Diabetes-related vision loss rose among young adults in global burden analysis

What happened

A global disease-burden analysis reports that blindness and vision loss attributed to type 2 diabetes became more prevalent among adults aged 20–39 between 1990 and 2021. The peer-reviewed study was published August 18, 2026, in the *International Journal of Ophthalmology*.

The modeled prevalence rose from 4.7 to 6.4 cases per 100,000 people in that age range. The corresponding rate of years lived with disability, or YLDs, increased from 0.5 to 0.7 per 100,000. Average annual percentage changes were approximately 1% for both measures, with the reported trends meeting the study’s threshold for statistical significance.

These are population-level estimates covering historical data through 2021—not counts observed in 2026. They nevertheless reinforce a public-health concern: developing type 2 diabetes earlier in adulthood creates more time for chronic microvascular complications to emerge and affect working-age healthspan.

How the study was conducted

Researchers analyzed cross-sectional estimates from the Global Burden of Diseases, Injuries, and Risk Factors Study 2021. The population comprised adults aged 20–39, with results stratified by age, sex, geographic region and sociodemographic index.

Prevalence measured how common diabetes-attributable blindness and vision loss were in the population. YLDs combined estimated prevalence with disability weights intended to represent the health loss associated with different levels of visual impairment. Joinpoint regression was used to identify changes in historical trends, while a Bayesian age-period-cohort model generated projections through 2045.

This was not a clinical cohort in which investigators examined participants’ eyes over time. GBD analyses synthesize available epidemiological sources and statistical models to estimate disease burden across places and years, including settings where direct measurements are sparse.

Where the burden changed

The study reported the highest regional prevalence in Southeast Asia and the highest country-level prevalence in Grenada. Central Latin America had the highest regional YLD rate, while Cambodia had the highest country-level YLD rate. Rates also increased in the high-sociodemographic-index group.

Women had steeper modeled increases than men. The average annual percentage change in prevalence was 1.16% for women and 0.88% for men; the corresponding YLD changes were 1.16% and 0.82%. Singapore was the only location reported to have declining trends in both prevalence and YLD rates.

Those comparisons do not reveal why locations or sexes differed. Possible contributors—including diabetes prevalence, duration, glucose and blood-pressure control, access to retinal screening, treatment availability and underlying data quality—were not tested as causal explanations by this analysis.

Practical meaning—and important limits

Diabetic retinopathy is a neurovascular complication whose risk is strongly related to diabetes duration and chronic hyperglycemia. The American Diabetes Association’s 2026 standards recommend an initial comprehensive eye examination when type 2 diabetes is diagnosed. The standards also support validated retinal-photography programs, including remote reading or authorized artificial-intelligence systems, as ways to expand screening access.

The World Health Organization similarly describes organized retinal screening and prompt treatment as tools for reducing diabetes-related visual impairment and blindness. The new analysis does not test a screening program, treatment or diagnostic technology, however. It quantifies a modeled historical burden.

Several limitations constrain interpretation. Estimates depend on the quantity and quality of underlying data, diagnostic definitions, disability weights and modeling assumptions. Sparse surveillance can widen uncertainty and make comparisons between countries less secure. Aggregated results cannot show whether individual cases received timely screening or treatment, and ecological trends cannot establish that any particular policy or risk factor caused the changes.

The projection to 2045 is also a scenario generated from past patterns, not an observed outcome. Future diabetes incidence, clinical care and screening coverage could alter that trajectory.

The most defensible conclusion is therefore narrower than a forecast of inevitable worsening: available global evidence suggests that diabetes-attributable vision loss increased among younger adults through 2021. Better age-specific surveillance will be needed to verify local trends and determine where prevention and eye-care systems are falling short.