The Diabetes Epidemic in Older Adults
Type 2 diabetes (T2D) affects more than 500 million people worldwide. It develops when the body stops using insulin effectively, causing blood sugar to rise to harmful levels. Over time, uncontrolled diabetes damages the eyes, kidneys, nerves, and heart.
Older adults are at particularly high risk. Yet catching diabetes before it fully develops — when lifestyle changes can still prevent or delay it — requires tools that actually work. BMI (body mass index, your weight relative to your height) and waist circumference are the most commonly used screening tools, but they only tell part of the story.
What We Thought We Knew
For years, doctors used BMI and waist circumference as the standard proxies for metabolic risk. If your BMI was high, you were flagged for closer monitoring. If your waist was large, that was a warning sign.
But here is the twist: BMI and waist circumference measure body size, not metabolic health. Two people with the same BMI can have very different levels of insulin resistance and fat distribution in their organs. The newer scores tested in this study go deeper, incorporating how the body handles fat and blood sugar together.
What Makes These Scores Different
The two standout scores — called TyG and TyHGB — both combine triglyceride levels (a type of fat in the blood) with other measurements.
TyG combines triglycerides with fasting blood glucose (blood sugar after fasting overnight). Think of it like a single number that captures how well your body is managing both fat and sugar at the same time — two interconnected metabolic pathways. When both are running hot, the risk of diabetes rises sharply.
TyHGB adds hemoglobin (a protein in red blood cells) into the equation. This gives it a longer-term view of blood sugar trends because hemoglobin reflects average glucose levels over months, not just one morning's reading.
Who Was Studied
Researchers analyzed data from 18,251 community-dwelling elderly adults enrolled in the BaHLS cohort study in Shenzhen, China. Participants were followed from baseline through the end of 2022. By that point, 1,350 people — about 7.4% of the group — had been newly diagnosed with type 2 diabetes. Researchers tracked how well each of the six metabolic scores predicted who would develop diabetes, comparing their accuracy using standard statistical tools.
Every one of the six scores was associated with diabetes risk. But TyG and TyHGB stood apart. For every one standard deviation increase in TyG, the risk of developing type 2 diabetes increased by 76% (hazard ratio 1.76). TyHGB came in close behind, with a 37% increase in risk per standard deviation.
By comparison, BMI was associated with only a 17% increase per standard deviation. Waist circumference added a bit more, at 23%. The gap is significant — it means TyG carries roughly four to five times more predictive signal per unit than BMI.
That's not the full story, though.
One of the most important findings was that TyG and TyHGB predicted diabetes even in people whose fasting blood glucose was still technically normal. This is a critical group: people who have not yet crossed the threshold into pre-diabetes, but whose metabolic health is quietly deteriorating. Catching risk at this stage — before blood sugar even starts to rise — offers the largest window for prevention.
The tools your doctor currently uses may not be capturing your full diabetes risk picture, especially if you are over 60.
TyG and TyHGB are not yet standard tools in most clinical settings. However, both use information from routine blood tests — a lipid panel for triglycerides and a hemoglobin test — that many patients already get regularly. If you are an older adult with any concern about diabetes risk, it may be worth asking your doctor whether your blood test results could be used to calculate these scores. In the meantime, the evidence for lifestyle interventions — reducing processed carbohydrates, increasing physical activity, managing weight — remains strong regardless of which screening tool is used.
This study was conducted entirely in elderly adults in one city in China (Shenzhen), which limits how well the results apply to younger populations or to people in other countries with different diets, genetics, and healthcare access. The study also could not fully account for all possible confounding factors. Additionally, the optimal cutoff values identified in the study would need validation in other populations before being used widely in clinical practice.
Researchers are calling for prospective validation studies in other countries and ethnic groups to test whether TyG and TyHGB perform as well across different populations. If the findings hold up, the next step would be developing standardized clinical guidelines for incorporating these scores into routine screening — potentially alongside or instead of BMI-based risk calculators. Given that the measurements needed are already commonly ordered, the barrier to adoption could be relatively low once the evidence base is firmly established.