For decades a Diagnosing diabetes relies largely on measuring blood sugar levels and checking whether they exceed the clinical threshold. But researchers are increasingly concerned that this approach misses millions of people who are already developing the disease.
Throughout the world, diabetes has become one of the most crucial health crises of today. According to the World Health Organization, 14% of adults had diabetes in 2022, up from 7% in 1990. More than 40 million people in the U.S. have diabetes, but about 11 million remain undiagnosed. There are over 115 million Americans estimated suffer from prediabetes, and about 80 percent do not know it. In the UK around 5.8 million people suffer from diabetes, of which as many as 1.3 million are considered undiagnosed.
“We’re talking about an epidemic that I think is much worse than the Covid pandemic,” says Michael Snyder, a professor of genetics at Stanford University. “We need new ways of approaching this problem.”
The danger is not just diabetes itself, but the damage that accumulates in silence in the years before diagnosis. Long-term high blood sugar levels raise the risk of heart disease, stroke, kidney failure, blindness and nerve damage. The earlier the disease is detected, the greater the chance of avoiding complications or avoiding diabetes altogether.
Diagnosis still relies largely on measuring blood glucose levels, most commonly using the HbA1c test, which assesses average blood sugar levels over the past few months. While it is widely used and generally reliable, it is not infallible. The results cannot reflect certain medical conditions or physiological factors that may affect blood sugar levels.
Scientists are increasingly concerned that existing diagnostic tools are also less effective in some populations. Latest research suggest In some black and South Asian people, HbA1c readings may be falsely low, delaying diagnosis until the disease is more advanced.
This disparity has fueled growing interest in more personalized and data-rich approaches to detecting diabetes: ones that combine biomarkers, wearable devices and artificial intelligence to identify risk earlier and understand the disease in greater detail.
At Stanford University, Snyder and colleagues studied whether continuous glucose monitors (CGM) – portable sensors that track glucose levels in real time – could reveal hidden metabolic patterns long before the conventional diagnosis of type 2 diabetes, which accounts for about 95 percent of cases. Although it is often associated with obesity – which is an crucial risk factor – thinner people can also develop type 2 diabetes. Snyder himself developed type 2 diabetes, even though he did not fit the stereotypical profile of the disease.
“Glucose regulation involves many organ systems: the liver, muscles, intestines, pancreas, and even the brain,” says Snyder. “There are many biochemical pathways and it is clear that glucose dysregulation cannot be isolated to one event.”
The Stanford team developed an artificial intelligence-based algorithm that analyzes patterns in CGM data to identify different forms of type 2 diabetes. In tests, the system identified some of these patterns with approximately 90% accuracy.
Scientists believe the discovery could lend a hand identify people who already have metabolic problems long before a conventional diagnosis of diabetes is made. “It’s a tool you can use to take preventive measures,” Snyder says. “If levels trigger a prediabetes warning, dietary or exercise habits can be adjusted, for example.”
CGMs are also becoming cheaper and more available, with many now available over the counter in the US. Snyder believes they could eventually become part of routine preventive health care. “In an ideal world, people would wear them once a year,” he says. “From our perspective, the goal is to keep people healthy rather than trying to fix them later.”