The HbA1c Test: What It Means and Why It Matters

UK readers: mg/dL values can be converted to mmol/L by dividing by 18. HbA1c percentages appear with mmol/mol equivalents where space allows. For UK-specific guidance, Diabetes UK and NICE are reliable sources.

Every person with diabetes hears about HbA1c, often before they fully understand what it measures. The test has shaped diabetes care for four decades. It drives medication decisions, insurance approvals, and clinical trial outcomes. Yet the number itself is frequently misread, misquoted, and sometimes simply wrong for biological reasons that have nothing to do with blood sugar control.

This guide walks through what HbA1c actually measures, what target ranges the major guidelines recommend, how often to test, when the result can't be trusted, how it compares to fingerstick and CGM data, and the fastest evidence-based ways to lower it. If you want to know your numbers properly, start here.

What Does HbA1c Actually Measure?

HbA1c measures the percentage of hemoglobin molecules in your red blood cells that have glucose permanently attached to them. Because red cells live about 120 days, this percentage mirrors your average blood glucose over the previous 2 to 3 months, weighted toward the most recent 4 weeks. (NIDDK, 2024)^[1]

The Biology of Glycation

Glucose doesn't just float around in blood. It slowly binds to proteins in a non-enzymatic process called glycation. When glucose attaches to the beta chain of hemoglobin A, it forms glycated hemoglobin, or HbA1c. The reaction is irreversible for the life of the red cell. The higher your average glucose, the more hemoglobin gets glycated. This ratio is what the lab reports.

Because the binding is proportional to ambient glucose concentration, HbA1c gives a stable, integrated picture that a single fingerstick simply can't. A fasting glucose of 110 mg/dL tells you about one moment. An HbA1c of 7.2% tells you about 8 to 12 weeks.

How Labs Standardize the Result

Every HbA1c test worldwide is now aligned to a single reference: the NGSP, which traces back to the DCCT landmark study of 1993. (NGSP, 2024)^[4] The DCCT proved that lowering HbA1c dramatically reduces retinopathy, nephropathy, and neuropathy risk in type 1 diabetes. That trial's assay became the global reference standard.

In Europe and several other regions, labs also report HbA1c in mmol/mol using the IFCC reference method. A value of 7.0% (53 mmol/mol) in NGSP units equals 53 mmol/mol IFCC. Both are correct. Patients moving between healthcare systems should know which unit their lab uses.

What Are the HbA1c Target Ranges?

The ADA sets the most widely used diagnostic cutoffs: less than 5.7% is normal, 5.7 to 6.4% is prediabetes, and 6.5% or higher on two separate tests confirms diabetes. (ADA Standards of Care, 2025)^[2] Treatment targets for people already diagnosed vary by individual factors like age, comorbidities, and hypoglycemia risk.

ADA vs AACE vs Individualized Targets

The ADA recommends a general treatment target of less than 7.0% for most adults with diabetes. (ADA, 2025)^[3] The American Association of Clinical Endocrinology (AACE) suggests a tighter target of 6.5% or less for many patients, provided it can be achieved without significant hypoglycemia. (Blonde et al., 2022)^[6]

Individualized targets are the real-world standard. An otherwise healthy 45-year-old with type 2 diabetes and low hypoglycemia risk might aim for 6.5%. A frail 82-year-old with cardiovascular disease and recurrent lows might safely sit at 8.0% (64 mmol/mol). Both are appropriate. The VADT and ACCORD trials showed that aggressive lowering in older patients with established disease can actually increase mortality. (ACCORD, 2008)^[7]

The Clinical Meaning of Each Point

Every 1% drop in HbA1c reduces microvascular complications by approximately 37% and diabetes-related deaths by 21%, per the UKPDS 35 landmark analysis. (Stratton et al., 2000)^[5] The DCCT showed similar benefits in type 1 diabetes: a drop from 9% to 7% cut retinopathy progression by 76%. (DCCT, 1993)^[8] These numbers are why clinicians push for lower readings, even when the drop seems small on paper.

How Does HbA1c Compare to Fingerstick and CGM Data?

HbA1c reflects 120 days of average glucose, while fingerstick readings capture a single moment and continuous glucose monitors (CGM) stream data every 1 to 5 minutes. The 10-week ADAG study derived the conversion formula eAG (mg/dL) = 28.7 x A1C - 46.7, establishing the link between the lab number and the readings you see daily. (Nathan et al., 2008)^[9]

HbA1c vs eAG

Estimated average glucose (eAG) converts the HbA1c percentage into mg/dL or mmol/L, the same units used by your meter. An HbA1c of 7.0% equals an eAG of 154 mg/dL. The number becomes tangible when it sits in the same units as daily readings. Many labs now report both values on the same sheet.

eAG has limits. It's an average, not a range. Two people can share an eAG of 154 mg/dL with very different glucose patterns. One may swing between 70 and 250 mg/dL hourly; the other may stay between 130 and 180 mg/dL consistently. Both have the same HbA1c. Their risk profiles are different.

Time-in-Range: The CGM Revolution

Continuous glucose monitors exposed a blind spot in HbA1c. Time-in-range (TIR), defined as the percentage of readings between 70 and 180 mg/dL, captures variability that a single averaged number hides. International consensus guidelines recommend TIR above 70% for most adults with type 1 or type 2 diabetes. (Beck et al., 2019)^[15]

A 10% increase in TIR roughly corresponds to a 0.5% drop in HbA1c. TIR above 70% typically aligns with HbA1c below 7%. But two patients with identical HbA1c can have very different TIR, and the one with more variability and more time in hypoglycemia faces different risks. (Battelino et al., 2019)^[16]

Glycemic Variability and the Glucose Management Indicator

CGM reports now include a Glucose Management Indicator (GMI), which estimates HbA1c from 14 or more days of continuous data. GMI and lab HbA1c can differ by up to 0.5% in individuals, usually due to red cell biology rather than measurement error. (Bergenstal et al., 2018)^[17] If your GMI and lab HbA1c don't line up, that's a clue worth investigating, not a contradiction to ignore.

How Often Should You Test Your HbA1c?

The ADA recommends testing HbA1c at least twice a year in people who are meeting treatment goals and have stable glycemic control, and every 3 months when therapy has changed or goals aren't being met. (ADA Standards of Care, 2025)^[3] Testing more often than quarterly rarely yields useful information because red cell turnover limits how fast the number can move.

The 120-Day Rule

Red blood cells live roughly 120 days. After any change in glucose control, it takes about 6 to 8 weeks before HbA1c starts to reflect the change meaningfully, and about 3 months before it stabilizes at the new level. Testing at 4-week intervals produces noise, not signal. That's why insurance plans typically cover HbA1c every 3 months, not every month.

Who Needs More Frequent Testing?

Three groups benefit from quarterly or more frequent testing: people starting a new medication, people with recent lifestyle changes like significant weight loss or new exercise, and pregnant women with pre-existing diabetes. In pregnancy, target HbA1c is below 6.0% (42 mmol/mol) where achievable without hypoglycemia, and monthly monitoring is often appropriate. (ADA Pregnancy, 2025)^[10]

When Does HbA1c Lie? Limitations and Interferences

HbA1c assumes normal red blood cell turnover. Any condition that changes red cell lifespan distorts the result. Iron-deficiency anemia can falsely raise HbA1c by up to 1.5%, while hemolytic anemia and recent blood loss can falsely lower it by similar margins. (Kim et al., 2010)^[11] These are not lab errors. The chemistry is correct. The interpretation isn't.

Anemia and Iron Deficiency

In iron-deficiency anemia, red cells live longer than 120 days because the bone marrow produces fewer replacements. Older red cells accumulate more glucose, so the glycation percentage rises even when blood glucose hasn't changed. Correcting the iron deficiency typically drops HbA1c by 0.5 to 1.5% without any change in diabetes management. (Coban et al., 2004)^[12]

The opposite happens in hemolytic anemia, chronic kidney disease with dialysis, and after recent bleeding or transfusion. Shorter red cell lifespan means less time for glycation, so HbA1c reads low even when glucose has been high. Fructosamine or continuous glucose monitoring give a more accurate picture in these patients.

Hemoglobinopathies

Sickle cell trait, sickle cell disease, hemoglobin C, E, and D variants all affect HbA1c assays to varying degrees. The NGSP maintains a list of which assay methods are accurate with which variants. (NGSP Interferences, 2024)^[13] A 2017 CDC analysis noted that sickle cell trait is present in roughly 1 in 13 Black Americans, making method selection clinically important in this population. (CDC, 2024)^[14]

Pregnancy and Other Red Cell Changes

During pregnancy, blood volume expands and red cell turnover accelerates in the second and third trimesters. This naturally lowers HbA1c by 0.3 to 0.5%. An HbA1c of 5.6% in a pregnant woman might reflect the same average glucose as 6.0% in a non-pregnant woman. Pregnancy-specific targets and frequent glucose self-monitoring are essential.^[10]

Other conditions to flag: chronic liver disease, severe hypertriglyceridemia, chronic aspirin therapy at high doses, and certain vitamin supplements. If your HbA1c doesn't match your fingerstick or CGM data, one of these may be the reason.

How Do You Lower Your HbA1c?

HbA1c responds to the same tools that lower daily glucose, just on a delayed timeline. A Cochrane meta-analysis found that structured exercise alone lowered HbA1c by an average 0.67% in people with type 2 diabetes, comparable to many oral medications. (Thomas et al., 2006)^[21] Weight loss, diet change, and medication adherence each contribute measurable reductions.

Exercise and Physical Activity

Both aerobic and resistance training lower HbA1c. A 2011 meta-analysis in JAMA found combined training reduced HbA1c by 0.66% over 12 weeks, while aerobic-only reduced it by 0.51% and resistance-only by 0.57%. (Umpierre et al., 2011)^[22] The ADA recommends 150 minutes of moderate-intensity aerobic activity per week plus 2-3 resistance sessions. Our piece on exercise and insulin sensitivity digs deeper into the mechanism.

Dietary Patterns

Low-carbohydrate and Mediterranean dietary patterns both lower HbA1c. A 2021 BMJ meta-analysis found low-carb diets reduced HbA1c by 0.47% at 6 months compared to control diets. (Goldenberg et al., 2021)^[23] The Mediterranean pattern produced a 0.30 to 0.47% reduction across multiple trials. (Esposito et al., 2009)^[24] See our guide on smart nutrition choices for practical meal strategies.

Carbohydrate quality matters as much as quantity. Our article on carbohydrates and blood sugar covers glycemic index, fiber, and portion strategies in detail. For most people, cutting sugar-sweetened beverages alone can drop HbA1c by 0.3% within 3 months.

Weight Loss

A sustained weight loss of 5 to 10% of body weight typically reduces HbA1c by 0.5 to 1.0% in people with type 2 diabetes. (Wing et al., 2011)^[25] The DiRECT trial showed that intensive weight management achieved diabetes remission in 46% of participants who lost at least 10 kg, with most remaining medication-free at 2 years. (Lean et al., 2018)^[26]

Medication Optimization

Metformin remains first-line for type 2 diabetes and lowers HbA1c by roughly 1.0 to 1.5% at full dose. GLP-1 receptor agonists and SGLT2 inhibitors now offer similar or greater HbA1c reductions plus cardiovascular and renal benefits. (Davies et al., 2022)^[27] Adherence is often the limiting factor. If your HbA1c isn't budging, the first question isn't "what new drug?" but "are you taking what's already prescribed?"

Sleep, Stress, and the Neglected Basics

Chronic poor sleep raises HbA1c. A 2015 meta-analysis found that short sleep duration (less than 6 hours) was associated with a 0.35% higher HbA1c compared to 7-8 hours. (Lee et al., 2017)^[28] Chronic stress elevates cortisol and drives hepatic glucose output. Mindfulness-based interventions have shown HbA1c reductions of 0.3 to 0.5% in some trials. (Noordali et al., 2017)^[29]

"Intensive blood-glucose control substantially reduces the risk of microvascular complications in patients with type 2 diabetes, but not incidence of macrovascular disease. The relationship is continuous: there is no threshold below which further lowering stops helping." - Adapted from the UKPDS 35 findings, Stratton et al., BMJ, 2000^[5]

Frequently Asked Questions