Do Vaccines Cause Autoimmune Diseases?

Large population studies consistently show that vaccines do not cause most autoimmune diseases. Across millions of tracked individuals, the rates of conditions like rheumatoid arthritis, multiple sclerosis, type 1 diabetes, and inflammatory bowel disease stay the same whether people are vaccinated or not. That said, the honest answer is not a clean “never.” A handful of genuine vaccine-autoimmune links have been identified over decades of surveillance, and they reveal something interesting about the immune system: in very specific circumstances, involving particular vaccines and particular people, an immune response meant to protect can occasionally misfire. Understanding where those rare exceptions sit, and why the broader pattern is one of safety, matters more than a simple yes or no.

What the Large Studies Show

The most useful evidence comes from studies that track large populations over time and compare autoimmune disease rates in vaccinated and unvaccinated groups. A population-based cohort in Hong Kong covering 1.1 million vaccinated people found that within 28 days of receiving either an mRNA or inactivated COVID-19 vaccine, the rates of autoimmune conditions requiring hospital care were below 9 per 100,000 and were not significantly higher than in unvaccinated individuals.¹ A Swedish study following recipients of the Pandemrix H1N1 flu vaccine for up to ten months found no change in risk for Guillain-Barré syndrome, multiple sclerosis, type 1 diabetes, or rheumatoid arthritis.²

A nationwide Korean cohort study looking at longer-term outcomes after mRNA COVID-19 vaccination found no increased risk for the vast majority of autoimmune connective tissue diseases, including psoriasis, rheumatoid arthritis, Crohn’s disease, ulcerative colitis, vitiligo, and more than a dozen others. The one exception was a modestly elevated risk of systemic lupus erythematosus, though even that signal was small.³ For vaccines that have been around much longer, a systematic review found no change in multiple sclerosis risk after vaccination against hepatitis B, HPV, seasonal flu, measles-mumps-rubella, tetanus, polio, or several other common immunizations.⁴ A separate case-control study reached a similar conclusion, finding no association between hepatitis B vaccination, HPV vaccination, or vaccination in general and the risk of central nervous system demyelinating diseases up to three years later.⁵

The pattern across these studies is remarkably consistent: for the overwhelming majority of autoimmune conditions, vaccination does not move the needle.

The Documented Exceptions

Against that reassuring backdrop, a few specific vaccine-autoimmune links have been confirmed through rigorous investigation. These are worth understanding individually, because they reveal how rare and how specific such events are.

MMR and Immune Thrombocytopenia

The measles-mumps-rubella vaccine can trigger immune thrombocytopenic purpura (ITP), a condition in which the immune system attacks platelets and causes easy bruising or bleeding. A large English study of children born between 2000 and 2019 found the risk was elevated after the first dose of MMR, with roughly one extra case for every 43,000 doses given. The second dose showed no increased risk at all.⁶ An earlier study had estimated the attributable risk at about one case per 25,000 vaccinations.⁷ In most children, MMR-associated ITP resolves on its own within weeks. It has remained an accepted risk in the vaccine’s safety profile since the 1990s, and it is one of the clearest examples of a vaccine triggering a genuine, if usually self-limiting, autoimmune response.

Pandemrix and Narcolepsy

The most consequential confirmed link between a vaccine and autoimmunity involves Pandemrix, an adjuvanted H1N1 influenza vaccine used during the 2009 pandemic. Following the vaccination campaign, the risk of narcolepsy type 1 rose five- to fourteen-fold in children and adolescents and two- to seven-fold in adults, with cases characterized by a loss of hypocretin-producing neurons in the brain.⁸ The elevated risk persisted for about two years after vaccination. Pandemrix stood out sharply from other influenza vaccines in post-marketing surveillance data.⁹

Researchers concluded that the adjuvant used in Pandemrix (called AS03) alone could not explain the narcolepsy cases. Instead, the evidence pointed to a combination of factors: genetic predisposition, the specific H1N1 antigens in that vaccine formulation triggering molecular mimicry against hypocretin neurons, and the adjuvant amplifying bystander immune activation.¹⁰ Pandemrix was withdrawn from use, and other H1N1 vaccines that used different adjuvants or formulations did not carry the same narcolepsy risk. The episode became a landmark case in vaccinology precisely because it was so unusual.

Adenoviral COVID-19 Vaccines and VITT

During the COVID-19 pandemic, the adenoviral vector vaccines (particularly the AstraZeneca/Oxford vaccine) were linked to a novel autoimmune clotting syndrome called vaccine-induced immune thrombotic thrombocytopenia, or VITT. In rare cases, the vaccine triggered autoantibodies against a protein called platelet factor 4, leading to severe blood clots combined with low platelet counts.¹¹ The mechanism has parallels to heparin-induced thrombocytopenia, another autoimmune clotting disorder, and researchers traced the chain of events to interactions between the adenoviral DNA cargo and platelet factor 4, which can break the immune system’s normal tolerance.¹² VITT was extremely rare but could be fatal, and its identification led to modified recommendations and eventually to the phasing out of adenoviral COVID-19 vaccines in most countries.

How a Vaccine Could Theoretically Trigger Autoimmunity

The immune system’s job is to recognize foreign invaders and mount a defense against them. Occasionally, parts of a vaccine resemble parts of the body’s own proteins closely enough that the immune response aimed at the vaccine accidentally targets self-tissue. This is called molecular mimicry, and it is the most studied mechanism behind vaccine-related autoimmune reactions.¹³

A second mechanism involves what immunologists call bystander activation. When a vaccine activates the immune system, immune cells in the surrounding tissue can get stirred up even if they are not specifically responding to the vaccine antigen. In animal experiments, this kind of nonspecific local activation of immune cells has been demonstrated following injection with certain adjuvants.¹⁴ In the case of mRNA COVID-19 vaccine-associated myocarditis, a recent study provided the most direct evidence yet for molecular mimicry, showing that immune cells from affected patients recognized vaccine-encoded spike protein sequences that resemble proteins found in heart muscle cells.¹⁵

These mechanisms explain how a vaccine could, in principle, cause autoimmunity. They do not explain why it almost never does. The reason is that the immune system has multiple layers of tolerance built in to prevent self-attack, and a vaccine-induced immune response is typically not strong enough or sustained enough to break through all of those safeguards. It takes a convergence of factors, including the right genetic background, the right antigen, and sometimes the right adjuvant, to produce the kind of perfect storm seen in the Pandemrix-narcolepsy cases.

The Role of Adjuvants

Adjuvants are substances added to some vaccines to boost the immune response. Aluminum salts have been the most commonly used adjuvant for decades. In 2011, researchers proposed a framework called ASIA (autoimmune/inflammatory syndrome induced by adjuvants) to organize case reports of autoimmune-type reactions that seemed linked not to the vaccine antigen itself but to the adjuvant.¹⁶ The ASIA concept groups together reactions from vaccine adjuvants, silicone implants, and other environmental immune stimulants under a single umbrella.¹⁷

ASIA remains controversial in mainstream immunology. Critics point out that the diagnostic criteria are broad enough to capture many conditions that would have occurred regardless of vaccination, and the framework has not produced the kind of strong epidemiological signal that would establish a clear causal link for most adjuvants. But the concept has been useful in one sense: it keeps the question alive and encourages investigators to examine adjuvant-specific effects rather than treating all vaccines as interchangeable. As the Pandemrix episode demonstrated, the adjuvant matters. The AS03 adjuvant in Pandemrix contributed to the narcolepsy signal in ways that other adjuvants did not.

Autoantibodies Are Not the Same as Autoimmune Disease

One finding that sometimes gets misinterpreted is that vaccines can trigger the formation of autoantibodies, proteins that react against the body’s own tissues. This sounds alarming, but healthy people produce transient autoantibodies all the time, especially after infections. A prospective one-year study tracked autoantibody development after two and three doses of the Pfizer-BioNTech mRNA COVID-19 vaccine. The vaccine did cause new autoantibodies to appear in a minority of recipients, but these were mostly transient, and after a full year of follow-up, no participant developed a new autoimmune disease, regardless of whether autoantibodies had formed.¹⁸ A separate analysis found that new-onset antinuclear antibody positivity occurred in only about 5% of participants after mRNA vaccination, and these antibodies had disappeared by three months.¹⁹

The gap between autoantibody formation and actual clinical disease is important. Autoantibodies are a lab finding. Autoimmune disease is a sustained, tissue-damaging process. The immune system has brakes, and in most people, a transient spike in autoantibodies gets shut down long before it causes symptoms.

If You Already Have an Autoimmune Disease

For people living with conditions like rheumatoid arthritis, lupus, or psoriasis, the question is more personal: will vaccination make my disease worse? The evidence here is more nuanced. A large registry study from Singapore found that roughly 18% of patients with autoimmune inflammatory rheumatic diseases experienced a flare after mRNA COVID-19 vaccination, with about 12% of those flares occurring within three months. However, most flares were mild to moderate, only 14% were severe, and hospitalization was rare. Patients with active disease at the time of vaccination and those on certain immunosuppressive medications had higher flare rates.²⁰ Interestingly, nearly 12% of patients in the same study actually improved after vaccination.

A broader adverse-event analysis across multiple autoimmune conditions found that aggravation of the existing condition was among the most commonly reported events following COVID-19 vaccination in these patients, spanning conditions from lupus to Hashimoto’s thyroiditis to ankylosing spondylitis.²¹ The key nuance is that flares are temporary disruptions, not a new disease. Professional guidelines universally continue to recommend vaccination for people with autoimmune conditions, particularly because the infections vaccines prevent can themselves trigger far worse flares or complications.

Why Genetics Shape Individual Risk

The fact that autoimmune reactions to vaccines are so rare but reproducible in certain individuals points strongly toward genetic susceptibility. Certain variants of HLA genes, which help the immune system distinguish self from non-self, have been linked to vaccine-associated autoimmune events. For COVID-19 vaccines specifically, certain HLA class I and class II alleles have been associated with vaccine-induced thyroiditis and with VITT.²² An older study of autoimmune encephalomyelitis triggered by the older Semple-type rabies vaccine (which used nerve tissue and is no longer used in most countries) found that specific HLA-DR alleles were more common in affected patients, and that these susceptibilities were distinct from those associated with multiple sclerosis, suggesting different genetic pathways can lead to different vaccine-associated outcomes.²³

This means that when a vaccine does trigger autoimmunity, it is not random bad luck applied uniformly. It is a genetically influenced susceptibility interacting with a specific immunological trigger. The hope in this line of research is that eventually genetic screening could identify individuals at elevated risk for particular vaccine reactions, allowing for alternative vaccine formulations or closer monitoring.

Infection Usually Poses a Greater Autoimmune Risk Than Vaccination

A question that often gets lost in the vaccine-autoimmunity debate is what happens when you catch the infection itself. For influenza, a simulation study concluded that under typical conditions, getting vaccinated actually reduced the overall risk of Guillain-Barré syndrome compared to remaining unvaccinated, because the risk of GBS from flu infection outweighed the tiny risk from the vaccine.²⁴ For COVID-19, a population-based study comparing vaccinated and unvaccinated individuals who contracted SARS-CoV-2 found that vaccination did not increase autoimmune risk and in fact appeared to reduce the risk of several infection-associated autoimmune conditions, including immune-mediated thrombocytopenia, Graves’ disease, and lupus.²⁵

This makes intuitive sense. An actual infection exposes the immune system to far more foreign protein, for far longer, with far more inflammation than a vaccine does. If molecular mimicry and bystander activation are the mechanisms by which immune responses can go awry, a full-blown infection provides a much larger and more sustained trigger for those mechanisms than a controlled vaccine dose.

The Coincidence Problem

One of the hardest challenges in vaccine safety research is separating genuine side effects from conditions that would have happened anyway. Autoimmune diseases tend to emerge in young adulthood, which is also when many vaccines are given. If you vaccinate millions of adolescents against HPV, a predictable number will develop autoimmune conditions in the following months purely by chance. A cohort study designed to illustrate this point tracked the background rates of autoimmune and other medical conditions in the age group targeted for HPV vaccination before the vaccine was introduced, precisely to document what events would be expected to occur in temporal proximity to vaccination regardless of any causal link.²⁶

This is why anecdotal reports of autoimmune disease following vaccination, while understandably concerning to the individuals involved, cannot establish causation on their own. It is also why experts in vaccine safety have emphasized that for autoimmune conditions, careful case review of medical records is essential, because electronic health records alone can misclassify timing and diagnosis.²⁷ Clinical trials large enough to detect events occurring in fewer than 1 in 100 vaccinees would need to be enormous, which is why post-marketing surveillance, rather than pre-approval trials, is where most rare autoimmune signals are detected.²⁸

Early reviews of the evidence captured this tension well: even the “frequent” autoimmune associations with viral vaccines involve small numbers of patients, and whenever controlled studies were conducted, the evidence for a causal link usually dissolved.²⁹

The Lupus Question

Systemic lupus erythematosus has received more attention in the vaccine-autoimmunity literature than most other conditions. The Korean cohort study mentioned earlier found a modestly elevated risk of lupus after mRNA COVID-19 vaccination.³ A 2017 meta-analysis found that vaccinations in general were associated with an increased risk of lupus, and a similar association with rheumatoid arthritis.³⁰ A more recent 2024 meta-analysis, though, found no statistically significant overall association between vaccination and lupus, while noting that hepatitis B vaccination specifically did appear to carry a higher odds ratio.³¹

The disagreement between these meta-analyses is not unusual for a question this difficult to study. Lupus has a long preclinical phase during which the disease is developing silently before symptoms appear. A person vaccinated during that preclinical phase might be diagnosed shortly after vaccination without the vaccine having played any causal role. Observational studies struggle to account for this. The hepatitis B signal is the most persistent finding, but even there, the evidence is based on relatively few studies with high variability between them. If there is a real connection, it is likely limited to specific vaccines and specific people, not vaccines as a category.

Sex Differences in Vaccine Immune Responses

Women generally mount stronger antibody responses to vaccines than men. This heightened immune reactivity likely explains why women also experience more side effects after vaccination and, in the broader population, have a much higher incidence of autoimmune disease in general. The mechanisms behind this difference involve hormonal effects (estrogen tends to enhance immune responses while testosterone suppresses them), X-chromosome-linked immune genes, and differences in the gut microbiome.³² Whether this translates into a higher risk of vaccine-associated autoimmune events in women specifically has not been definitively established, but the biological plausibility is strong. The sex disparity in autoimmune disease is one of the reasons that teasing apart vaccine effects from background rates is so complicated, particularly for conditions like lupus that already overwhelmingly affect women.