HbA1c (blood sugar)

HbA1c reflects your average blood sugar over recent months. This score estimates your genetic tendency toward higher or lower long-term blood sugar.

What this HbA1c score reads

This is PGS000304, used by Xie and colleagues in a 2020 study of genetic risk scores for complex traits in young people. It sums 43 variants associated with HbA1c at genome-wide significance, from GWAS data on 121,963 people, about 72% of European ancestry.

HbA1c is haemoglobin with sugar attached, and it is read as a summary of average blood glucose over the preceding two to three months. That framing hides a subtlety this score depends on.

How well it predicts HbA1c

In the study's European-ancestry validation samples, the score explained about 2.8% of the variation in HbA1c. That is a small fraction, and it means two people with opposite percentiles will usually have HbA1c values that overlap completely.

Why a genetic HbA1c tendency is not the same as blood sugar

HbA1c depends on two things: how much glucose is in the blood, and how long red blood cells live to accumulate it. Variants that shorten or lengthen red cell survival, or that change haemoglobin itself, shift HbA1c without shifting glucose at all.

This score contains variants of both kinds. One of them is the HFE C282Y variant behind hereditary hemochromatosis, which this site covers on its own page. So a high percentile here can reflect red cell and iron biology rather than a tendency toward high blood sugar, and that distinction matters if an HbA1c test is ever used to diagnose diabetes in you.

What this score does not capture

Sickle cell trait and thalassaemia traits, common in some populations, can make HbA1c readings unreliable, and are not read here. Neither is anaemia, recent blood loss, pregnancy or kidney disease, all of which distort HbA1c. If your HbA1c and your glucose measurements disagree, that conflict is a clinical question, not a genetic one.

Ancestry and accuracy

The underlying GWAS was about 72% European ancestry, and the validation was in European-ancestry youth only. Nothing here has been shown to hold in other populations, which is a particular limitation for a measure whose interpretation already varies with haemoglobin variants that differ in frequency between populations.

Variants in this score you can read on their own

This score includes variants that this site also explains individually, where you can see what each genotype means on its own: Fasting glucose / type 2 diabetes (MTNR1B) (rs10830963) · Hereditary hemochromatosis (HFE gene) (rs1800562).

How much of this score your file covers

42 of the published score's 43 variants are present in a standard consumer DNA file (98%). The remaining one cannot be read from one, so a result computed here approximates the published score rather than reproducing it.

Common questions

Does a high HbA1c score mean I am heading for diabetes? No. The score explains under 3% of HbA1c variation, and part of what it captures is red cell biology rather than blood sugar. A fasting glucose or HbA1c measurement answers the question this score only gestures at.

Why is the hemochromatosis variant in a blood sugar score? Because HbA1c is a property of haemoglobin inside red blood cells, variants affecting iron handling and red cells can shift the measurement without changing glucose. That is a known feature of HbA1c genetics, and it is one reason a genetic HbA1c tendency should not be read as a glucose tendency.

Related

Autoimmune protection (IL23R R381Q) · Body mass (BMI) tendency · Colorectal cancer risk (8q24) · Depression · Longevity-associated variant (FOXO3) · Rheumatoid arthritis · Venous thromboembolism (clots)