Type 2 diabetes has long been linked to poor physical outcomes, but a new nationwide study suggests it may also increase the risk of very late-onset schizophrenia.
Researchers in Israel followed nearly 100,000 adults, aged 50 to 72 at enrollment, for more than 14 years to investigate whether diabetes plays a role in psychosis onset. The study, published in Schizophrenia Bulletin, found that people who developed type 2 diabetes had a 50% higher risk of being diagnosed with schizophrenia compared to those without diabetes.
Background
Traditionally, the association between schizophrenia and diabetes has been described the other way around. People with schizophrenia are known to have a two- to three-fold higher risk of developing type 2 diabetes, often attributed to antipsychotic medication, sedentary lifestyle, and obesity. Some studies have even reported abnormal glucose regulation in individuals before their first psychotic episode.
But whether diabetes itself raises the risk of schizophrenia has remained an open question. Genetic evidence and family studies have hinted at shared risk factors: siblings of schizophrenia patients show higher diabetes rates, and common genetic variants appear to increase the risk of both disorders. Until now, however, no large-scale prospective study had tested the hypothesis directly.
The Study
The research team analyzed data from Meuhedet Healthcare Services, one of Israel's national health maintenance organizations, which covers around 14% of the country's population. The cohort included 99,567 individuals born between 1932 and 1952 who had no history of diabetes or schizophrenia at the start of follow-up in 2005. Participants were tracked until early 2020, with an average follow-up of more than 14 years.
During the study period, the incidence of schizophrenia was 0.26 per 10,000 person-years among those with diabetes and 0.12 among those without. After adjusting for 20 potential confounders - including smoking, obesity, cardiovascular disease, depression, and socioeconomic status - the researchers found a hazard ratio of 1.53, meaning that diabetes increased the risk of schizophrenia by 53%.
Robust Findings
The team conducted nine complementary analyses to test the robustness of the results. They looked at sex differences, excluded early cases to minimize reverse causation, restricted analyses to individuals with multiple schizophrenia diagnoses, and examined the effect of antidiabetic medication. Across these tests, the association remained consistent.
Interestingly, the increased risk was stronger in women than in men. This finding may help explain why schizophrenia diagnoses after age 60 are more common among females, possibly linked to hormonal changes such as estrogen loss.
Antidiabetic medications, including insulin and glucose-lowering drugs, were not associated with increased schizophrenia risk, suggesting that the link stems from the disease itself rather than its treatment.
Possible Explanations
Several mechanisms may explain the association. One possibility is that diabetes and schizophrenia share genetic pathways, a theory supported by genome-wide studies showing overlapping risk regions. Another is that the metabolic effects of diabetes - such as inflammation, vascular changes, or cognitive decline - create vulnerabilities that can trigger psychosis in later life. Stress-related biological processes common to both conditions may also contribute.
The researchers caution that while diabetes appears to increase risk, their findings do not prove causality. It remains unclear whether diabetes directly causes schizophrenia, unmasks latent vulnerabilities, or interacts with genetic and environmental factors to influence onset.
Implications
The study provides the strongest evidence to date that diabetes is linked not only to cardiovascular and metabolic disease but also to serious mental health outcomes. With global diabetes rates rising sharply, the findings carry public health importance.
"Clinicians treating older adults with type 2 diabetes should be alert to potential psychotic symptoms," the authors write. Monitoring for hallucinations, delusions, or cognitive decline may allow for earlier intervention and better patient outcomes.
Limitations
The study has limitations. Schizophrenia diagnoses were based on health records rather than clinical interviews, and some cases may have been missed. The analysis also lacked detailed information on ethnicity, HbA1c levels, or symptom subtypes, making it difficult to explore finer-grained mechanisms. Still, the large sample size, long follow-up, and extensive confounder adjustments make the findings highly robust.
Conclusion
The Israeli nationwide cohort study concludes that type 2 diabetes is associated with a significant, long-term increase in the risk of very late-onset schizophrenia. As populations age and diabetes prevalence grows, the study underscores the need for integrated care models that bridge endocrinology, psychiatry, and geriatrics.
