According to research reviewed by Gram, early-onset type 2 diabetes is rising worldwide, and high body mass index showed the most consistent link to it across 204 countries from 1990 to 2021. Diets low in whole grains also appeared linked, though that connection varied by region.
Type 2 diabetes has long been linked with older age, but new research shows it is increasingly appearing in people between 15 and 39 years old. According to research reviewed by Gram, a large analysis of data from 204 countries and territories between 1990 and 2021 found that rates of this early-onset form have been climbing worldwide. The strongest and most consistent factor linked to these cases was high body mass index, a measure of body weight relative to height. Diets low in whole grains also showed a connection, but that link varied depending on where people lived and how the data were analyzed. The researchers used several machine learning tools to test which factors best predicted rising rates. Because the study looked at whole populations rather than individual people, it cannot prove that any single factor causes the disease. Still, the findings point to body weight and diet as important areas for prevention.
Key Statistics
A 2026 ecological analysis of Global Burden of Disease data covering 204 countries and territories found that early-onset type 2 diabetes incidence among people aged 15 to 39 increased globally from 1990 to 2021.
In the same 2026 study, high body mass index was one of only two exposures that ranked highest across all three machine learning models and remained consistently linked to early-onset type 2 diabetes incidence.
The Quick Take
- What they studied: The researchers wanted to know whether type 2 diabetes is becoming more common in young adults around the world, and which everyday factors, such as body weight and diet, are linked to those rising rates.
- Who participated: The study did not enroll individual people. Instead, it used national health estimates for 204 countries and territories, covering adults aged 15 to 39 from 1990 to 2021.
- Key finding: Early-onset type 2 diabetes rose globally over the 32-year period. High body mass index was the factor most consistently linked to higher rates across all the analytical methods used, while low whole-grain intake was linked in a less consistent way.
- What it means for you: Maintaining a healthy body weight and eating plenty of whole grains are sensible habits that support overall health. Because this study looked at national patterns, it cannot tell you your personal risk, so talk with a doctor about blood sugar testing if you have concerns.
The Research Details
This was an ecological study, which means it looked at groups of people, in this case entire countries, rather than following individuals over time. The researchers drew on the Global Burden of Disease 2021 database, a large collection of health estimates maintained by international research teams. They tracked how often early-onset type 2 diabetes was diagnosed in each country and how those rates changed from 1990 to 2021.
To find out which factors were linked to these rates, the team tested many possible exposures, such as body weight, diet, and other population-level traits. They used three types of machine learning models, which are computer programs that look for patterns in large datasets. They trained the models on earlier years, checked them on a middle period, and tested them on the most recent years. This approach helped show which models could best predict future data rather than simply fitting past data.
The team then ran additional checks to see whether the results held up. They repeated the analysis after randomly resampling countries, combined the rankings from different models, and used statistical models that account for differences between countries and between years. These steps help separate patterns that appear consistently from those that appear by chance.
Tracking rates over several decades across many countries reveals long-term trends that a single study in one city or one clinic could miss. Testing predictions on years the models had not seen is a stronger check than simply describing past data. Repeating the analysis with different methods also makes it less likely that one statistical choice drove the results.
The study uses a large, well-known data source and applies several independent checks, which increases confidence that the main patterns are real. However, national estimates for some countries may be less precise than others, and the data reflect what was recorded, not necessarily what happened. Because the study is observational, it can show that factors and disease rates tend to rise together, but it cannot prove that one causes the other. The journal’s impact factor is an approximate estimate and is not a measure of this study’s individual quality.
What the Results Show
The study found that early-onset type 2 diabetes incidence increased across the world between 1990 and 2021. Some Pacific Island countries had persistently high rates throughout the period, while countries such as Egypt, Greenland, and Afghanistan showed rapid increases. These patterns indicate that the rise is not limited to one region.
Of all the factors tested, high body mass index and diets low in whole grains were the only two that remained among the strongest predictors after the researchers corrected for the many comparisons they made. Among the three machine learning models, XGBoost made the most accurate predictions for future years. Both factors ranked highest across models and across time periods, which made them stand out from the rest.
When the researchers used a simpler statistical model that accounts for shared changes across all countries in a given year, both body weight and whole-grain intake were positively associated with higher rates. When they further accounted for differences that stay constant within each country, high body mass index remained a significant factor, while the whole-grain link was less consistent. This suggests that population body weight had the most steady connection to the disease.
The team also built combined exposure scores to map how multiple risk factors clustered geographically. These scores differed widely from place to place, though the general patterns looked similar regardless of how the factors were weighted.
The combined exposure scores revealed that some regions face several risk factors at once, which could compound their risk of early-onset diabetes. This geographic picture may help public health planners decide where to focus prevention efforts. The findings also highlight that the rise in early-onset cases is a global concern rather than a problem confined to high-income nations.
Type 2 diabetes is widely known to be strongly linked with excess body weight, and this study supports that long-standing understanding at the population level. The link between whole-grain intake and diabetes has been reported in many earlier studies, which makes the weaker and more variable connection in this analysis worth noting. Research on early-onset cases specifically is less developed than research on adults in general, so this study adds useful global context.
Because the study compared countries rather than individuals, it cannot determine how any one person’s weight or diet affects their own risk. Data quality varies across countries, and some estimates rely on modeling rather than direct measurement. The study shows associations, not cause and effect, and other unmeasured factors such as physical activity, genetics, sleep, and access to healthcare may also play a role. The machine learning approach is useful for finding patterns, but its predictions depend on the quality of the data it receives.
The Bottom Line
Based on this research, maintaining a healthy body weight appears to be an important factor in preventing type 2 diabetes, and the evidence for this link is strong and consistent (high confidence). Eating more whole grains is a reasonable part of a healthy diet, though the evidence specifically linking whole-grain intake to early-onset diabetes is less consistent (moderate confidence). Anyone with symptoms such as unusual thirst, frequent urination, or fatigue should consult a healthcare provider for a blood sugar test.
Young adults, parents, and anyone concerned about weight and metabolic health may find these findings useful. Public health planners and clinicians may also use the global patterns to focus prevention efforts. People who already have a diabetes diagnosis should follow their own treatment plan rather than changing medication or diet based on this study alone.
Changes in body weight and diet can begin to affect blood sugar regulation within weeks, but measurable changes in long-term diabetes risk typically take months to years of sustained effort. Because this study covers population trends, it does not offer a timeline for individual outcomes. A doctor can help set realistic goals based on personal health history.
Frequently Asked Questions
Is type 2 diabetes becoming more common in young adults?
Research shows early-onset type 2 diabetes, affecting people aged 15 to 39, is rising globally. An analysis of 204 countries and territories from 1990 to 2021 found persistently high rates in Pacific Island countries and rapid increases in Egypt, Greenland, and Afghanistan.
What is the biggest risk factor linked to early-onset type 2 diabetes?
High body mass index, a measure of body weight relative to height, showed the most consistent link to early-onset type 2 diabetes across machine learning models and analytical methods. The study is observational, so it shows association at the population level rather than proving cause.
Does eating whole grains lower the risk of early-onset type 2 diabetes?
This study found that low whole-grain intake was linked to higher early-onset type 2 diabetes rates in some settings, but the connection was less consistent than body weight. Because this was a population-level analysis, it cannot confirm that whole grains directly prevent the disease in individuals.
Can a population-level study tell me my personal diabetes risk?
A population study cannot predict one person’s risk. This analysis compared national averages across 204 countries, so it describes broad patterns. A doctor can assess personal risk using blood sugar tests, body weight, family history, and lifestyle habits.
Want to Apply This Research?
- Log your body weight once a week at the same time of day, and record the number of servings of whole grains (such as oats, brown rice, or whole wheat bread) you eat each day. Reviewing these numbers monthly can show whether your habits are moving in a healthy direction.
- Try swapping one refined grain serving per day, such as white bread or white rice, for a whole-grain option. Pair this with a short daily walk to support healthy body weight. Set a reminder in the app to check off each change so the habit becomes routine.
- Review your weight and whole-grain logs every month and look for trends over three to six months rather than day-to-day fluctuations. Schedule regular checkups with a healthcare provider, who can order blood sugar tests to track your personal risk over time. Share your app summaries with your doctor to make visits more productive.
This article summarizes an observational study of national health data and is for general information only. It does not establish that any food or body weight change causes or prevents diabetes in an individual. Anyone with symptoms of diabetes or concerns about their blood sugar should consult a qualified healthcare provider before making changes to diet, exercise, or medication.
This research translation is published by Gram Research, the science division of Gram, an AI-powered nutrition tracking app.