HbA1c and Time in Range:
What Gets Measured Gets Managed
Everything in one place. Read the plain version with Jude, earn your way into the evidence with Grace, then the full model with John. Stop wherever you have enough.
How we teach: three rules, borrowed from Taleb
You earn each level by showing you understand it, not by scrolling past it. We only teach what we would use on ourselves and the people we love.
Understanding beats memory and luck, so the checks reshuffle every time you retry. A pass means you got it, not that you guessed it. And we teach you to tell a trend (signal) from one reading (noise).
We give you the scaffolding and get out of your way. Roam where your curiosity leads, go as deep as you want, and ask Grace anything. We will not teach a bird how to fly.
Want this for your own numbers, in your units? Ask Grace, then take it to your care team.
One page, three depths
This guide compounds: each layer rests on the one beneath it. Read Jude’s plain version, then pass a short understanding check to open Grace, then another to open John. You can roam freely within a layer; you cannot skip ahead a layer, because the next one would not make sense and you would be standing on a gap.
The whole thing, in plain words
Two numbers tell you most of what you need about glucose over time. Time in Range is how much of the day your glucose sits in a good zone, between 3.9 and 10.0 mmol/L (70 to 180 mg/dL); your here-and-now number. HbA1c is a three-month report card from the lab; your over-time number. “What gets measured gets managed” is true, with one catch: measuring everything just wears you out. Measure the few that matter, and read them kindly.
Two people can spend the same time in range and still get different HbA1c results, because bodies attach glucose to blood cells at different rates. So your own pattern matters more than the textbook average; a higher-than-expected HbA1c is often your biology, not a failing.
Here is the big idea, and it is good news. If your HbA1c is high, bringing it down protects your heart and eyes a great deal; the first steps down do the most good. But once it is already in a good place, pushing lower helps only a little more and can cause more hypos. (These are average patterns across large studies, not a prediction about your own risk.) So the aim is a steady, good-enough number, set with your diabetes team, not the lowest possible one.
Most of the protection is banked by the time you reach a good-enough range; the heart is full there. Going lower to the floor adds no extra protection and brings more hypos. Aim for steady and good-enough, set with your team.
Does this match the life of the person living it? If a number ever makes you feel like a failure, the number is being used wrongly. Aim for steady and liveable, not flawless.The Pemberton lens, lived recognisability, one of the four GNL appraisal lenses.
The numbers underneath
The two numbers translate into each other
From Beck 2019, in a large type 1 group,1 every 10 percentage points of time in range is worth about 0.6% on HbA1c, and 70% time in range lands, on average, at about 53 mmol/mol (7.0%), an average glucose near 8.6 mmol/L (154 mg/dL).
| Time in Range | HbA1c mmol/mol | % | Mean glucose | Across real people (95%) |
|---|---|---|---|---|
| 90% | 42 | 6.0 | 7.0 mmol/L (126) | 28 to 56 (4.7 to 7.3%) |
| 80% | 48 | 6.5 | 7.8 mmol/L (140) | 33 to 62 (5.2 to 7.8%) |
| 70% | 53 | 7.0 | 8.6 mmol/L (154) | 38 to 67 (5.6 to 8.3%) |
| 60% | 57 | 7.4 | 9.4 mmol/L (169) | 43 to 73 (6.1 to 8.8%) |
| 50% | 63 | 7.9 | 10.2 mmol/L (183) | 49 to 77 (6.6 to 9.2%) |
Same time in range, different HbA1c
At 70% time in range the average HbA1c is 53, but real people span 38 to 67 mmol/mol (5.6 to 8.3%). That is biology, not error, and it is why your own pairing beats the average.
What a change in HbA1c does to your heart
Using the long DCCT/EDIC data, the risk of a heart event rises steeply as HbA1c climbs.2 Across the full span from 10% (86 mmol/mol) to 5% (31), relative risk falls about five-fold on this model, but the steps are not equal: the drop near the top removes far more than the drop near the bottom.
Relative risk, the shape of risk, not a personal probability. Coming down from a high HbA1c removes a lot; the move from 6% (42) to 5% (31) removes the least.
A big-sounding percentage is not the same as a big difference. Whenever you are handed a risk number, ask the two questions that keep everyone honest: out of how many, and over how long?The Goldacre lens, evidence-grade discipline, one of the four GNL appraisal lenses.
The model, the numbers, the limits
The cardiovascular-risk model
DCCT/EDIC found the risk of a first cardiovascular event rose by a hazard ratio of 1.38 for every 1% (about 11 mmol/mol) higher HbA1c.3 Turned into a ladder from 10% to 5% (showing every 1% step), relative to the 5.0% (31) floor:
| HbA1c % | mmol/mol | Mean glucose | Relative CV risk (×) | Risk removed by this 1.0% step |
|---|---|---|---|---|
| 10.0 | 86 | 13.4 (240) | 5.00 | n/a |
| 9.0 | 75 | 11.8 (212) | 3.63 | 1.37 |
| 8.0 | 64 | 10.2 (183) | 2.63 | 1.00 |
| 7.0 | 53 | 8.6 (154) | 1.90 | 0.73 |
| 6.0 | 42 | 7.0 (126) | 1.38 | 0.52 |
| 5.0 | 31 | 5.4 (97) | 1.00 | 0.38 |
D The hazard ratio is Grade A (Bebu 2020); the ladder built from it is a Grade-D GNL model. Relative risk = 1.38 to the power of the HbA1c difference in %.3 It shows the shape of risk, not a personal probability. Mean glucose via ADAG (Nathan 2008).4
Each further drop buys less while its cost rises. The trial that proved tight control works also recorded about three times the rate of severe hypoglycaemia at the lowest HbA1c.5
Why you need both numbers
HbA1c is the strongest single predictor of long-term microvascular complications.2 5 Time in Range adds a here-and-now view HbA1c cannot give; as a long-term predictor, though, most of its signal turns out to be HbA1c information seen through a different window (Beck 2019; Lachin 2022).1 6 They overlap heavily, are not rivals, and neither replaces the other.
Your target, and where to settle
Guidance treats 53 mmol/mol (7.0%) as the strongly-evidenced target for most people (ISPAD in paediatrics, with equivalent adult targets in ADA and NICE), and the tighter 48 mmol/mol (6.5%) only where CGM and an automated system make it safely reachable, and on weaker evidence.7 A high glycator holds a higher time in range to bank the same protection; a low glycator has more room. Same destination, different routes, set with your team.
Where to settle is a conversation, not a race to the floor: the gain shrinks and the hypo cost rises as you approach it.
It is the rare, large swing that does the lasting damage, not the average Tuesday. Keep the highs short and the lows rare, and protect hardest against the catastrophic low you cannot afford.The Taleb lens, robustness to outliers, one of the four GNL appraisal lenses.
A model is only as honest as its assumptions. This rides on a single hazard ratio from one historic cohort; that is a clean lens, not the whole view. Name what would strengthen it, and never sell the model as the territory.The Hayes lens, technical and methodological rigour, one of the four GNL appraisal lenses.
The whole guide, summarised
Glucose never lies; it just keeps an honest record over months. Read it kindly, and aim for the number you can live with.
Find your glycator pattern
Enter at least three pairs: a lab HbA1c (mmol/mol) and the average CGM glucose (mmol/L) for that same 90-day window, spread over at least nine months. This estimates whether you glycate higher or lower than the textbook. It is a teaching tool, not a target, and not a formal GNL Explorer.
No values are pre-filled; you enter every number fresh. Population-average method (the glycation gap between your measured HbA1c and the HbA1c your mean glucose predicts, via ADAG). The formal Explorer is deferred until peer-reviewed publication and cross-sensor validation. This is a taste of the explorers inside Learn with Grace; confirm anything here with your care team.
This page is the taster. The full journey, three modules and their 30 questions, with your progress saved, lives in Learn with Grace. Glucose never lies; come and learn to read it.
References
Evidence grades A (strongest) to D (editorial or working analysis).
- Beck RW, et al. The Relationships Between Time in Range, Hyperglycemia Metrics, and HbA1c. J Diabetes Sci Technol. 2019;13(4):614-626 (the TIR-to-HbA1c regression); and Validation of Time in Range as an Outcome Measure. Diabetes Care. 2019;42(3):400-405, DOI 10.2337/dc18-1444. A
- Nathan DM. The DCCT/EDIC study. Diabetologia. 2021;64:1049-1058; DCCT, N Engl J Med. 1993;329(14):977-986. A
- Bebu I, Schade D, Braffett B, et al; DCCT/EDIC Research Group. Risk factors for first and subsequent CVD events in type 1 diabetes. Diabetes Care. 2020;43(4):867-874. DOI: 10.2337/dc19-2292. A
- Nathan DM, et al; ADAG Study Group. Translating the A1C assay into estimated average glucose. Diabetes Care. 2008;31(8):1473-1478. A
- DCCT Research Group. N Engl J Med. 1993;329(14):977-986. (Severe hypoglycaemia cost of intensive control.) A
- Lachin JM, et al; DCCT/EDIC Research Group. Association of estimated time-in-range capillary glucose with HbA1c and complications. Diabetes Care. 2022;45(10):2445-2448, DOI 10.2337/dc21-2298. A
- de Bock M, Agwu JC, et al. ISPAD 2024 Clinical Practice Consensus Guidelines: Glycemic Targets. Horm Res Paediatr. 2024;97(6):546-554; adult targets per ADA and NICE NG17. A
One page, three voices: Jude, Grace, John. Population-average, not personalised.
Estimate your GMI from a mean glucose
Enter the mean glucose from your report. The Glucose Management Indicator (GMI) is the HbA1c that this mean glucose tends to predict across a population (Bergenstal 2018). It is an estimate, not your laboratory HbA1c; the two can differ in any one person, which is the point, not a fault.
Formula: GMI(%) = 3.31 + 0.02392 × mean glucose in mg/dL; mg/dL = mmol/L × 18.018; IFCC(mmol/mol) = 10.929 × (GMI% minus 2.15). Educational only and population-average; it is never a dose to take, and your laboratory HbA1c is confirmed with your care team.