HbA1c (haemoglobin A1c) is the cornerstone blood test for diagnosing and monitoring diabetes in the UK, yet there are well-recognised circumstances in which HbA1c results are not accurate. When red blood cell lifespan, structure, or turnover is altered — by conditions such as anaemia, haemoglobin variants, kidney disease, or pregnancy — the test can produce falsely high or falsely low readings that do not reflect true average blood glucose levels. Understanding why HbA1c may be unreliable, which alternative tests exist, and when to raise concerns with your GP is essential for safe, informed diabetes care.

Why HbA1c Results May Not Always Be Accurate

HbA1c is inaccurate when red blood cell lifespan or turnover is abnormal, causing results to be falsely elevated or falsely lowered regardless of true average blood glucose levels.

HbA1c (glycated haemoglobin) is a widely used blood test that reflects average blood glucose levels over the preceding two to three months. It works by measuring the proportion of haemoglobin — the oxygen-carrying protein in red blood cells — that has become chemically bound to glucose. Because red blood cells typically survive for around 90 to 120 days, the test provides a useful long-term snapshot of glycaemic control, making it central to both the diagnosis and monitoring of type 2 diabetes.

In the United Kingdom, HbA1c is reported in millimoles per mole (mmol/mol), in line with the International Federation of Clinical Chemistry (IFCC) standard. A result of 48 mmol/mol (6.5%) or above on two separate occasions is used to diagnose type 2 diabetes in appropriate clinical circumstances. A result of 42–47 mmol/mol (6.0–6.4%) indicates non-diabetic hyperglycaemia (sometimes called 'pre-diabetes'), which carries an increased risk of developing type 2 diabetes. These thresholds are set out in NICE guideline NG28.

However, HbA1c is not infallible. Its accuracy depends on red blood cells behaving in a predictable, consistent manner. When anything alters the lifespan, structure, or turnover rate of red blood cells, the HbA1c result may no longer accurately reflect true average blood glucose levels. In such cases, the result can be either falsely elevated (suggesting poorer glucose control than actually exists) or falsely lowered (masking genuinely poor control).

It is important to understand that a misleading HbA1c result does not necessarily indicate a laboratory error. Rather, it reflects a biological limitation of the test itself. Clinicians are trained to recognise circumstances in which HbA1c may be unreliable and to seek alternative or supplementary investigations accordingly. Patients who are aware of these limitations are better placed to have informed conversations with their healthcare team about their diabetes management and diagnostic results.

Medical Conditions That Can Affect HbA1c Readings

Haemolytic anaemia, blood transfusion, iron deficiency, haemoglobin variants (HbS, HbC, HbE), splenectomy, and chronic kidney disease are among the conditions that can significantly distort HbA1c readings.

A range of medical conditions can interfere with HbA1c accuracy, broadly by altering red blood cell lifespan, turnover rate, or haemoglobin structure:

Conditions that may cause falsely LOW HbA1c:

• Haemolytic anaemia — accelerated destruction of red blood cells means fewer older, more glycated cells are present, lowering the measured HbA1c

• Recent blood transfusion — donor red blood cells dilute the patient's own glycated haemoglobin

• Acute blood loss — rapid red cell replacement with younger cells reduces the proportion of glycated haemoglobin

• Erythropoietin (EPO) therapy — stimulates production of new red blood cells, lowering average cell age and HbA1c

• Haemoglobin variants such as sickle cell disease or haemoglobin C disease — structural differences can interfere with certain laboratory assay methods

• Pregnancy — increased red cell turnover and haemodilution can lower HbA1c, making it unreliable for monitoring; HbA1c is not used to diagnose gestational diabetes (NICE NG3)

• Advanced liver disease — shortened red cell survival can reduce HbA1c

• Certain medicines — dapsone and some antiretroviral agents can cause haemolysis, lowering HbA1c

Conditions that may cause falsely HIGH HbA1c:

• Iron deficiency anaemia — reduced red cell turnover leads to older cells accumulating more glycation, generally causing a falsely elevated HbA1c; this is the more consistently reported effect in the literature

• Vitamin B12 or folate deficiency — impaired red cell production prolongs cell lifespan, increasing glycation

• Chronic kidney disease (CKD) — anaemia associated with CKD, dialysis, and ESA therapy can lower HbA1c; however, in some patients, reduced red cell turnover may raise it. It is worth noting that most modern IFCC-aligned and NGSP-certified assays have minimal susceptibility to carbamylation interference from uraemia, though discordance between HbA1c and glucose readings remains common in CKD and warrants careful interpretation

• Splenectomy — removal of the spleen prolongs red cell survival, artificially elevating HbA1c

Haemoglobin variants deserve particular attention. The NHS Sickle Cell and Thalassaemia Screening Programme highlights that certain variants — including HbS, HbC, and HbE — are prevalent in specific ethnic populations and can significantly distort HbA1c measurements depending on the assay used. Some laboratory methods (such as boronate affinity chromatography) are less susceptible to variant interference than others. UK laboratories participating in UK NEQAS external quality assurance schemes are expected to flag results where variant interference is suspected. Clinicians should always consider a patient's ethnic background and haematological history when interpreting results, and check the local laboratory's method and any interference notes reported alongside the result.

What NHS Guidelines Say About HbA1c Limitations

NICE NG28 states HbA1c must not be used diagnostically in pregnancy, suspected type 1 diabetes, haemoglobin variants, recent blood transfusion, or when symptoms have been present for fewer than two months.

NICE guidance on the diagnosis and management of type 2 diabetes (NG28) and type 1 diabetes (NG17) acknowledges that HbA1c is not appropriate for all patients. NICE explicitly states that HbA1c should not be used as a diagnostic test in the following circumstances:

• Symptoms of diabetes present for fewer than two months

• Children and young people

• Pregnant women or those who may be pregnant

• Suspected type 1 diabetes — where clinical features suggest type 1, urgent same-day specialist assessment is required rather than reliance on HbA1c

• Patients with haemoglobin variants or conditions affecting red cell turnover

• Patients who have recently received a blood transfusion

• Acute pancreatic damage or pancreatectomy

• Steroid-induced or other acute-onset hyperglycaemia

The NHS also recognises that HbA1c can be affected by ethnicity-related haemoglobin variants. UK laboratories are expected to use IFCC-aligned, NGSP-certified methods and to report potential interference when variant haemoglobins are identified. The 2011 WHO-aligned UK guidance on using HbA1c for diagnosis, adopted by Public Health England, sets out these exclusions and the conditions under which plasma glucose testing should be used instead.

The Joint British Diabetes Societies (JBDS) similarly advise that clinicians exercise caution when interpreting HbA1c in the presence of any condition known to affect red cell biology. The key clinical principle is that no single test should be interpreted in isolation. Where HbA1c is unreliable or borderline, NICE recommends repeating the test or using alternative measures to confirm or exclude a diagnosis of diabetes. This approach helps prevent both under-diagnosis — which carries serious long-term health risks — and over-diagnosis, which can cause unnecessary anxiety and treatment burden.

Alternative Tests Used When HbA1c Is Unreliable

Validated alternatives include fasting plasma glucose (≥7.0 mmol/L diagnostic), random plasma glucose (≥11.1 mmol/L with symptoms), OGTT, fructosamine for short-term monitoring, and CGM for established diabetes.

When HbA1c is considered unreliable, clinicians have several validated alternatives available to assess glycaemic status:

Random plasma glucose In a patient with classic symptoms of diabetes (such as thirst, polyuria, unexplained weight loss, or fatigue), a random plasma glucose of 11.1 mmol/L or above is sufficient to diagnose diabetes without the need for a fasting or confirmatory test. This is an important diagnostic route when HbA1c cannot be relied upon.

Fasting plasma glucose (FPG) This measures blood glucose after an overnight fast of at least eight hours. A result of 7.0 mmol/L or above on two separate occasions (or once with symptoms) is diagnostic of diabetes. It is straightforward and widely available, though it only reflects glucose at a single point in time.

Oral glucose tolerance test (OGTT) The OGTT involves measuring blood glucose before and two hours after consuming a standardised 75 g glucose drink. A two-hour result of 11.1 mmol/L or above confirms diabetes. This test is particularly important in pregnancy: NICE NG3 sets out specific OGTT thresholds for diagnosing gestational diabetes, where HbA1c is not recommended as a diagnostic tool.

Fructosamine Fructosamine measures glycated serum proteins (primarily albumin) and reflects average glucose control over the preceding two to three weeks. It is used as a monitoring tool — not for diagnosis — in situations where HbA1c is unreliable, such as haemolytic anaemia or during pregnancy. However, it can be affected by conditions that alter serum protein levels, such as liver disease or nephrotic syndrome, and is not a NICE-endorsed diagnostic test.

Continuous glucose monitoring (CGM) For patients with established diabetes — particularly type 1 — CGM devices provide real-time and retrospective glucose data. Metrics such as time in range (TIR) are increasingly used alongside or instead of HbA1c to guide management, especially where HbA1c is known to be unreliable. NICE NG17 supports CGM use in adults with type 1 diabetes, and NICE NG28 sets out criteria for CGM and intermittently scanned CGM (isCGM) in type 2 diabetes.

The choice of alternative test depends on the clinical context, the reason HbA1c is unreliable, and local laboratory capabilities. A specialist or GP with a special interest in diabetes can advise on the most appropriate approach.

When to Speak to Your GP About HbA1c Concerns

Patients with haemoglobin variants, anaemia, CKD, or results inconsistent with home glucose readings should discuss HbA1c reliability with their GP, and anyone with symptoms of type 1 diabetes requires same-day urgent assessment.

Most patients will not need to question their HbA1c result routinely. However, there are specific circumstances in which it is entirely appropriate — and important — to raise concerns with your GP or diabetes care team.

Seek urgent same-day medical help if you have:

• Symptoms of marked hyperglycaemia — such as severe thirst, frequent urination, unexplained weight loss, vomiting, drowsiness, or confusion

• Ketones detected in your urine or blood

• Any concern that you or your child may have type 1 diabetes — this requires same-day specialist assessment and should not wait for routine HbA1c testing

If you are pregnant and concerned about blood glucose or gestational diabetes, contact your maternity or diabetes team promptly, as specialist assessment is needed.

Consider speaking to your GP if:

• You have been diagnosed with a haemoglobin variant (such as sickle cell trait, HbC, or thalassaemia) and have not been told how this might affect your diabetes monitoring

• You have anaemia (iron deficiency, B12 or folate deficiency, or haemolytic anaemia) and are unsure whether your HbA1c results are reliable

• You are pregnant or planning a pregnancy and have diabetes or are at risk of gestational diabetes

• You have chronic kidney disease and your HbA1c results seem inconsistent with your home glucose readings

• Your HbA1c result does not match your self-monitored blood glucose readings or how you feel day-to-day

• You have recently had a blood transfusion and are due a diabetes review

It is also worth raising concerns if you feel your diabetes management decisions — such as changes to medication — are being made solely on the basis of an HbA1c result that you believe may not be accurate for you. Shared decision-making is a core principle of NHS care, and patients have the right to ask questions and seek clarification.

Your GP may refer you to a diabetes specialist, a haematologist, or a specialist nurse depending on the underlying concern. Early discussion can prevent inappropriate treatment changes and ensure your monitoring plan is tailored to your individual circumstances.

How Clinicians Interpret HbA1c Alongside Other Results

Clinicians cross-reference HbA1c with full blood count, reticulocyte count, haemolysis markers, renal function, haemoglobin electrophoresis, and CGM data to identify glycaemic discordance and ensure accurate interpretation.

Experienced clinicians rarely rely on HbA1c in isolation. Instead, they interpret it within a broader clinical picture that includes the patient's symptoms, medical history, medication use, and other laboratory findings. This holistic approach is essential for accurate diagnosis and safe ongoing management.

When reviewing an HbA1c result, a clinician will typically consider:

• Full blood count (FBC) — to identify anaemia, abnormal red cell indices (such as mean corpuscular volume), or signs of haemolysis

• Reticulocyte count — elevated reticulocytes suggest increased red cell turnover, which can lower HbA1c

• Haemolysis markers — lactate dehydrogenase (LDH), haptoglobin, and bilirubin may indicate active haemolysis affecting HbA1c reliability

• Serum ferritin, B12, and folate — nutritional deficiencies that affect red cell lifespan and HbA1c reliability

• Renal function (eGFR and urea) — to assess for CKD-related factors, including anaemia and dialysis status

• Haemoglobin electrophoresis or HPLC — to identify haemoglobin variants in at-risk patients

• Medication history — including dapsone, antiretrovirals, EPO therapy, and other agents known to affect red cell lifespan

• Self-monitored blood glucose logs or CGM data — to cross-reference with the HbA1c result

• Laboratory method notes — clinicians are encouraged to review the assay method used and any interference flags reported by the laboratory, particularly where variant haemoglobins are possible

The concept of glycaemic discordance — where HbA1c and other glucose measures give conflicting information — is increasingly recognised in diabetes care. When discordance is identified, clinicians are guided to investigate the underlying cause rather than simply accepting one result over another.

Ultimately, HbA1c remains a valuable and well-validated tool when used appropriately. Its limitations do not diminish its clinical utility; rather, they underscore the importance of contextualised, patient-centred interpretation. Patients who understand these nuances are better equipped to engage meaningfully with their care team, ask the right questions, and contribute to decisions about their own health management.

Frequently Asked Questions