An HbA1c That Has Not Fallen on Treatment
Before anything else, check when the repeat was taken. HbA1c reports roughly three months and weights the most recent weeks most heavily, so a test at six weeks is still half-reading the level from before treatment started. A result that looks like failure at six weeks can be a good response measured too early — and that single fact accounts for a large share of unnecessary dose changes.
The pattern on your report
- HbA1c High · moderate Key
- Fasting glucose High Key
- eGFR Normal Key
- Hemoglobin Normal Key
Printed as: eGFR in mL/min/1.73m2— Determines which treatments are appropriate and at what dose.Fasting glucose in mmol/Lor mg/dLHemoglobin in g/Lor g/dL— Anemia distorts the HbA1c, so a low value changes how much the result can be trusted.HbA1c in mmol/molor %— Reports roughly three months, weighted toward the recent past, so a repeat needs enough time to have passed to mean anything.
Why the numbers look like this
Sugar attaches to hemoglobin gradually and irreversibly over a red cell's life. Because the circulation always holds cells of every age, the value at any moment is an average across all of them, weighted toward the recent past because more cells are alive to have recorded it.
That weighting is what sets the timescale. Changing glucose control today starts changing the number immediately, but the full effect is not visible until the cells that lived through the old conditions have been replaced.
So a repeat needs enough time to have passed. Testing earlier gives a value part way through the transition, and reading it as an endpoint is a measurement error, not a treatment failure.
Where enough time has passed and nothing has moved, the explanations divide cleanly. Either the treatment is not being taken or not reaching an effective dose, or something is actively working against it, or the number itself is unreliable for one of the reasons that distort it.
Not being flagged is not the same as normal
Different laboratories report in different units and some report both, which means a value that looks unchanged may not have been compared like with like. Repeating at the same laboratory makes a small difference genuinely interpretable. The size of change that matters also depends on the starting point: a fall from a very high value is expected to be larger than the same treatment produces from a modestly raised one, so a proportional view is more informative than an absolute one.
What else on the report can hide this
Establish the interval since treatment started or changed before anything else follows from the result.
Then raise the tablets themselves, and raise them so that saying 'I stopped' carries no penalty. Not taking them is common, under-reported, and entirely understandable when side effects are involved.
A full blood count and iron studies matter because anemia and iron deficiency both distort the value, and iron deficiency in particular pushes it up, which can look exactly like treatment failing.
A medication review covers the drugs working against control, with steroids the most important. Weight change over the same period explains a great deal on its own.
Where the response is genuinely inadequate on an adequate dose, the answer is usually to add a second agent rather than to keep pushing the first. Modern options do more than lower the number: several also reduce cardiovascular and kidney risk independently, which matters more than the HbA1c itself.
What usually causes it
Listed from most to least common — not from most to least serious.
- Very common
Retested too soon
A test at six weeks is still substantially reading the old level. This accounts for a large share of apparent failures and of the dose changes that follow.
- Very common
Not taking the treatment
Common, under-reported, and often driven by side effects. The answer depends entirely on how the question is asked.
- Very common
The dose is not yet effective
Many treatments are started low and need escalating. A partial fall that stalls fits this rather than resistance.
- Common
Weight gain over the same period
Works directly against the treatment. Recording weight alongside the result explains a great deal without any test.
- Common
Steroid treatment
Raises glucose substantially, predominantly after meals, and courses are often not mentioned as medication.
- Common
Iron deficiency
Lengthens red cell survival and pushes HbA1c up, which imitates failing control precisely.
- Common
Another drug working against control
Some antipsychotics, thiazides and immunosuppressants among others. A medication review identifies them.
- Uncommon
Progressive loss of insulin production
Control that was stable for years drifting despite good adherence. It changes what treatment is appropriate rather than how hard to push the current one.
- Uncommon
Latent autoimmune diabetes in adults
Suspected in a leaner adult whose control deteriorates faster than expected on tablets. Antibody testing distinguishes it.
What is usually checked next
- The interval since treatment started or changed The number needs roughly three months to reflect the new level, and testing early looks like failure.
- An open conversation about how the treatment is taken The commonest genuine reason, and side effects are usually behind it.
- Full blood count and iron studies Iron deficiency pushes the value up and imitates failing control exactly.
- A medication review and a weight comparison Steroids and weight gain between them explain most of the rest, and neither needs a test.
- Consider adding a second agent rather than escalating further Several modern options reduce cardiovascular and kidney risk independently of the number.
When to seek care sooner
- Emergency Vomiting with abdominal pain and rapid breathing
- Emergency Confusion or drowsiness
- Same day Excessive thirst with passing large volumes of urine and weight loss
- Same day Blurred vision that has changed quickly
- Same day A foot ulcer, or a wound that is not healing
- Soon Control deteriorating quickly despite good adherence
Questions worth bringing to your appointment
- How long was it between starting treatment and this test?
- Should it be repeated after a full three months?
- Has my iron been checked, since it can push the number up?
- Have I gained weight or started any steroid over this period?
- Would adding a second medicine be better than increasing this one?