Type 1 diabetes is caused by the immune system mistakenly attacking and destroying the insulin-producing cells in the pancreas. Unlike type 2 diabetes, which is driven by lifestyle and metabolic factors, type 1 is an autoimmune disease with roots in genetics, immune system malfunction, and environmental triggers that researchers are still working to fully understand. No single cause explains every case. Instead, a combination of inherited risk and outside triggers sets the process in motion.
The Autoimmune Attack on Insulin-Producing Cells
The pancreas contains clusters of cells called islets, and within those islets, beta cells produce insulin. In type 1 diabetes, a specific type of immune cell (CD8+ T cells) identifies proteins on beta cells as foreign threats and begins killing them. These immune cells use the same tools they’d normally deploy against a virus: they punch holes in cell membranes and trigger self-destruction pathways. The attack is precise. The immune system targets beta cells specifically while leaving other neighboring cell types in the islets unharmed.
What makes the damage especially hard to reverse is that it spreads beyond the cells directly under attack. Research published in the journal Diabetes found that beta cells near the ones being destroyed, even those not in direct contact with immune cells, start malfunctioning too. These bystander cells show signs of inflammation, reduced insulin stores, and impaired ability to manufacture new insulin protein. So the immune attack has a ripple effect: it doesn’t just kill individual cells, it degrades the function of surviving ones. By the time symptoms appear, roughly 80 to 90 percent of beta cell function is already lost.
Genetic Risk Factors
Genes play a major role in who develops type 1 diabetes, though they don’t guarantee it. The strongest genetic link involves a set of immune system genes called HLA, which help the body distinguish its own cells from invaders. Two specific gene variants, known as HLA-DR4-DQ8 and HLA-DR3-DQ2, are found in the vast majority of people diagnosed. A Swedish study found these two variants account for 89% of newly diagnosed type 1 diabetes cases. Carrying both variants together raises the odds roughly eightfold compared to the general population.
On the flip side, certain gene variants appear to be protective. One called DQ6 was associated with near-complete protection in some populations studied, meaning people who carry it almost never develop the disease. This helps explain why type 1 diabetes runs in families but doesn’t follow a simple inheritance pattern. A child with a parent who has type 1 diabetes faces a higher risk than the general population, but the overall chance is still relatively low, typically between 3 and 8 percent depending on which parent is affected. Identical twins share about a 50% concordance rate, which confirms that genes matter enormously but also that something beyond genetics is needed to trigger the disease.
Viral Infections as a Trigger
Among environmental triggers, viral infections have the strongest evidence. A group of viruses called enteroviruses, particularly Coxsackie B viruses, have been found directly inside the beta cells of people with recently diagnosed type 1 diabetes. Research published in PNAS examined pancreatic tissue from six people who had just developed the disease and found active Coxsackie B4 virus in beta cells of half of them. The infection was remarkably specific: the virus was found inside insulin-producing cells but not in other cell types in the same islets. Electron microscopy showed that 76 to 88 percent of beta cells in infected samples contained viral material.
The virus appears to cause harm through several possible routes. It can directly kill beta cells through infection. It can also trigger a lower-grade inflammation that damages cells slowly, releasing internal cell contents that the immune system then misidentifies as threats. This “bystander damage” model has been supported by animal studies. A third possibility is molecular mimicry, where viral proteins resemble beta cell proteins closely enough that the immune system, after learning to fight the virus, accidentally targets beta cells too. These mechanisms aren’t mutually exclusive, and different people may experience different combinations.
Importantly, the virus alone doesn’t seem to be enough. The PNAS researchers found that inflammation from viral infection, without a pre-existing autoimmune tendency, did not destroy beta cells. Both insulitis (inflammation in the islets) and autoimmunity appear to be necessary. This fits with the broader understanding that type 1 diabetes requires a genetic predisposition plus an environmental spark.
Autoantibodies: Early Warning Signs
Years before symptoms appear, the immune system produces detectable antibodies against beta cell proteins. Four main autoantibodies have been identified: those targeting an enzyme called GAD65, a protein called IA-2, insulin itself, and a zinc transporter protein known as ZnT8. These antibodies don’t cause the disease directly, but they signal that the autoimmune process is underway.
The number of different autoantibodies a person has matters more than which specific ones are present. Someone with a single autoantibody has a modest risk of progressing to diabetes, but the presence of two or more dramatically increases the likelihood. Children tend to develop multiple autoantibodies more often than adults. Screening for these markers is increasingly used in research settings and family screening programs to identify people in the earliest stages of the disease, sometimes years before they would need insulin.
The Gut Microbiome Connection
The community of bacteria living in the gut appears to differ in people who develop type 1 diabetes compared to those who don’t. Research published in The Lancet found that at the time of diagnosis, people with type 1 diabetes had higher levels of a bacterial group called Bacteroides, while healthy controls had more Prevotella. Levels of Bifidobacterium, a group of bacteria generally considered beneficial, were lower in those who developed the disease.
More telling is what happens in the period between when autoantibodies first appear and when diabetes is diagnosed. During this window, researchers observed a drop in overall bacterial diversity along with an increase in pro-inflammatory organisms and shifts in metabolic byproducts found in both stool and blood. Whether these microbial changes help cause the disease or simply reflect the same immune dysfunction driving it remains unclear. But the consistency of the pattern across multiple study populations, including children in Finland, Sweden, Colorado, and Washington state, suggests the gut environment plays some role in how quickly or aggressively the autoimmune process unfolds.
The Hygiene Hypothesis
Type 1 diabetes rates have been climbing steadily in industrialized countries for decades, and the disease is more common in firstborn children and in wealthier families. This pattern led to the hygiene hypothesis: the idea that growing up in cleaner environments with fewer childhood infections may leave the immune system poorly calibrated, making it more likely to turn against the body’s own tissues.
The theory has intuitive appeal, and some data supports pieces of it. Having older siblings and being exposed to more infections early in life does seem to influence immune development. However, the specific immunological reasoning doesn’t hold up neatly for type 1 diabetes. The disease involves a particular branch of the immune response, and as researchers noted in The BMJ, the evidence actually contradicts some of the predicted immune shifts the hygiene hypothesis would require. The rising incidence is real, but the explanation is likely more complex than reduced infection exposure alone.
Vitamin D and Other Proposed Factors
Vitamin D has received significant attention as a potential protective factor, partly because type 1 diabetes is more common in northern countries with less sunlight. Some observational studies have found associations between low vitamin D levels and higher risk. However, a large genetic analysis published in PLOS Medicine used a technique called Mendelian randomization to test whether vitamin D levels actually cause changes in risk, rather than simply correlating with it. The result: decreased vitamin D levels did not have a substantial impact on type 1 diabetes risk in the populations studied. This suggests that the geographic patterns in type 1 diabetes rates are driven by something other than vitamin D alone.
Other environmental factors under investigation include early dietary exposures (such as the timing of introducing cow’s milk or gluten to infants), childhood growth rate, and maternal factors during pregnancy. None of these have been established as definitive causes, but they reflect the broader principle that type 1 diabetes likely results from multiple small environmental nudges acting on a genetically susceptible immune system.
When It Develops in Adults
Type 1 diabetes is often thought of as a childhood disease, but it can develop at any age. When autoimmune diabetes appears in adults, it sometimes progresses more slowly and is initially misdiagnosed as type 2 diabetes. This slower form is sometimes called latent autoimmune diabetes in adults, or LADA. Roughly 10% of people diagnosed with type 2 diabetes actually have islet autoantibodies similar to those seen in type 1.
LADA shares the same fundamental cause, an immune attack on beta cells, but research from the American Diabetes Association has identified differences in which autoantibodies are present, which beta cell proteins the immune system targets, and how much insulin resistance accompanies the condition. These immunological differences suggest that LADA may involve a partially distinct disease process rather than simply being a delayed version of childhood-onset type 1 diabetes. People with LADA typically retain some insulin production for longer, which is why they can initially manage without insulin injections, but most eventually need insulin as beta cell destruction continues.