An Average Blood Sugar of 139 Is What A1c?

An average blood sugar of 139 mg/dL corresponds to an A1c of roughly 6.5%, based on the widely used conversion developed in the A1c-Derived Average Glucose (ADAG) study. That number is clinically meaningful: 6.5% is the threshold most guidelines use to diagnose diabetes. But the relationship between average glucose and A1c is not as clean as the formula suggests, and understanding where the number comes from, what can throw it off, and what it means for your health matters more than the math itself.

Where the 6.5% Comes From

The conversion between average blood sugar and A1c rests on a large international study published in 2008, which tracked participants wearing continuous glucose monitors while also getting regular A1c blood draws. The researchers found a strong straight-line relationship between the two measurements. Using that data, they derived a formula that lets you move back and forth between average glucose and A1c. When you plug 139 mg/dL into that formula, the result is an A1c of about 6.5%.¹

The formula works well at a population level, meaning it gives a good estimate for most people most of the time. But the fit is not perfect. In the original study, the correlation between average glucose and A1c was strong but not absolute, which means two people with the same average glucose could end up with somewhat different A1c readings. That built-in wiggle room is worth remembering: your personal number could be a bit above or below the formula’s prediction.

Why 6.5% Is a Loaded Number

An A1c of 6.5% is not just a data point. It is the line most professional guidelines draw between prediabetes and diabetes. If your A1c comes back at 6.4%, you are classified as prediabetic. At 6.5%, you are classified as diabetic. The difference between those two labels can change the medications your doctor prescribes, the monitoring schedule you follow, and even your insurance coverage.

That bright-line cutoff is clinically useful but somewhat arbitrary. Research on the best A1c threshold for diagnosing diabetes has produced slightly different numbers depending on the population studied. One analysis found that an A1c of 6.1% optimized both sensitivity and specificity for detecting diabetes, suggesting the commonly used 6.5% threshold may miss some cases.² The practical takeaway: if your average glucose is sitting around 139 and your A1c is hovering near 6.5%, you are in a zone where the exact label matters less than the trend. Whether that number is climbing, holding steady, or coming down tells you more than whether it technically crosses the diagnostic line on any single blood draw.

What A1c Actually Measures

A1c is not a direct blood sugar reading. It measures how much glucose has chemically attached itself to your hemoglobin, the oxygen-carrying protein in red blood cells. Glucose sticks to hemoglobin slowly and continuously over the life of each red blood cell, which lasts about 120 days.³ The more glucose circulating in your blood over those months, the more hemoglobin gets coated. That is why A1c reflects an average, roughly weighted toward the most recent four to eight weeks.

This averaging effect is both the test’s greatest strength and its biggest blind spot. A1c smooths out daily highs and lows into a single number, which makes it convenient. But it also means two people with an identical A1c of 6.5% could have very different daily glucose patterns. One might hold steady around 139 all day. The other might swing between 70 and 210 repeatedly. Both get the same A1c, yet the second person’s pattern of wide glucose swings may carry additional cardiovascular risk that the A1c alone does not capture.⁴

When the Formula Gets It Wrong

Because A1c depends on hemoglobin, anything that changes your red blood cells can throw off the result. If your red blood cells are turning over faster than normal, each cell has less time to accumulate glucose, and your A1c will read artificially low even if your actual blood sugar is elevated. This happens in conditions like hemolytic anemia, significant blood loss, and chronic kidney disease, where shortened red blood cell survival skews the measurement downward.⁵ Conversely, iron-deficiency anemia can push A1c artificially high because red blood cells last longer when iron is scarce.

Certain hemoglobin variants, common in people of African, Mediterranean, and Southeast Asian descent, can also interfere with A1c measurement depending on the lab method used. The result might read falsely high or falsely low. If you have a known hemoglobin variant, your doctor may need to use a specific assay method or interpret the result with extra caution.

Racial and Ethnic Differences in the Glucose-A1c Relationship

Even without hemoglobin variants, the average-glucose-to-A1c conversion does not land in the same place for everyone. Studies have consistently found that Black individuals tend to have higher A1c readings than white individuals at the same average blood glucose level. One large analysis found that for a given glucose concentration, the average A1c in Black participants was about 0.4 percentage points higher than in white participants.⁶

A separate cross-sectional analysis confirmed this pattern across people with normal glucose tolerance, prediabetes, and diabetes alike. Even after adjusting for plasma glucose levels and other factors known to affect A1c, Black participants showed consistently higher readings, with the gap growing larger in people with diabetes.⁷ The reasons are not entirely settled, but they appear to involve differences in how readily glucose attaches to hemoglobin, not just differences in glucose levels themselves.

What this means practically: if you are Black and your A1c comes back at 6.5%, your actual average glucose might be somewhat lower than 139. The standard conversion could be overstating your glucose exposure. This does not mean A1c is useless, but it does mean the formula-based estimate should be treated as an approximation, not a precise measurement, especially when you are sitting right near a diagnostic cutoff.

Pregnancy Changes the Math

Pregnant women face a different version of the same problem. During the second trimester, blood volume expands and red blood cell turnover increases, which tends to push A1c downward even as actual blood sugar rises. Research from Massachusetts General Hospital found that by the late second trimester, average glucose measured during oral glucose tolerance testing was about 13 mg/dL higher than what the A1c-based estimate predicted. The discrepancy was minimal in the first trimester and postpartum but widened significantly in mid-pregnancy, especially in women with lower hemoglobin levels.

For a pregnant woman, relying on A1c alone to monitor glucose control could mask rising blood sugar levels during the critical period when gestational diabetes screening typically occurs. That is why most obstetric guidelines rely on direct glucose testing during pregnancy rather than A1c.

The Health Risks at This Level

An average glucose of 139 and an A1c near 6.5% place you at the threshold where diabetes-related complications begin to rise meaningfully. But cardiovascular risk does not suddenly appear at the diabetes cutoff. It scales upward across the whole range of blood sugar levels, even in people without diabetes. A meta-analysis of prospective studies found that people without diagnosed diabetes who had the highest blood glucose levels after a glucose challenge faced roughly a quarter to a third greater risk of cardiovascular disease compared with those who had the lowest levels. That elevated risk persisted even after adjusting for other heart disease risk factors like cholesterol and blood pressure.⁸

This is actually encouraging in a sense: if risk is a continuum, then any reduction in average glucose you achieve, even if it does not bring you all the way back to the normal range, still lowers your risk. You do not have to hit some magic number for the effort to count.

What You Can Do About It

If your average blood sugar is 139 and your A1c is around 6.5%, you are in the range where lifestyle changes can have a real impact and where early intervention pays off the most. The classic combination of dietary changes and regular exercise has been studied extensively in people at this stage. In a long-running Chinese trial involving people with impaired glucose tolerance, exercise alone reduced the risk of progressing to full diabetes by about 46%, while diet alone reduced it by about 31%. Combining the two reduced the risk by about 42%.⁹

Other trials have confirmed that combined diet and exercise interventions produce the greatest improvements in A1c, and that the benefits hold even without dramatic weight loss.¹⁰ That last point surprises many people. You do not need to lose 30 pounds for your glucose control to improve. Regular physical activity changes how your muscles take up glucose in ways that show up on your next A1c regardless of what the scale says.

Depending on your overall risk profile and how long your glucose has been elevated, your doctor may also consider medication. Metformin is the most commonly prescribed first-line drug at this stage, though the decision depends on factors beyond just the number.

Why Big Glucose Swings Matter Alongside the Average

One of the limitations of focusing purely on average glucose or A1c is that it tells you nothing about the spikes and dips throughout the day. Research increasingly suggests that glucose variability, meaning how much your blood sugar swings up and down, may matter for your health independently of where the average lands. Markers of oxidative stress and blood vessel damage appear to track more closely with glucose peaks and wide swings than with chronically elevated but stable glucose.¹¹

This is one reason continuous glucose monitors have become so popular, even among people with type 2 diabetes who are not on insulin. A monitor worn on your arm that reads glucose every few minutes can reveal patterns that fingerstick checks and quarterly A1c tests simply miss. You might learn that your average is 139 because you sit around 110 most of the day but spike to 220 after dinner, and that specific insight lets you target the problem with a focused change, like adjusting what you eat at that meal or taking a walk afterward.

Continuous Monitors and the GMI

If you wear a continuous glucose monitor, you will encounter a metric called the Glucose Management Indicator, or GMI. This is essentially a CGM-derived version of A1c: the device calculates your average sensor glucose and converts it into a percentage that looks and feels like an A1c result. The idea is to give you an ongoing estimate of your A1c without waiting for a lab draw.

In practice, though, GMI and lab-measured A1c frequently disagree. One clinic-based study found that lab A1c was on average about 0.34 percentage points higher than the CGM-derived GMI.¹² Researchers have been working to update the GMI formula to better align with lab results across a wider range of glucose levels.¹³ If your GMI says 6.3% but your lab A1c comes back at 6.7%, neither number is necessarily wrong. They are measuring related but slightly different things, and the biology connecting sensor glucose in your tissue fluid to hemoglobin glycation in your bloodstream introduces some natural mismatch.

For someone with an average glucose around 139, this discrepancy is worth keeping in mind. Your CGM app might show a GMI just under the diabetes threshold while your lab A1c sits just above it, or vice versa. Treat either number as an estimate, not a verdict.

When A1c Is Unreliable, Alternatives Exist

For people whose A1c cannot be trusted due to red blood cell disorders, kidney disease, recent blood loss, or other conditions that alter hemoglobin turnover, alternative blood tests can fill the gap. Fructosamine and glycated albumin both measure glucose attaching to blood proteins other than hemoglobin, and because those proteins turn over faster (on the order of two to three weeks rather than three months), they reflect a shorter window of glucose control.¹⁴

These tests are not used as widely as A1c, partly because the reference ranges are less well established and partly because fewer large outcome studies have linked them to long-term complications. But they can be especially useful for tracking how quickly glucose control is changing after a new medication or a dietary overhaul, since they respond faster than A1c. Another marker, 1,5-anhydroglucitol, drops when blood sugar spikes above the kidney’s reabsorption threshold, making it a rough indicator of how often you are experiencing postmeal glucose spikes.¹⁵

The Value of Understanding Your Own Number

There is an underappreciated psychological dimension to all of this. Research has shown that when people with poorly controlled diabetes who were unfamiliar with what A1c meant received education about the relationship between A1c and average glucose, their subsequent A1c levels dropped significantly. Interestingly, people who already understood the term did not see the same improvement, suggesting that the act of connecting the abstract lab number to a concrete daily glucose level may itself be motivating.

Cost Considerations With Glucose Monitoring

Continuous glucose monitors are not cheap, and access depends heavily on your insurance situation. But the economics are shifting. A microsimulation model looking at U.S. patients with type 2 diabetes on basal insulin found that CGM use actually reduced total healthcare costs over ten years compared with traditional fingerstick monitoring, despite the higher upfront cost of the devices. The savings came primarily from fewer diabetes complications, particularly kidney failure and heart attacks, because the better glucose data led to better management decisions.¹⁶ A separate modeling study from China reached similar conclusions, finding that CGM was cost-effective across a range of starting A1c levels, with greater benefit at higher baselines.¹⁷

For someone sitting at an average glucose of 139 and an A1c around 6.5%, a continuous monitor is not always covered by insurance, since many plans restrict coverage to people already on insulin or with more advanced diabetes. But the trend is toward broader coverage as the cost-effectiveness data accumulates. If you are in this glucose range and curious about your daily patterns, even a short trial with a monitor (some are now available by prescription for two-week cycles) can reveal information that quarterly lab tests never will.