The Autoimmune Protocol (AIP) in Hashimoto's Thyroiditis: What Two Studies Actually Showed

In Brief

The autoimmune protocol is the most radical of the popular diets in Hashimoto's. Where a gluten-free diet removes one protein, this one removes all grains, legumes, dairy, eggs, nuts and seeds, nightshades, alcohol, coffee, sweeteners and a range of additives. The promise is correspondingly large: halt the autoimmune process with food. What stays on the plate is meat, fish, vegetables other than nightshades, fruit and fats, and the protocol itself is built in phases: elimination is meant to be followed by bringing foods back. In the two existing studies the elimination phase ran 10 and 12 weeks — every figure discussed below comes from those durations.

The idea the protocol rests on is increased intestinal permeability: remove the irritating foods, the reasoning goes, and antigen traffic falls, taking the autoimmune response down with it. That mechanism is covered separately, in the piece on intestinal permeability. This article is about something else: whether the diet built on it works.

That promise can only be tested against data, and data exist. There is very little of it, its design is weak, and it does not say what the promise says.

How Much Evidence There Actually Is

There are two interventional studies of the autoimmune protocol in Hashimoto's thyroiditis. Not «few», not «insufficient» — literally two.

• Abbott and colleagues, 2019: 17 women aged 20–45 recruited, 16 completed, 10 weeks (Abbott RD, Cureus 2019).
• Ihnatowicz and colleagues, 2023: 28 people including one man, 12 weeks (Ihnatowicz P, Ann Agric Environ Med 2023).

Both are single-arm: measurements before and after, with no one to compare against. No randomisation, no blinding, no group that carried on eating normally.

No randomised controlled trial of the autoimmune protocol exists. Not in thyroiditis, not in inflammatory bowel disease, not in rheumatoid arthritis — not in any condition. There are no meta-analyses of the protocol either.

The entire evidence base for the protocol consists of four interventional studies totalling 68 participants: two in thyroiditis, one in inflammatory bowel disease with 15 patients, and one in rheumatoid arthritis with 9 (Pardali EC, Metabol Open 2025). For scale: that headcount would not fill a single arm of an average randomised trial.

What Changed — and What Did Not

The first study measured three things at once: well-being, thyroid function and inflammation. The results diverged.

The table reads as follows. People felt markedly better — both in quality of life and in symptom count, and the effect size there is large. Meanwhile the thyroid did not move on a single measure. Thyroglobulin antibodies even rose on average, although the spread in that group was twice the mean itself, so the direction of the shift cannot be interpreted.

Of the pre-specified measures, the only one that changed significantly was the systemic inflammation marker hs-CRP: down 29%, from 1.63 to 1.15 mg/L (p = 0.022). The effect size was small, and the design does not permit «the protocol reduced inflammation» — only «it fell during the intervention». The authors also mention shifts in the white cell differential in their conclusions, but those hold up less well: the white cell count did not change significantly in the main analysis, and the rise in lymphocyte share did not survive correction for multiple comparisons, as the authors themselves state.

The Second Study: Antibodies Went Up

The second and larger study produced a result that usually gets lost in retellings. Over 12 weeks TSH and free T3 and T4 fell while staying within the reference range, thyroid volume decreased, thyroglobulin antibodies decreased slightly — and thyroid peroxidase antibodies rose significantly (Ihnatowicz P, Ann Agric Environ Med 2023).

The same authors note that participants' body weight fell and attribute it to a caloric deficit. That detail matters: some of the observed shifts may follow from weight loss rather than from the composition of the diet.

The summary across both studies is simple. In one, antibodies did not change; in the other, they rose. A study showing them fall does not exist.

Why the «Diet» Here Is Not Only a Diet

Even the improvement in well-being cannot be attributed to food removal alone, and that is visible in how the intervention is described. In the first study participants received not a list of prohibitions but a ten-week supported programme:

• a team of a physician, nutritional therapy practitioners and certified health coaches;
• education and support by email and in a private group;
• a separate lifestyle component: sleep, stress management, movement, time outdoors, support systems;
• in week five, individual dietary recommendations built from each participant's own test results.

Separating the contribution of food, coaching, group support and physician attention is impossible in this design even in principle. The paper's own title reflects that honestly: it studied the protocol as part of a multi-disciplinary intervention, not the diet by itself.

One more thing worth knowing, disclosed in the publication itself: a co-author is a co-founder of the company whose coaching programme was the intervention under study.

What These Studies Do Not Contain

• No control group in either study.
• No follow-up longer than 12 weeks.
• No hard outcomes: levothyroxine dose, progression to overt hypothyroidism, ultrasound change as a primary endpoint — none of these were measured.
• Almost no men: the first study was women-only, the second had one man among 28.
• No data on what happens after the programme ends.

What the Guidelines Say

Nothing. That is a result, not an empty cell.

The autoimmune protocol appears in no American or European Thyroid Association guideline — neither as a recommended option nor as one considered and rejected. In the most recent European Thyroid Association guideline on levothyroxine the phrases «autoimmune protocol», «elimination diet» and «Paleo» do not occur once; diet is discussed there solely as a factor in drug absorption.

Absence from guidelines does not mean «proven not to work». It means the evidence base was not even sufficient for the protocol to enter the discussion. Authors sympathetic to the approach acknowledge this too: a systematic review of the Paleolithic diet in autoimmune thyroid disease found eight studies in total, six of which were case reports (Hollywood JB, J Am Nutr Assoc 2023).

Risks and Costs

Strict elimination is not free, and it is worth knowing the price in advance.

In fairness, the opposite finding deserves equal space. In a large French cohort of 20,456 people without celiac disease, those who restricted gluten ate more fruit and vegetables and less sugar, salt and alcohol on average — their diet was closer to a healthy pattern, not worse (Perrin L, Br J Nutr 2019). Elimination does not impoverish a diet automatically; it impoverishes it when what was removed is not replaced.

The Reintroduction Phase

The autoimmune protocol is built from three phases: elimination, reintroduction, maintenance. Bringing foods back is not an option or a sign of weakness — it is a constitutive part of the design.

In practice this is the phase most often left unfinished, and here it has to be said plainly: there are no numbers on it. No data exist on what share of people outside a study protocol reach the reintroduction stage. There are no long-term data on the maintenance phase either.

What is known is the danger of getting stuck in phase one. The review authors put it bluntly: the protocol is not a dietary pattern that can be implemented for life, and remaining in the elimination phase puts a person at risk of nutrient deficiencies and malnutrition (Pardali EC, Metabol Open 2025).

What to Test Before Starting

• Tissue transglutaminase antibodies and total IgA — before removing gluten. On a gluten-free diet celiac disease stops being diagnosable, and the person risks going without an answer for a long time. In autoimmune thyroid disease celiac disease is biopsy-confirmed in about 1.6% — pooled across 6,024 patients — and more often in children with thyroiditis, at 6.2% (Roy A, Thyroid 2016).
• TSH, free T4, TPO antibodies — the baseline without which later changes have nothing to be compared to.
• Vitamin D, ferritin, vitamin B12 — the things elimination can worsen and that are worth knowing in advance.
• Weight — to separate the effect of dietary composition from the effect of a caloric deficit.

What Is Actually Supported in Thyroiditis

A guide that leaves the reader with nothing but prohibitions is useless. So here is what is known about supplements.

There is a second side to it. A separate meta-analysis found no clinically meaningful benefit from selenium — not in well-being, not in TSH among untreated patients, not on ultrasound (Winther KH, Endocrine 2017). A Cochrane review of four early trials in 463 participants states the central point: not one of them tested whether the levothyroxine dose or quality of life changes (van Zuuren EJ, Cochrane Database Syst Rev 2013). And selenium carries a harm signal beyond the thyroid: in a randomised trial, taking 200 µg daily over an average of 7.7 years was accompanied by a higher incidence of type 2 diabetes (Stranges S, Ann Intern Med 2007).

Who It Makes Sense For — and Who It Does Not

The Bottom Line

The autoimmune protocol works — but not where it promises to. The two quality-of-life measurements deserve as much trust as the absence of a control group allows: people felt better. In the first study the thyroid felt nothing: neither TSH, nor free hormones, nor antibodies moved. In the second there were shifts — TSH and free hormones fell while staying within range, and thyroid volume decreased — but TPO antibodies rose significantly in that same study and body weight fell, so part of the change may follow from a caloric deficit rather than from what was eaten.

This is not a verdict against the diet or a reason to forbid it. It is a reason to name the price and the goal honestly. If the goal is to feel better, and the person is ready for a time-limited, supported programme that ends with foods coming back, that decision has a weak but real basis. If the goal is to halt the autoimmune process, there is no basis — and the price is months of restriction, a risk of deficiencies and a celiac diagnosis deferred.

The difference between those two goals is the entire point of this guide.