Why One Diabetes Threshold Does Not Fit Every Population

The short version is this. The body's handling of blood sugar rests on two linked systems: how sensitive your tissues are to insulin, and how vigorously your pancreas responds by releasing it. The balance between those two systems is not the same across the world's populations. When you compare groups, a person from one background can show the same fasting glucose as a person from another while sitting at a meaningfully different distance from diabetes. That is why a single global threshold, applied without context, can both miss people who are at real risk and flag people who are not.

This is not a fringe idea. It comes out of a large body of physiological measurement, and it matters for how we screen, diagnose, and counsel patients. It is also educational here, not medical advice. If you have questions about your own glucose numbers, talk to a clinician who can see your full picture.

The two dials behind every glucose reading

Picture glucose control as two dials that move together. The first is insulin sensitivity: how readily muscle, liver, and fat take up glucose when insulin tells them to. The second is the insulin response, sometimes called beta-cell function or insulin secretion: how much insulin the pancreas puts out to get the job done.

In healthy people these two dials are coupled in a predictable way. When sensitivity falls, the pancreas compensates by secreting more insulin, and glucose stays in range. The relationship is curved, not straight. Researchers often describe it as a hyperbola, and the product of the two measures is sometimes called the disposition index, a rough gauge of how well the whole system is holding up. Type 2 diabetes tends to develop when the pancreas can no longer ramp up secretion fast enough to cover falling sensitivity. The dial that was supposed to compensate runs out of room.

The trouble is that a fasting glucose value, or even a single oral glucose tolerance test, does not tell you which dial is doing what. Two people can land on the same number by very different routes. One may be highly insulin resistant but blessed with a pancreas that compensates hard. Another may be quite insulin sensitive but secreting relatively little. They look identical on the lab slip and are not in the same situation at all.

What the cross-population evidence actually shows

When you line up studies from around the world, the coupling between sensitivity and secretion does not sit in the same place for every group. Some populations, on average, show lower insulin sensitivity at a given body size, with the pancreas compensating through higher secretion. Others tend to sit at the opposite end, more sensitive but with a more modest insulin response. East Asian populations, for instance, have repeatedly shown relatively lower insulin responses, while several South Asian and other groups show pronounced insulin resistance. These are population averages with wide individual scatter, not labels that tell you about any one person.

A systematic review and meta-analysis in Diabetes Care, which I co-authored, set out to map exactly this. It pooled studies measuring insulin sensitivity and insulin response across ethnic groups and examined how the two related from one population to the next. The work is widely cited. The finding is consistent: the sensitivity-response relationship is not a single universal curve. It shifts by population.

That shift has a practical edge. If your diagnostic cutoffs, your risk calculators, and your body-mass-index thresholds were calibrated mostly in one population, they will not translate cleanly to another. This is part of why some health systems now use lower BMI thresholds to flag diabetes risk in South Asian and East Asian patients. The biology was telling us, in effect, that the same waistline carries different metabolic weight in different people.

From averages to individuals

None of this means ethnicity is destiny, and it is not a substitute for measuring the actual person in front of you. Population averages hide enormous individual variation, and ethnicity is a crude stand-in for a tangle of genetic, developmental, dietary, and social factors. The honest reading of the evidence is humbler and more useful: a one-size-fits-all threshold is a starting point, not a verdict, and clinicians should hold it loosely when the patient's background suggests the curve sits elsewhere.

This is where precision medicine earns its name. The goal is not to swap one rigid cutoff for several rigid cutoffs by group. It is to measure both dials where we can, account for the patient's context, and stop treating a single glucose number as if it told the whole story. The richer the picture, the earlier and more accurately we can catch the people whose pancreas is quietly losing its ability to compensate, often years before a standard test crosses the line.

That earlier, more individual picture is what good decision-support tools are built to support, helping a clinician weigh more of the relevant signal at once rather than leaning on a lone threshold.