A1C testing traces its origins to 1969, when Iranian-born researcher Samuel Rahbar identified unusually high levels of a fast-moving hemoglobin fraction in patients with diabetes. That laboratory observation, made using a technique called chromatography, set off a chain of developments that would take roughly a decade to reach routine clinical use and another three decades to become a formal diagnostic criterion for diabetes itself. The journey from obscure biochemical curiosity to one of the most widely ordered blood tests in the world is a story of slow scientific persuasion, messy measurement problems, and landmark clinical trials.
The Accidental Discovery
In the late 1960s, hemoglobin variants were a hot research area, driven mainly by interest in conditions like sickle cell disease. Samuel Rahbar, working at the University of Tehran, was using electrophoresis to separate hemoglobin fractions in blood samples. He noticed that samples from diabetic patients consistently showed an elevated band of hemoglobin that migrated faster than normal adult hemoglobin. By 1969, Rahbar and colleagues had published their finding that patients with diabetes had higher concentrations of these “fast hemoglobin” fractions compared to people without diabetes.1Diabetes Research and Clinical Practice. Haemoglobin A1c: Historical overview and current concepts At the time, nobody fully understood what this modified hemoglobin was or why it accumulated. But the observation planted a seed: something in the blood of diabetic patients was chemically altering their hemoglobin, and the degree of alteration seemed to track with how poorly their blood sugar was controlled.
How Glucose Modifies Hemoglobin
The mechanism turned out to be straightforward in principle. Glucose floating in the bloodstream gradually attaches itself to hemoglobin molecules inside red blood cells. Specifically, it binds to one or both ends of the beta chains of normal adult hemoglobin, forming a compound through a nonenzymatic reaction.2Laboratory Medicine. Glucose and Hemoglobin A1c The process begins with a reversible attachment, producing what’s called “labile” A1C. Over time, this labile form undergoes a slower, largely irreversible rearrangement into stable A1C.3PubMed Central. Insights into the Progression of Labile Hb A(1c) to Stable Hb A(1c) via a Mechanistic Assessment of 2,3-Bisphosphoglycerate Facilitation of the Slow Nonenzymatic Glycation Process Because red blood cells live for about 120 days, the amount of stable A1C in a blood sample reflects the average glucose exposure over the previous two to three months. That built-in time window is what makes the test so useful: it captures a long-term trend rather than a snapshot of what your blood sugar happened to be that morning.
For decades, the underlying model was simple: higher average blood sugar means more glucose sticking to hemoglobin, period. Recent work has started to question whether the relationship is quite that clean, pointing out that glucose transport into red blood cells may not be entirely passive and that intracellular glucose metabolism could play a role.4PubMed. HbA1c reconsidered: glycated haemoglobin in the era of dynamic erythrocyte glucose metabolism For practical purposes, though, the basic correlation between average blood sugar and A1C remains the foundation of how doctors use the test.
Entering the Clinical Lab
The gap between Rahbar’s 1969 discovery and routine clinical use was about eight years. Glycated hemoglobin, initially measured as total HbA1, first appeared in routine clinical laboratories around 1977.5PubMed Central. HbA1c standardisation: history, science and politics The early methods were crude by modern standards. Labs used various separation techniques, each exploiting a slightly different property of the glycated hemoglobin molecule. Some relied on the electrical charge difference between glycated and non-glycated hemoglobin. Others used affinity-based approaches that detected the sugar molecule itself. This diversity created a serious practical problem: two labs running the same patient sample could return meaningfully different numbers.
Throughout the 1980s and into the 1990s, the test gained popularity despite the measurement inconsistencies. Clinicians could see that patients whose A1C was trending down tended to develop fewer complications, even before rigorous trial data confirmed it. But the lack of uniformity across methods meant that an A1C of 7% from one laboratory might not mean the same thing as 7% from another. A patient who moved across town and switched labs could see their A1C “change” without any real change in blood sugar control.6PubMed Central. A review of the challenge in measuring hemoglobin A1c
The Trials That Proved It Mattered
The test’s clinical credibility was cemented by two landmark studies. The Diabetes Control and Complications Trial (DCCT), which enrolled people with type 1 diabetes, and the UK Prospective Diabetes Study (UKPDS), focused on type 2 diabetes, both demonstrated that lowering A1C reduced the risk of microvascular complications like eye disease, kidney damage, and nerve damage. In the UKPDS, the intensively treated group maintained a median A1C of about 7.0% compared with 7.9% in the conventionally treated group, a difference of roughly one percentage point sustained over ten years, and this gap corresponded to meaningfully fewer complications.7PubMed. Haemoglobin A1c–a marker for complications of type 2 diabetes: the experience from the UK Prospective Diabetes Study (UKPDS)
These trials did more than validate A1C as a useful marker. They gave the medical community concrete targets. The idea that getting a patient’s A1C below 7% (or below some other threshold, depending on the guideline and the patient) could prevent blindness, kidney failure, and amputations transformed A1C from a monitoring tool into a treatment goal. Health systems started building quality metrics around it. Insurance companies used it to evaluate physician performance. For better or worse, A1C became the single number most associated with diabetes management.
Solving the Standardization Mess
The measurement inconsistency problem that plagued early A1C testing demanded a systematic fix. In the United States, the National Glycohemoglobin Standardization Program (NGSP) was established to bring order to the chaos. The NGSP worked with manufacturers to calibrate their instruments against a common reference, and introduced proficiency testing using whole blood samples so that individual labs could be assessed for accuracy.8Clinical Chemistry. The National Glycohemoglobin Standardization Program: Over 20 Years of Improving Hemoglobin A1c Measurement Japan and Sweden pursued their own standardization efforts in parallel.
Internationally, the International Federation of Clinical Chemistry (IFCC) developed a reference method that could serve as a gold standard across countries. Over time, the combined efforts of the NGSP and the IFCC network brought A1C measurement into much tighter agreement worldwide. The improvement was dramatic: labs that once produced results differing by a full percentage point or more now routinely agree within a fraction of a point. Without this standardization work, A1C could never have become a diagnostic criterion, because you cannot diagnose a disease based on a number that varies depending on which lab runs the test.
From Monitoring Tool to Diagnostic Criterion
For decades, diabetes was diagnosed solely by measuring blood glucose. Patients had to fast overnight for a fasting glucose test, or sit through a two-hour oral glucose tolerance test. Both approaches capture a single moment, and both are sensitive to what the patient ate, how stressed they were, or whether they happened to be fighting an infection. A1C, reflecting months of glucose exposure, seemed like a more stable alternative. In 2010, the American Diabetes Association formally added A1C to its diagnostic criteria, recommending that a value of 6.5% or higher be used to diagnose diabetes.9PubMed Central. Implications of using hemoglobin A1C for diagnosing diabetes mellitus This was a watershed moment. The test that had spent 40 years as a monitoring tool for people who already had diabetes was now being used to identify the disease in the first place. No fasting required, no glucose drink, just a routine blood draw at any time of day.
When A1C Gets It Wrong
A1C is not infallible, and its limitations are important to understand. The test assumes that your red blood cells behave normally and live for roughly the standard 120 days. Anything that changes red blood cell lifespan or hemoglobin structure can throw off the result.
Hemoglobin variants are among the most studied confounders. The most common variants worldwide are HbS (associated with sickle cell trait and disease), HbC, HbD, and HbE. Each involves a single amino acid change in the hemoglobin beta chain, and the effect on A1C measurement depends on which variant is present and which lab method is used.10PubMed Central. A review of variant hemoglobins interfering with hemoglobin A1c measurement Modern testing systems have improved substantially. Across five widely used platforms, the common variants HbS, HbC, HbD, and HbE no longer cause clinically meaningful interference.11Laboratory Medicine. Interference of hemoglobin variants with HbA1c measurements by six commonly used HbA1c methods But not all systems handle them equally well, and rarer variants like Hb Okayama and Hb Lepore can still produce misleading results.12Clinica Chimica Acta. Analytical interference of 33 different hemoglobin variants on HbA1c measurements comparing high-performance liquid chromatography with whole blood enzymatic assay Elevated fetal hemoglobin, seen in conditions like thalassemia, also tends to push A1C readings downward regardless of the method used.
Kidney disease is another well-known confounder. In people with chronic kidney disease, particularly those who are also anemic, the correlation between A1C and actual blood sugar weakens considerably. Research has shown the correlation dropping from around 0.70 in people without kidney disease or anemia to about 0.35 in those with both severe kidney disease and anemia.13PubMed. How does CKD affect HbA1c? Anemia in general, whatever its cause, tends to shorten red blood cell lifespan, giving glucose less time to accumulate on hemoglobin and potentially producing a falsely low A1C. Iron deficiency without anemia can push the result in the opposite direction, though the relationship is complex and weakened in people who also have kidney problems.14Diabetes Care. Association Between Iron Deficiency and A1C Levels Among Adults Without Diabetes in the National Health and Nutrition Examination Survey, 1999–2006
Racial and Ethnic Differences in A1C
One of the more uncomfortable realities about A1C is that the same average blood sugar produces different A1C values depending on a person’s racial or ethnic background. In a study of people with impaired glucose tolerance (not yet diabetic), adjusted A1C levels were about 5.78% for white participants, 5.93% for Hispanic participants, 6.00% for Asian participants, 6.12% for American Indian participants, and 6.18% for Black participants, even after accounting for differences in glucose levels, body weight, blood pressure, and insulin resistance.15PubMed Central. Differences in A1C by Race and Ethnicity Among Patients With Impaired Glucose Tolerance in the Diabetes Prevention Program
Research using continuous glucose monitoring to pin down actual average blood sugar has confirmed the gap. For a given A1C level, mean glucose concentrations were lower in Black participants than in white participants, with A1C running about 0.4 percentage points higher in Black individuals for the same mean glucose.16PubMed. Racial Differences in the Relationship of Glucose Concentrations and Hemoglobin A1c Levels Among people with type 2 diabetes, these discrepancies persisted across multiple racial and ethnic groups even when estimated mean plasma glucose was similar.17The Journal of Clinical Endocrinology & Metabolism. Racial and Ethnic Differences in Mean Plasma Glucose, Hemoglobin A1c, and 1,5-Anhydroglucitol in Over 2000 Patients with Type 2 Diabetes
The clinical implications are real. A Black patient with an A1C of 6.5% may have lower actual average blood sugar than a white patient with the same number. Using a single A1C cutoff for diagnosis could mean that some populations get diagnosed earlier, or treated more aggressively, than their glucose levels warrant. This issue has fueled ongoing debate about whether diagnostic thresholds should be adjusted by race, whether alternative markers should be used alongside A1C, or whether the current one-size-fits-all approach is good enough. No consensus has been reached.
Point-of-Care Testing
For most of its history, A1C required a blood sample sent to a central laboratory, with results returning hours or days later. Point-of-care devices, which can return results from a fingerstick in minutes right in the exam room, have changed the workflow considerably. Evaluation of one such device showed that fingerstick results correlated almost perfectly with laboratory reference values, with minimal bias.18PubMed Central. Accuracy and Precision of a Point-of-Care HbA1c Test
The appeal is obvious: a patient walks in, gets their A1C, and discusses the result with their doctor in the same visit. No waiting, no follow-up phone call, no delayed treatment decisions. Cost-effectiveness analysis has suggested that point-of-care A1C testing in primary care produces better outcomes than traditional lab testing, though at modestly higher cost per patient.19Frontiers in Pharmacology. Cost-Effectiveness of Point-of-Care A1C Tests in a Primary Care Setting For patients who struggle with follow-up appointments or live far from a lab, having the test done on the spot can mean the difference between adjusting treatment promptly and losing months to inertia.
Continuous Glucose Monitoring and Time in Range
The newest challenge to A1C’s dominance comes from continuous glucose monitors, wearable sensors that track blood sugar every few minutes around the clock. These devices produce a metric called “time in range,” the percentage of the day a person’s glucose stays within a target zone, usually 70 to 180 mg/dL. Time in range captures information that A1C cannot: it reveals how often glucose spikes dangerously high or drops dangerously low, patterns that get averaged away in a single A1C number.
Real-world studies in people with type 1 diabetes have confirmed that time in range correlates with A1C, though the relationship is not perfect and varies among individuals.20BMJ Open Diabetes Research & Care. Time in range–A1c hemoglobin relationship in continuous glucose monitoring of type 1 diabetes: a real-world study Two patients with identical A1C values could have very different time-in-range profiles, one stable and the other swinging between highs and lows. For this reason, many diabetes specialists now use time in range alongside A1C rather than relying on A1C alone. Whether continuous glucose monitoring metrics will eventually replace A1C as the primary standard remains an open question, but the direction of travel is clear: the field is moving toward richer, more granular measures of glucose control.
Alternatives for When A1C Falls Short
For patients in whom A1C is unreliable, clinicians have other options. Fructosamine and glycated albumin both reflect glycemic control but over a shorter window, roughly two to three weeks, because they measure glucose attachment to serum proteins rather than hemoglobin.21PubMed Central. Advantages and pitfalls of fructosamine and glycated albumin in the diagnosis and treatment of diabetes These markers are particularly useful in patients with red blood cell disorders, kidney disease, recent blood transfusions, or rapid changes in glucose control where A1C might lag behind reality.22PubMed Central. Alternative biomarkers for assessing glycemic control in diabetes: fructosamine, glycated albumin, and 1,5-anhydroglucitol Another marker, 1,5-anhydroglucitol, works differently: it drops when glucose is high and recovers when glucose normalizes, making it especially sensitive to recent glucose spikes. None of these alternatives has the decades of outcomes data backing A1C, which is the main reason A1C remains the default.
How Patients Understand Their A1C
Despite the test’s central role in diabetes care, patients and clinicians do not always share the same understanding of what the number means. In qualitative research exploring how people with diabetes assess their own control, most participants reported being familiar with A1C, but their understanding of what the value actually represented varied widely.23PubMed Central. Exploring how patients understand and assess their diabetes control Some grasped that it reflected long-term average blood sugar. Others confused it with their most recent glucose reading or treated it as a vague indicator of whether they were “doing well.” Patients also relied on other cues to judge their control, including how consistently they took their medications, whether they felt symptoms they attributed to diabetes, and what their home glucose meter showed. The A1C result was one piece of a larger self-assessment puzzle, not the definitive answer that clinicians sometimes treat it as.
A1C Testing in Veterinary Medicine
The reach of A1C testing extends beyond human medicine. Dogs and cats develop diabetes too, and veterinarians have been exploring whether human A1C assays can work in other species. In dogs, a commercially available human A1C analyzer has been validated for use, with diabetic dogs showing significantly higher A1C values than non-diabetic dogs and a cutoff above 6.2% suggesting canine diabetes.24PubMed Central. Evaluation of a human glycated hemoglobin test in canine diabetes mellitus Further research has shown that A1C can identify at-risk dogs, including overweight and elderly animals, whose values run higher than healthy controls even before overt diabetes develops.25PubMed Central. The role of HbA1c in identifying dogs at high risk for diabetes despite normal blood glucose levels
Cats present a different challenge. Their red blood cells have a shorter lifespan than those of dogs or humans, and anemia is common in sick cats. When anemia is present, it can artificially push A1C upward, the opposite of the usual direction in humans. A weak negative correlation between A1C values and packed cell volume has been observed in anemic cats.26PubMed. Establishment of a feline glycated hemoglobin reference interval for a novel dried-blood-spot assay and the effects of anemia on assay results Fructosamine remains the more commonly used long-term glucose marker in feline practice for exactly this reason. The fact that a test originally discovered through human hemoglobin research now has applications in companion animals speaks to how fundamental the underlying chemistry is: wherever there is hemoglobin and glucose, glycation happens.