Threshold performance of anthropometric indices for type 2 diabetes in Asian adults: a 20-year systematic review and meta-analysis

For decades, doctors have relied on a simple number to gauge a person's risk of developing type 2 diabetes: the body mass index, or BMI. This calculation uses a person's height and weight to place them into categories of underweight, normal weight, overweight, or obese. The logic seemed straightforward: the more weight a person carries, the higher their risk. However, a growing body of evidence has shown that this rule does not apply equally to everyone. In adults of Asian ancestry, the curve shifts. People in these populations can develop diabetes at a much lower body weight than their counterparts in Europe or North America. A body size that might be considered healthy for one group can signal a serious health threat for another. This discrepancy has created a difficult problem for public health officials and clinicians: how do you screen for a disease that hides in bodies that do not look "overweight" by traditional standards?

To solve this, researchers have turned to other ways of measuring body shape. Instead of just weighing a person, they look at where fat is stored. Some measures focus on the waist, such as the waist circumference, which measures the girth around the middle. Others compare the waist to the hips, or the waist to the total height. These measurements aim to catch "central adiposity," a condition where fat accumulates around the internal organs, a pattern that is particularly dangerous for metabolic health. The question has been whether these tape-measure tools are better than the scale at predicting who will develop diabetes, and if so, what specific number on the tape measure should trigger a warning.

A comprehensive review of twenty years of research, covering studies from 2006 to 2026, has now brought these scattered findings together. The authors analyzed data from over 145,000 adults of Asian ancestry, drawn from studies across Asia and in diaspora communities. They looked at how well various body measurements could distinguish between people who had diabetes and those who did not. The goal was not to create a perfect diagnostic test, but to find the most reliable screening tool—a simple rule that could tell a doctor which patients need a blood test to confirm a diagnosis. The review examined the performance of the body mass index alongside newer indices like the waist-to-height ratio, the waist-to-hip ratio, and several complex calculations that mix body measurements with blood lipid levels.

The results paint a clear, if modest, picture. No single body measurement is a perfect predictor. The best tools found in the study performed only "fairly" well, meaning they would miss about one in three people with diabetes and incorrectly flag about one in three people without the disease. However, among the imperfect options, one tool stood out as the most consistent. The waist-to-height ratio, which is calculated by dividing a person's waist measurement by their height, proved to be the most effective screening index. In the studies reviewed, this ratio consistently outperformed the body mass index. While the body mass index is a measure of total weight relative to height, the waist-to-height ratio specifically targets the dangerous fat stored around the abdomen. The researchers found that this simple ratio was better at identifying diabetes risk across different groups of men and women, and across various countries in Asia.

The study also looked at more complex tools that require a blood test for lipids, such as the lipid accumulation product. These indices, which combine waist size with cholesterol and triglyceride levels, did show slightly better performance than the waist-to-height ratio in direct comparisons within the same groups of people. However, the authors note a significant practical limitation: these tools require a blood draw, which defeats the purpose of a quick, non-invasive screening tool. If a patient is already having their blood drawn for lipids, the doctor might as well test for diabetes directly. Therefore, for a quick check in a community setting or a doctor's office, the tape measure remains the most practical choice.

When it comes to the specific numbers that should trigger a warning, the review offers a clear recommendation for the waist-to-height ratio. The researchers found that a ratio of 0.5 is the most defensible boundary for screening. This means that if a person's waist is half their height or more, they should be considered at risk and offered a diabetes test. This rule is simple to remember and apply. The data showed that while the optimal number varied slightly between men and women in different studies, the 0.5 threshold sits comfortably within the range of what the data supports, balancing the need to catch as many cases as possible without causing too many false alarms.

For the body mass index, the findings reinforce a shift that has already begun in medical guidelines. The review confirms that the traditional threshold of 25 for overweight is too high for Asian adults. The data suggests that a body mass index of 23 is a more appropriate point to start testing for diabetes. This number is lower than the standard used for European populations, reflecting the fact that Asian adults develop the disease at a lower body weight. The authors emphasize that this lower number is a trigger for action—a signal to check blood sugar levels—rather than a new definition of obesity. It is a safety net designed to catch the disease earlier, before it causes damage.

The review also highlights the limitations of the current research. The studies included in the analysis showed a great deal of variation, with results differing widely depending on the country, the age of the participants, and how the measurements were taken. This variation means that the numbers found in the study are not absolute laws but rather the best estimates available from the existing literature. Furthermore, the review noted a significant gap in the data: very few studies from South Asia, a region with a very high burden of diabetes, were included in the final analysis. This suggests that the current recommendations might need adjustment as more data becomes available from those specific populations.

Ultimately, the study concludes that while body measurements are useful for deciding who needs a diabetes test, they cannot replace the test itself. A tape measure or a scale can tell a doctor who is at risk, but only a blood test can confirm the disease. The waist-to-height ratio of 0.5 and a body mass index of 23 emerge as the most reliable guides for this initial screening step. These tools offer a practical way to adapt medical advice to the reality of Asian bodies, ensuring that the warning signs are not missed simply because a person does not look "overweight" by older, Western standards. The path forward involves using these simple, accessible measures to identify those who need further care, while continuing to gather more data to refine these rules for every corner of the Asian world.