’ view, not be proven equivalent for the person taking one of them every morning.

What We See in Practice

In our practice, we generally favor a consistent branded T4 formulation such as Synthroid or Tirosint over generic levothyroxine, and when a patient isn’t doing well on T4 therapy, formulation is one of the things we’ll raise with the prescribing provider. We frequently see patients who report feeling better and more stable on a consistent branded product, even when the dose looks equivalent on paper.

That’s a pattern we see, not proof that generic levothyroxine is worse — and plenty of patients do very well on it. The large studies back that up: an analysis of nearly 16,000 adults found essentially identical rates of normal TSH among people who switched manufacturers and people who didn’t (Brito et al., JAMA Internal Medicine, 2022).

An average across 16,000 people, though, is a statement about a group, not about you. If your response changes after a pharmacy switch, or has never been good in the first place, formulation belongs on the list of things to revisit — alongside dose, timing and absorption.

Tirosint and Tirosint-SOL Are Two Different Products

Tirosint is a capsule containing a viscous liquid; its complete list of inactive ingredients is gelatin, glycerin and water — no lactose, dyes, or starch. Tirosint-SOL is an oral solution in single-dose ampules containing only glycerol and water, available in a wider range of strengths for finer dose adjustment.

Both matter for the same reason: a tablet has to break down in the stomach before the hormone is available to absorb, and a liquid or gel capsule doesn’t. That makes absorption less dependent on what’s happening in your stomach — which is exactly the problem for some people. In a pooled analysis of six studies covering 141 patients whose TSH was off on tablets, switching to a liquid form at the same dose brought TSH down substantially (Virili et al., Frontiers in Endocrinology, 2018). Among patients on acid-blocking medications, that switch dropped average TSH from 5.4 to 1.7 (Vita et al., JCEM, 2014).

These are small studies, mostly comparing patients to themselves before and after a switch, and mostly from a handful of research groups. The direction is consistent and the mechanism makes sense. What they don’t show is any advantage for someone who absorbs levothyroxine normally — so this isn’t a general upgrade. The short ingredient list is a real advantage for a specific person, though: most levothyroxine tablets contain lactose and colorants, and some people react to them.

Sometimes the problem is not T4 itself. It may be the formulation, the absorption, the consistency of the product, or the fact that T4-only therapy isn’t producing the best response in that person — four different problems with four different solutions.

What Is the Best Time to Take Thyroid Medication?

Levothyroxine is absorbed in the small intestine, and food and several common substances reduce that absorption meaningfully. Inconsistent timing can look, on paper, exactly like an inadequate dose.

Morning versus bedtime has been studied properly. A randomized crossover trial found bedtime dosing lowered TSH modestly compared with morning — but no difference in how patients actually felt (Bolk et al., Archives of Internal Medicine, 2010). Other trials have favored morning. Neither wins, and what the evidence really points to is the variable that matters: the interval between the dose and food, not the position of the hands on the clock. The best time is whenever you can reliably take it on an empty stomach, separated from what interferes, day after day.

Coffee reduces absorption by roughly a third when taken together (Benvenga et al., Thyroid, 2008) — one of the more common unrecognized reasons a dose seems inadequate. Calcium and iron supplements bind levothyroxine in the gut; taking calcium carbonate alongside it for three months raised mean TSH from 1.6 to 2.7 mIU/L and moved a fifth of patients out of range (Singh et al., JAMA, 2000). Acid-suppressing medications, particularly proton pump inhibitors, reduce absorption by changing the stomach environment — commonly missed, because PPIs get taken long-term without anyone connecting them to a dose that keeps climbing. High-fiber foods, soy and certain binding medications matter too.

If timing has been inconsistent — and for most people it has been at some point — sort that out before concluding the medication isn’t working.

Why Do Some Patients Still Feel Hypothyroid on Thyroid Medication?

This is the question that brings most people to an article like this. It’s also more common than it sounds when you’re sitting in an exam room being told your labs are fine: somewhere around 5–10% of people on levothyroxine with a normal TSH still report significant symptoms. A minority, but a substantial one, and professional societies now acknowledge it rather than explaining it away.

Two things are true at once here, and most of the argument online comes from people holding only one of them: a normal TSH doesn’t prove that persistent symptoms are thyroid-related — and it doesn’t rule it out either. Here’s how we work through it.

The dose or the delivery may still be inadequate. Dose, timing, adherence, absorption, formulation, and whether body weight or physiology has changed since the dose was set. Unglamorous, and where a meaningful share of “treatment-resistant” cases actually resolve.

T4 alone may not be producing the response we want. Some people function better with a small amount of T3 in their replacement — examined below. But “poor converter” is a label that gets applied far too fast, and it isn’t a fixed personal trait. Conversion shifts with illness, inflammation, significant caloric restriction, certain medications, nutrient status and physiologic stress. One low-normal Free T3 in someone who’s been sick or under real strain is not a diagnosis.

Absorption may be impaired. Everything in the timing section applies, plus conditions that genuinely change how much gets in: celiac disease, atrophic gastritis, H. pylori, lactose intolerance, and altered anatomy after bariatric surgery. Atrophic gastritis with H. pylori raised required daily dose by roughly a third in one classic study (Centanni et al., NEJM, 2006). There’s an underused diagnostic point here — an unexplained, persistently high levothyroxine requirement is itself a clinical clue, and a reason to look for celiac disease or a gastric cause rather than continuing to raise the dose.

The symptom may not be coming from the thyroid at all. This one gets the least attention in functional medicine spaces and deserves the most. Fatigue, hair loss, brain fog, feeling cold and struggling with weight are what a long list of conditions look like from the outside: iron deficiency (sometimes with a ferritin that reads as technically normal), low B12 or folate, genuinely inadequate sleep, insulin resistance, perimenopause, not eating enough, gut problems, and several other hormone conditions. Having thyroid disease doesn’t protect you from also having one of these — if anything it makes them easier to miss, because there’s already an explanation on file.

We fairly often meet patients who are certain their medication has stopped working, when the thyroid is being replaced perfectly well and something else is driving how they feel. Adding more thyroid hormone in that situation doesn’t help, and it can do harm.

This is the part of thyroid care that isn’t really about the thyroid, and it’s a substantial part of what we evaluate.

How Long Does It Take to Know if a Medication Is Working?

Two timelines are running, and confusing them causes unnecessary discouragement.

Thyroid hormone levels take time to reach a new steady state after any change — roughly six weeks, based on T4’s week-long half-life. Recheck sooner and the number doesn’t yet mean what people think it does.

Symptoms lag further. Labs frequently normalize before a person feels normal; energy, cognition, hair and body composition can take months. That gap is one of the most common reasons people conclude a medication isn’t working when it still is.

When a T3-containing strategy is being evaluated, the window is longer still — the joint British Thyroid Association and Society for Endocrinology statement suggests a minimum of three to six months, with TSH kept in range, before deciding whether it helped (Ahluwalia et al., Clinical Endocrinology, 2023).

Our approach is to give a change enough time to be informative, then reassess deliberately — symptoms, the full panel, and the autoimmune picture together — rather than letting a plan drift without anyone deciding whether it worked.

Do Some People Need T3?

Some patients do appear to do meaningfully better with some T3 included. That isn’t the same as saying everyone needs it.

T3 can be supplied three ways: as separately prescribed synthetic T3 alongside T4, as part of a combination approach where each hormone is dosed independently, or through a desiccated thyroid product containing both.

One thing worth saying up front, because it runs against the internet’s instincts: when T3 helps, a small amount is usually what helps. Guideline-based approaches substitute a modest quantity of T3 for part of the T4 dose — the British Thyroid Association and Society for Endocrinology statement describes substituting liothyronine at roughly one-seventeenth of the current levothyroxine dose, a ratio anchored to human physiology rather than chosen arbitrarily.

What the Research Actually Says About T4 + T3

This is where most content online picks a side it can’t support. The honest answer is more interesting.

On average, combination therapy has not outperformed T4 alone. Between sixteen and twenty randomized trials have compared the two, and when researchers pool the results there’s no meaningful difference in quality of life, symptom scores, mood or memory and thinking (Millan-Alanis et al., Thyroid, 2021).

One recent trial tested the popular theory directly. Researchers gave 141 patients whose thyroids had been removed either combination therapy or a placebo, with nobody knowing who got what, and succeeded in shifting their T3-to-T4 balance closer to what a healthy thyroid produces. It made no difference to how they felt or to markers of thyroid activity in their tissues (Brigante et al., European Journal of Endocrinology, 2024). The “restore the natural ratio and people will feel better” idea got a fair test and didn’t pass it.

This is why guidelines don’t recommend combination therapy as routine treatment.

And yet patients in those same trials consistently preferred it. Pooling eleven randomized trials involving 1,135 patients, preference favored T3-containing therapy by roughly two to one — about 52% preferring combination, 24% preferring T4 alone, 24% with no preference (Beltrão et al., JCEM, 2025). The individual trials show the same shape: in a double-blind crossover comparing desiccated thyroid with levothyroxine in 70 patients, symptom and neurocognitive measures showed no difference, yet 48.6% preferred the desiccated thyroid and 18.6% preferred levothyroxine (Hoang et al., JCEM, 2013).

The most informative trial compared all three approaches head to head over 22 weeks each. Group outcomes were similar — but the patients who had been most symptomatic on levothyroxine were the ones who preferred and responded to the T3-containing options (Shakir et al., JCEM, 2021). That finding is contested and wasn’t the primary endpoint, but it’s the closest thing in the literature to an evidentiary basis for what many clinicians observe.

So the measurements say no difference and the patients say they can tell. Both are real results from the same studies, and there are at least four explanations, none established: the questionnaires may not capture what patients experience; preference may reflect feeling the T3 — warmth, a faster heart rate, a sense of activation — rather than being better on it; there may be a genuine subgroup benefit diluted in the average; or expectation may play a part. A large trial designed to answer this, enrolling patients who remain symptomatic on levothyroxine, is underway; results aren’t available yet.

Two conclusions are both wrong. That combination therapy is better and more people should be on it — the trials don’t support that, and the patients who don’t benefit are invisible in the enthusiastic version of the story. And that it’s useless because average outcomes were neutral — an average tells you what to do by default, not what to do for someone who has already failed the default.

One thing worth knowing before you ask about switching: changing preparations often doesn’t resolve symptoms either. The British Thyroid Association and Society for Endocrinology note that most patients still report persistent symptoms on liothyronine, combination therapy, or desiccated thyroid.

On genetic testing: you may see DIO2 gene testing marketed for this decision. The laboratory science is real, but whether the variant predicts who benefits from T3 hasn’t replicated reliably — a study of more than 12,000 people found no association with thyroid hormone levels, quality of life or cognition (Wouters et al., Thyroid, 2017) — and no major guideline recommends using it to guide treatment. We don’t select thyroid medication based on it.

When Combination Therapy Comes Up

The situations are fairly specific: persistent symptoms despite apparently adequate T4 treatment, a poor clinical response on T4 alone, a lab and symptom picture together suggesting some T3 may help, documented prior benefit from T3, or a need for finer control than a combined product allows. Its main practical advantage over a glandular product is that the two hormones can be adjusted independently — if someone needs more T4 but is doing fine on their current T3, that’s a change you can make.

Where professional bodies address this, they describe a deliberate, supervised, time-limited trial in a patient with confirmed hypothyroidism whose symptoms persist on adequate levothyroxine and in whom other explanations have been excluded — TSH kept in range, reassessment at three to six months, return to levothyroxine alone if there’s no improvement. That’s close to how we’d want a strategy evaluated with a prescribing provider.

What it is not: TSH is normal, the patient is tired, therefore add T3. That sequence skips the entire evaluation.

Is T3 Alone a Good Treatment?

Generally no — and the reason is how the hormone behaves, not anything sinister about it. Because T3 peaks fast and clears within about a day, taking it once daily gives you a spike and then a dip rather than the steady level T4 provides. That means splitting doses, monitoring more closely, and a real risk of palpitations, tremor, anxiety and insomnia if levels run high. TSH also becomes a less reliable guide when there’s no T4 in the regimen, which takes away one of the main safety checks. Guidelines reserve T3 on its own for people who genuinely can’t tolerate levothyroxine.

T3 isn’t a bad hormone. It’s an essential one that happens to be a harder tool to handle.

Women, Men and Thyroid Hormone Needs

Two separate things get conflated here constantly, so we’ll keep them apart deliberately.

Estrogen and Thyroid-Binding Proteins

This part is established physiology. Most thyroid hormone in your blood is bound to carrier proteins, and only the unbound fraction reaches tissues. Oral estrogen raises one of those carriers, thyroxine-binding globulin.

What happens next depends on whether you have a working thyroid. In a study following women on oral estrogen for 48 weeks, those with normal thyroid function saw binding globulin and total T4 rise while free T4 and TSH didn’t move — their glands simply made a bit more hormone. Women on a fixed dose of thyroid medication had no gland to compensate: free T4 fell and TSH climbed from an average of 0.9 to 3.2 (Arafah, NEJM, 2001). Same physiology, two outcomes, determined by whether the thyroid can respond.

Route matters. Oral estrogen passes through the liver first, and that first pass drives the effect; transdermal estrogen largely bypasses it and isn’t expected to do the same (Mazer, Thyroid, 2004).

This also causes unnecessary alarm on labs: total T4 and total T3 rise on oral estrogen, oral contraceptives, and in pregnancy with no change in actual thyroid status. A high total T3 in a woman on the pill is usually a binding-protein finding, not hyperthyroidism.

Practically: starting or stopping oral estrogen is a good reason to recheck thyroid labs in the weeks that follow. It’s a common and easily missed reason a dose that used to be right stops being right.

A Separate Question: Women and T3

Everything above is a binding-protein effect. It says nothing about how well someone converts T4 to T3, and it is not the reason some women do better with T3 in their regimen. The claim that “estrogen blocks thyroid hormone, so women need T3” is a misreading of the physiology, and a common one.

With that separation clear: in our practice, we see that women are more likely than men to ultimately do better with at least some T3 included in their replacement.

That’s a pattern we see, not a rule. No trial has shown a sex difference in who responds to combination therapy, and being a woman would never on its own be a reason to add T3. It’s one of the things we weigh when we’re trying to work out why a particular patient isn’t responding the way we’d expect.

On firmer ground: women moving through changing hormonal environments — contraception, pregnancy, perimenopause, hormone therapy — have more variables acting on their thyroid over the years, and that alone justifies closer monitoring.

Many male patients do very well on T4-based therapy, particularly when they appear to convert efficiently and feel well; some still benefit from T3-containing treatment. We’d avoid the common claim that testosterone automatically drives faster conversion — that’s more confident than the evidence supports.

Desiccated Thyroid: What It Is and Where It Fits

Desiccated thyroid extract goes by several names — DTE, NDT, natural desiccated thyroid, glandular thyroid — all describing the same category. The products most commonly prescribed in the U.S. are Armour Thyroid and NP Thyroid, both made from porcine thyroid tissue and containing T4 and T3.

They contain the same hormone per grain — 38 mcg of T4 and 9 mcg of T3, per both FDA labels — but differ in inactive ingredients and available strengths, which is a real reason someone might tolerate one and not the other. (Claims circulate online about these products being free of particular allergens; those claims don’t appear on the FDA labels, and we’d suggest confirming with the manufacturer rather than relying on them — particularly for anyone with celiac disease or a food allergy.)

Why It Works Well for Some Patients

We don’t view desiccated thyroid as a fallback for people who’ve run out of options. We work with these medications regularly and see people do well on them.

The likely reason isn’t mysterious: they already contain some T3. For a patient who does better with T3 in the picture, a product that includes it may simply produce a better response than T4 alone — the same phenomenon discussed above, in a different package. Both hormones come in one medication, dosing is straightforward, and some patients report feeling substantially better, which is worth taking seriously as clinical information.

The mainstream position is also more nuanced than the internet suggests. In September 2025 the American Thyroid Association — which does not recommend desiccated thyroid as first-line therapy — stated that “while it is not recommended by the ATA as first-line therapy, the ATA recognizes that it is the treatment of choice for some patients,” and that “the ATA supports the principle of personalized patient therapy.”

Is Desiccated Thyroid More “Natural”?

Here’s where the conversation usually turns ideological, so let’s replace the ideology with a number.

Desiccated thyroid contains roughly four parts T4 to one part T3. The human thyroid secretes something on the order of eleven to fourteen parts T4 for each part of T3. Proportionally, desiccated thyroid delivers considerably more T3 than a human thyroid does — and it comes from a pig, whose thyroid isn’t calibrated to human physiology. Whatever else it is, it is not a replica of human thyroid secretion.

That’s not an argument these products are bad. It explains things patients notice: why some people feel a distinct effect on desiccated thyroid, why T3 rises noticeably in the hours after a dose, and why TSH often runs low on a dose that feels subjectively good. Those are predictable consequences of the ratio — which is also the product’s main structural limitation. The ratio is fixed, so you can’t increase one hormone without increasing the other, and some patients find the T3 content is simply more than suits them.

Meanwhile, synthetic medications contain hormone molecules corresponding to human T4 and T3. “Synthetic” describes how the molecule was manufactured, not whether your receptors recognize it. They do.

On standardization, the claim that desiccated thyroid is wildly inconsistent describes an older era, when these products were standardized by iodine content rather than hormone content. That changed in 1985, and modern products are assayed for actual T4 and T3. The current concern is narrower: it’s a biological product, and there have been documented potency problems — FDA recalls in 2020 and 2021 for products testing above or below labeled content, and another for superpotency in August 2026. A real quality issue worth knowing about, but not the same as the products being unstandardized.

Is Desiccated Thyroid Better Than Synthroid?

Not universally, and the question is framed in a way that can’t be answered. Some patients clearly feel better on desiccated thyroid; others feel better on Synthroid, on Tirosint, on separately dosed T4 and T3 — and feeling better on synthetic therapy is a legitimate outcome, not a failure to appreciate the product. Both are reporting accurately about themselves; neither has discovered a general truth.

The table below shows where each approach tends to fit, and pregnancy is the one situation where the choice genuinely narrows.

Is Armour Thyroid Being Taken Off the Market?

No. Armour Thyroid and NP Thyroid remain available, and no patient has been told to stop taking them. Something real did change, though, and the headlines have been more alarming than the situation.

These products have never been FDA-approved. That isn’t new — it’s a legacy of how they’ve been marketed for decades, and it’s stated on the label: “This drug has not been found by FDA to be safe and effective.” If you take one, that sentence is already in your package insert.

In August 2025, the FDA notified manufacturers that animal-derived thyroid products fall under the pathway for biological products rather than the ordinary drug-approval pathway, because they contain a protein from the source tissue — which would require a different and more involved application than anyone had been pursuing. Roughly 1.5 million patients filled prescriptions for these products in 2024.

In March 2026, the FDA revised its position substantially, removing language about transitioning patients to alternatives and moving to an open-ended, risk-based enforcement approach. No deadline was set, and the agency has not required these products off the market.

As of September 2026, the FDA’s promised guidance still had not been issued, there’s no enforcement deadline, and the FDA’s advice to patients is to talk with their doctor — not to stop. At least one manufacturer has completed a clinical study working toward what would be the first approved desiccated thyroid product. The reclassification itself is disputed, and the situation is evolving, so anything written about it carries a date. This section reflects the FDA’s position as of September 2026.

Why Your Pharmacy Experience May Not Match the Headlines

This is the most practically useful part of this section, and almost nobody explains it. Three separate things affect whether you can get your medication.

Regulatory status is whether the FDA permits a product to be marketed. That’s what the news coverage is about, and these products remain on the market.

Insurance formulary decisions are whether your plan or pharmacy benefit manager covers a medication — business decisions made independently of the FDA, and they change regularly. One major pharmacy benefit manager removed desiccated thyroid from its standard formulary in April 2026, with coverage potentially available through prior authorization. A patient whose medication suddenly requires prior authorization or costs far more is experiencing a coverage decision, not a withdrawal.

Pharmacy stocking decisions are whether an individual pharmacy keeps something on the shelf. A product can be entirely available and still need ordering.

Any of these can make a medication feel like it’s disappearing. Usually it isn’t — and it’s worth finding out which you’re dealing with, because the solutions are completely different: an appeal or prior authorization, a different pharmacy, or a real conversation about alternatives. None of it is a reason to stop or switch on your own. Stopping thyroid hormone abruptly creates a more immediate problem than anything currently happening at the FDA.

How We Decide Whether a Strategy Is Working

We look at three things together, and no one of them is sufficient alone.

The patient feels and functions well — energy, cognition, bowel function, temperature tolerance, sleep, hair, menstrual health, capacity to do what they want to do. A lab value that leaves someone unable to function isn’t a success. It can’t be the only measure either, for the reason above: these symptoms aren’t specific to thyroid disease.

The laboratory pattern looks appropriate — TSH, Free T4, Total T4, Free T3, Total T3 where relevant, and other markers depending on context, interpreted together and with attention to what medication someone is on and when they took it. No single thyroid marker tells the entire story.

When Hashimoto’s is present, the autoimmune picture is not worsening — TPO and thyroglobulin antibodies, the clinical picture, and imaging when indicated.

All three, not one of the three.

Why TSH Does Not Always Tell the Whole Story

TSH isn’t a thyroid hormone at all. It’s a pituitary hormone — the signal the pituitary sends to the thyroid, turned up when it senses too little thyroid hormone and turned down when it senses enough. That makes it an indirect measurement, and a very good one for the job it was designed to do. The 2014 American Thyroid Association guidelines call TSH “the most reliable marker of adequacy of replacement treatment,” and as a general statement we don’t disagree.

What TSH can’t tell you directly is how much thyroid hormone is circulating, in what form, how much is getting into tissues, or how much thyroid signaling is actually happening once it arrives. Those are separate questions, and there’s now a reasonable body of human evidence that they don’t always have the same answer.

Why the Pituitary and the Rest of the Body May Not Read Thyroid Hormone the Same Way

Here’s the piece of physiology that explains most of this.

T4 has to be converted into T3 before it does much, and a lot of that conversion happens inside cells rather than in the bloodstream. The enzyme responsible is type 2 deiodinase, usually shortened to D2. Nearly every tissue uses it — including the hypothalamus and pituitary, the very tissues deciding how much TSH to release.

The interesting part is what T4 does to that enzyme. In most tissues, rising T4 actually reduces D2 activity: T4 tags the enzyme for destruction, so as more T4 arrives, each cell converts a progressively smaller share of it. That’s a sensible brake for keeping local T3 steady.

In the hypothalamus, at least in animal studies, that brake appears to work differently — D2 there is relatively protected from being shut down by T4 (Werneck de Castro et al., Journal of Clinical Investigation, 2015). If the same holds in people, the consequence is uncomfortable: when you give levothyroxine, the pituitary may be able to make plenty of T3 for itself, and quiet down, before the rest of the body has caught up.

This isn’t a fringe idea. The 2021 joint consensus document from the American, British and European thyroid associations states it directly: “D2-mediated T3 production is a more efficient process in the hypothalamus/pituitary unit when compared with other D2-containing tissues,” with the result that “TSH secretion is normalized before T3 levels are fully restored in the plasma and other tissues.”

Two honest caveats. The detailed mechanism has been worked out in rodents, not in human brain tissue — nobody has biopsied a human hypothalamus to confirm it. And rats get a larger share of their T3 directly from the thyroid than humans do, so the translation isn’t automatic. Treat this as a well-supported explanation rather than a proven one in people.

What has been measured in people is the consequence.

What Actually Happens in People Taking Levothyroxine

Several human studies, using different designs, point the same direction.

In the largest, 1,811 patients with no functioning thyroid taking levothyroxine were compared with nearly 3,900 people with healthy thyroids, matched for the same TSH. The treated patients had higher Free T4, lower Free T3, and a markedly lower Free T3-to-Free T4 ratio. About 15% had a Free T3 below the reference range entirely, with a TSH that looked perfectly fine (Gullo et al., PLOS ONE, 2011). A U.S. population study using the same matched-on-TSH approach found the same pattern (Peterson et al., JCEM, 2016).

The cleanest demonstration compares people to themselves. In 135 patients studied before and after complete thyroid removal, those whose post-operative TSH landed in the normal range had a Free T3 significantly lower than their own pre-surgery level. Getting their original T3 back required a dose that pushed TSH somewhat below the reference range (Ito et al., European Journal of Endocrinology, 2012).

There are hints this matters functionally, though the evidence thins out here. Resting energy expenditure in levothyroxine-treated women with normal TSH ran about 4–6% lower than in women with healthy thyroids, and tracked with their Free T3 (Samuels et al., Thyroid, 2016). A review pooling 99 studies found modestly higher total and LDL cholesterol in treated patients at normal TSH (McAninch et al., JCEM, 2018). These are small differences in surrogate markers, not evidence that anyone feels worse, and we won’t present them as more than that.

So a Normal TSH Doesn’t Settle the Question

This is the practical conclusion, and it runs against something a lot of patients have been told is beyond discussion.

Someone can have a normal TSH and still have relatively low circulating thyroid hormone, a gland that is progressively failing, active autoimmune thyroid disease, structural abnormalities visible on ultrasound, or a clinical picture that supports treatment. A single number doesn’t rule those out.

So when the question is whether replacement is warranted, or whether someone is adequately replaced, what we look at includes Free T4 and Free T3; Total T4 and Total T3 where they add something; thyroid antibodies; binding proteins when there’s reason to think they’re distorting the picture; symptoms and clinical history; which medication someone takes, at what dose, and when they last took it; pregnancy status or plans; and thyroid ultrasound or other structural findings when they’re indicated.

That isn’t testing for the sake of testing. It’s recognizing that TSH answers one question well and several other questions not at all.

Reading TSH When T3 Is Part of the Regimen

There’s a specific wrinkle worth understanding if you take liothyronine or desiccated thyroid.

Immediate-release T3 produces a rise and fall in the blood rather than the steady supply a working thyroid provides — and the pituitary’s response doesn’t track that pulse in real time. In a study that gave healthy volunteers a single dose of liothyronine and then followed them for four days, serum T3 peaked at about 2.5 hours. TSH didn’t reach its lowest point until 12 hours, and stayed below baseline for two to three days (Jonklaas et al., Therapeutic Drug Monitoring, 2015).

That gap matters for interpretation. A TSH drawn on a T3-containing regimen reflects an accumulated, delayed signal from the preceding days — not the amount of T3 in that tube at that moment. It’s one reason TSH often looks disproportionately low in people taking T3, a pattern we see regularly in our practice, and one reason the 2021 consensus recommends standardizing when blood is drawn and measuring T3 at a trough.

We want to be equally clear about the other direction, because this is exactly where someone could talk themselves into trouble. A suppressed TSH is not automatically benign, and it is not something to wave off. Genuine over-replacement carries real cardiovascular and bone consequences — covered in the next section, with the numbers. The point isn’t that a low TSH can be ignored. It’s that it has to be read alongside T3 and T4 levels, the medication and dose, the timing of the last dose relative to the draw, symptoms, heart rate and cardiovascular history, bone-health risk, and how the person is doing overall.

Same TSH, Different Physiology

One more study brings this together.

In a double-blind crossover trial, 14 hypothyroid patients took liothyronine and levothyroxine in turn, with doses adjusted so that TSH was the same on both — the researchers reported no difference in TSH between the two treatments. Even so, patients weighed about two kilograms less on liothyronine, with total cholesterol roughly 11% lower, LDL about 13% lower, and apolipoprotein B about 18% lower. Heart rate, blood pressure, insulin sensitivity and exercise tolerance didn’t differ (Celi et al., JCEM, 2011).

Fourteen people, six weeks per arm, surrogate endpoints, and a three-times-daily dosing schedule nobody actually uses — so this demonstrates a principle rather than recommending a treatment. But the principle is what matters here: the same TSH does not necessarily mean the same thyroid hormone physiology throughout the body.

None of this makes TSH useless, and we’d be unhappy if that were the takeaway. It remains the best single screening test for thyroid disease and an essential part of monitoring anyone on replacement. What it isn’t is the entire assessment. Pituitary euthyroidism and whole-body euthyroidism are related ideas — they are not automatically the same one.

Can You Take Too Much — or Too Little?

Yes to both, and this section matters as much as any other.

Signs of excess include palpitations, tremor, anxiety, heat intolerance, sweating, insomnia, loose stools and unintended weight loss. Some overlap with the symptoms people are trying to treat, which is part of why “increase the dose until she feels better” is a poor strategy.

Precision matters here, and writing on both sides tends to get it wrong. The best data comes from nearly 18,000 patients on long-term levothyroxine. Compared with those whose TSH was in range, patients whose TSH was suppressed — below 0.03 — had higher rates of cardiovascular disease, arrhythmias and fractures. Patients whose TSH was low but not suppressed, between 0.04 and 0.4, showed no significant increase in any of these (Flynn et al., JCEM, 2010). A meta-analysis of thirteen cohorts found the same concentration of fracture risk below a TSH of 0.10 (Blum et al., JAMA, 2015), and a long-running study found a threefold atrial fibrillation risk at 0.1 or below, with no significant increase in the group whose TSH was low but above it (Sawin et al., NEJM, 1994).

A modestly low TSH and a fully suppressed TSH are not the same thing. Treating them as equivalent produces unnecessary anxiety in people who are fine and unnecessary dose reductions in people who were doing well.

The other half is less discussed and equally important. In that same study of 18,000 patients, an elevated TSH — under-replacement — carried risk at least as high as suppression, and higher for cardiovascular disease. Being under-treated is not the safe option; it’s a different way to be inadequately treated.

Regimens containing T3 need extra attention, because levels spike after each dose and because TSH is a less straightforward guide when T3 is in the mix. That’s an argument for watching more closely, not for avoiding it.

Hashimoto’s and Hypothyroidism Are Not the Same Problem

These terms get used interchangeably and describe different things. Hashimoto’s is an autoimmune process — the immune system directing activity against thyroid tissue. Hypothyroidism is a functional state — not enough thyroid hormone available. Hashimoto’s causes hypothyroidism in many people over time, but someone can have Hashimoto’s for years before hormone production becomes inadequate, and that earlier window is where a lot of meaningful clinical work happens.

Once thyroid tissue has been sufficiently damaged, the situation changes. Nutrition, gut treatment, targeted supplementation and lifestyle change are all worth doing — and they don’t restore destroyed thyroid tissue. A gland that can no longer produce enough hormone won’t be argued back into producing enough. We do not view needing thyroid hormone as a failure to find the “root cause.” Replacement and broader functional care aren’t competing approaches.

On antibodies: TPO and thyroglobulin antibodies are part of the autoimmune picture and we watch them. They are not a dosing target — they don’t need to reach zero, individual fluctuations don’t represent treatment failure, and thyroid hormone shouldn’t be adjusted from an antibody level.

Positive antibodies alone also don’t create an indication for thyroid hormone. For women with normal thyroid function who are antibody-positive and dealing with infertility or recurrent pregnancy loss, high-quality randomized trials have shown that levothyroxine does not improve fertility or pregnancy outcomes, and the 2026 American Thyroid Association pregnancy guidelines advise against offering it in that situation, recommending monitoring instead. Antibodies are worth knowing about and following. Treating them with thyroid hormone, absent actual thyroid dysfunction, doesn’t appear to help.

Why Gut Health Still Matters

The relationship runs both directions: hypothyroidism slows gastrointestinal motility, contributing to constipation, and GI dysfunction affects both medication absorption and the nutrients thyroid physiology depends on. It isn’t thyroid medication or gut health — sometimes both need attention. What treating the gut won’t do is remove the need for replacement in someone whose thyroid can no longer make enough hormone. More in our article on thyroid and gut health.

Thyroid Medication When Trying to Become Pregnant

Pregnancy is the one situation where the usual individualized framing genuinely narrows, and understanding why makes the recommendation make sense rather than feel arbitrary.

The fetus doesn’t produce its own thyroid hormone in early pregnancy — the fetal thyroid doesn’t begin meaningful production until around the twelfth week. Through the first trimester, a period of intensive brain development, it depends entirely on the mother.

Here’s the part that drives medication choice: what reaches the fetus is primarily T4. According to the 2026 American Thyroid Association guidelines, the fetal central nervous system is relatively impermeable to T3, and most of the T3 in the fetal brain during pregnancy is produced locally from maternal T4 rather than crossing over directly. The placenta actively breaks down T3. So the developing brain takes in maternal T4 and manufactures its own T3 locally — which makes maternal T4 availability the input that matters.

The 2026 guidelines recommend treating hypothyroidism in pregnancy with levothyroxine and advise against using other thyroid preparations such as LT3 (liothyronine) or desiccated thyroid during pregnancy. They extend this to before conception, advising that women planning pregnancy who take liothyronine or desiccated thyroid switch to levothyroxine before beginning fertility treatment. Obstetric guidance agrees: the American College of Obstetricians and Gynecologists advises avoiding these preparations in pregnancy, because their high T3-to-T4 ratio can push maternal T3 to supraphysiologic levels while leaving maternal T4 low.

The clearest way to express the concern: a T3-heavy regimen can produce a normal-looking TSH while maternal T4 — the form the fetal brain relies on — runs relatively low. TSH does not warn you about this.

There are no trials directly comparing these medications in pregnancy, so this recommendation rests on the physiology and on strong agreement among experts about what that physiology means. The ATA calls its own recommendation strong while noting the underlying evidence is low quality — which isn’t a contradiction. When the mechanism is well understood and the stakes involve a developing brain, waiting for a trial isn’t the right call. More on the wider picture in the effects of thyroid disease on menstruation, fertility, and pregnancy.

If You Used a Different Medication in a Past Pregnancy

If you took desiccated thyroid or a T3-containing medication during a previous pregnancy, this guidance is not a judgment about the care you received, and it is not a reason to worry retrospectively about your child.

Recommendations here have developed as evidence and expert consensus have developed, and previous decisions were made in the context of the guidance and clinical circumstances available at the time. Guidance evolving is how medicine is supposed to work. Many women have had healthy pregnancies on a range of thyroid regimens. This is information for planning forward, not for looking backward.

Our Approach

When pregnancy is being actively pursued, our approach generally leans toward a T4-based strategy and closer monitoring. If someone is on a T3-containing or glandular product and is planning pregnancy or preparing for fertility treatment, that’s a conversation we want to have with the prescribing provider early — well before conception rather than after a positive test.

Pregnancy also changes thyroid physiology substantially. Requirements commonly increase, sometimes early and quickly, and monitoring intensifies accordingly — the 2026 guidelines suggest checking TSH at pregnancy confirmation and roughly every four weeks through the first half of pregnancy, using the same laboratory assay throughout. Ovarian stimulation raises estrogen sharply and can itself push TSH up, another reason to have thyroid status settled before fertility treatment begins. Treatment shouldn’t be adjusted casually during pregnancy, and it should be coordinated between the prescribing provider and the obstetric team.

So Which Thyroid Medication Is Best?

The table isn’t a ranking. It describes different tools and where each tends to fit.

Our Approach to Thyroid Medication at Genesis Functional Wellness

Here’s where we stand, including what we don’t believe:

We do not believe everybody should take T4. We do not believe everybody needs T3. We do not believe glandular thyroid is inherently better because it is “natural,” and we do not believe synthetic medication is inherently better simply because it is standardized.

Thyroid medications are tools, and the appropriate tool depends on the patient.

What we evaluate: symptoms and how someone is actually functioning; a complete thyroid panel rather than TSH alone; response to previous treatment; reproductive stage and pregnancy goals; absorption and gastrointestinal considerations; formulation history, including changes they may not have been told about; and the autoimmune picture when Hashimoto’s is present.

Medication decisions are coordinated with the prescribing medical provider. Our role is to evaluate thyroid physiology thoroughly, work out what’s actually driving a patient’s clinical picture, and bring specific, well-reasoned recommendations into that conversation — including recommending a different formulation or strategy when the current one isn’t producing the response we’d expect. For many patients this turns out to be the practical advantage: they keep the physician they already have, and add an evaluation that goes further than a TSH. As our medical team grows, more of this management will happen directly within the practice.

Sometimes a strategy that looks appropriate on paper turns out not to be the right tool for that patient. Starting with one medication and later changing it doesn’t mean the first decision was a mistake — it means treatment is being adjusted on new information. Thyroid replacement is a clinical process, not a personality type. Nobody is a “Synthroid person” or an “Armour person.”

The Bottom Line

Thyroid medication doesn’t need to be an argument between synthetic and natural, T4 and T3, or conventional and functional medicine. The people who insist otherwise, in either direction, are usually arguing about something other than your physiology.

What matters is understanding what hormone a medication provides, how a particular person processes it, how they respond clinically, what the labs show, whether something else is interfering, and whether the medication still fits as physiology changes.

You Don’t Have to Change Doctors to Get a Closer Look

One of the most common reasons people put off having their thyroid properly evaluated is the assumption that it means starting over somewhere new — leaving a physician they like, or explaining a decade of history to someone who wasn’t there for any of it.

It doesn’t have to work that way. Precision Thyroid Care is built around the evaluation: comprehensive thyroid testing, ultrasound when it’s indicated, a careful read of how you’re actually responding to the treatment you’re on, and ongoing monitoring as things change. When medication needs to be reconsidered, we work with the clinician who writes your prescription rather than around them.

Explore Precision Thyroid Care →

Have multiple overlapping health concerns? Learn about Comprehensive Functional Wellness →