Hypoglycaemia:
Knowing the Low, and Treating It Calmly
The most safety-critical topic in diabetes, taught kindly. Read the plain version with Jude, earn your way into the evidence with Grace, then the full depth 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 to understand a hypo in 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
A hypo is when your glucose drops too low. The widely-used marker is below 3.9 mmol/L (70 mg/dL), often rounded to “below 4”. It is the one thing in diabetes that can need help in minutes rather than weeks, so it is worth knowing well, and worth meeting calmly rather than with fear.
Your body usually warns you first. Common early signs are shaking, sweating, a thumping heart, hunger, going pale, feeling anxious or irritable. As glucose falls further the signs become more about the brain: trouble concentrating, slurred or muddled words, confusion, blurred vision. Different people feel different signs, and your own pattern is the one to learn. The simplest rule is the kindest one: if you feel low, check; if you cannot check, treat anyway.
For treatment, the approach almost every diabetes team teaches has a simple shape, and it is worth recognising so the steps make sense rather than feeling like a panic. The widely-taught approach is: take a fast-acting glucose (something that is mostly pure glucose, not fat or fibre), wait a short while without piling on more, then re-check, and repeat only if you are still low. The waiting is the hard part; glucose takes a little time to arrive, and treating again too soon is the usual reason a hypo turns into a high afterwards. Pure glucose works faster and more reliably than sweets or juice, which is why teams favour it.1
This page explains the why behind that approach; it does not set your dose or your plan. The exact amount, the exact wait, and what to carry are personal, and they belong with your diabetes care team. What this layer gives you is the shape, so the plan they agree with you makes sense.
Does this match the life of the person living it? A hypo is frightening in the moment, and fear makes people over-treat and then ride a high for hours. The calm version, a measured amount and a short wait, is not just tidier; it is what actually feels better by the evening.The Pemberton lens, lived recognisability, one of the four GNL appraisal lenses.
The numbers underneath
Two thresholds, not one: below 3.9 and below 3.0
International consensus (ATTD/IHSG, Battelino 2019; ADA Standards of Care 2024) splits the low range into two levels.1 Below 3.9 mmol/L (70 mg/dL) is Level 1, the alert value: the point at which it is sensible to act. Below 3.0 mmol/L (54 mg/dL) is Level 2, clinically significant hypoglycaemia, low enough to impair thinking and to matter in its own right. A third level, severe hypoglycaemia, is defined by needing someone else’s help, regardless of the reading.
| Level | Threshold | What it means |
|---|---|---|
| Level 1, alert | below 3.9 mmol/L (70 mg/dL) | Time to act; the standard treatment threshold |
| Level 2, significant | below 3.0 mmol/L (54 mg/dL) | Low enough to impair thinking; counts on its own |
| Level 3, severe | any low needing help | Defined by needing assistance, not by a number |
Why 3.9 and not the physiological figure? The body’s own counter-measures begin around 3.3 mmol/L (60 mg/dL), and clear symptoms often appear below 3.0. The 3.9 line is set deliberately higher to give continuous sensors a safety margin: a true 3.2 can read as high as 4.0 and still count as accurate, so treating at “below 4” tends to catch real lows early.2
The low you do not feel: nocturnal hypoglycaemia
Lows at night are common and easy to miss, because sleep removes the warnings. This is where modern pump technology earns its place. In the ASPIRE In-Home trial, a pump that suspended insulin at a low threshold cut the amount of overnight low (the nocturnal hypoglycaemia area-under-the-curve) by about 38%, with average HbA1c essentially unchanged.3 In children, a predictive version, suspending before the low arrives, cut time spent below 3.9 by about 50 to 54% across 3,420 nights, with no severe lows and no rise in ketones.4
A schematic of the overnight pattern, not data from one person. Predictive suspension keeps the trace above the low zone; the older approach allows a long dip while you sleep. Shapes are illustrative.
When the warnings fade: impaired awareness
Some people gradually stop feeling their lows. This is impaired awareness of hypoglycaemia, and it is not carelessness; it is a measurable change in the body’s alarm system. It affects roughly 15 to 30% of adults with type 1 diabetes, less often in CGM users.5 It can be screened in clinic in under a minute with the Gold score (a single one-to-seven question; four or more means awareness is impaired). It matters because people with impaired awareness carry about a six-fold higher risk of a severe low.5 6 The reassuring part, covered in the next layer, is that it is often reversible.
A “six-fold higher risk” sounds alarming until you anchor it: six times a small number is still a small number for most people on most nights. The figure earns its keep not as a scare, but as a reason to screen awareness and to take a faded warning seriously rather than ignore it.The Goldacre lens, evidence-grade discipline, one of the four GNL appraisal lenses.
The broken alarm, and the cost of chasing low
Why the warnings fail, and why that is reversible
In type 1 diabetes the body’s defence against a low is damaged in two steps. First, the glucagon response (the hormone that should release stored glucose) is lost early and permanently. Second, the back-up adrenaline response gets blunted by recent lows, which also blunts the symptoms you feel. Each low makes the next one quieter. Cryer named this self-feeding loop hypoglycaemia-associated autonomic failure (HAAF), and it is the mechanism behind impaired awareness.6
The clinically important part is that the adrenaline-and-symptom side is reversible. Scrupulously avoiding lows for a few weeks to months restores the warning symptoms and the hormonal response, in both short-duration and long-duration type 1 diabetes (Fanelli 1993; Cranston 1994).7 The glucagon deficit does not come back, but the alarm you actually rely on does. That reframes impaired awareness from a permanent loss into a treatable state, and it makes hypo avoidance, not just lower averages, a first-order therapeutic aim.
The cost of chasing a low number: the DCCT lesson
The trial that proved tight glucose control prevents long-term complications also recorded its price. In the DCCT, the intensive-therapy group reached a mean HbA1c near 7% (53 mmol/mol) and gained large protection against eye, kidney and nerve damage; but they experienced severe hypoglycaemia at about three times the rate of the conventional group, roughly 61 versus 19 events per 100 patient-years.8 Risk of severe low rose continuously as HbA1c fell. This is the central tension of the topic: the same effort that protects the long term raises the short-term danger, unless the lows are managed deliberately.
The DCCT threefold severe-hypo cost is the historical worst case, measured before continuous sensors and automated insulin delivery existed. The modern lesson is not “loosen control”; it is “earn the low average without buying the lows”, which is exactly what the technology in the layer above is designed to do.
Prevention is the goal, and it is now measurable
The modern target is to keep time below range small: international consensus sets time below 3.9 mmol/L at no more than 4% of the day, and time below 3.0 at no more than 1%.1 UK real-world data on 15,777 people validated this: a time-below-range of about 4% over two weeks rules out high severe-hypo risk with roughly 97% certainty, and a faded warning (Gold score) adds risk on top, independently, so both should be reviewed together.9 The older trap was that pushing time-in-range up tended to drag time-below-range up with it; without the right algorithm the two move together, not apart. Low-glucose-suspend and predictive-suspend systems are the first tools that reliably break that link, lowering the lows without raising the average.3 4 9
It is the rare, catastrophic low, the one that needs another person, or comes in the night, that does the lasting harm, not the average mild dip. Prevention should be weighted toward the tail you cannot afford, not toward shaving the typical Tuesday.The Taleb lens, robustness to outliers, one of the four GNL appraisal lenses.
Several of these numbers rest on small or older studies: the reversibility evidence is a handful of clamp studies in a few dozen people; the DCCT cost predates modern insulin and sensors. They are honest signposts, not settled constants. Name the limits, and never sell a population average as a personal certainty.The Hayes lens, technical and methodological rigour, one of the four GNL appraisal lenses.
The whole guide, summarised
Glucose never lies; a low is information, not a failure. Learn your own signs, meet the low calmly, and build your plan 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. Want to ask about a hypo on your own terms? Ask Grace, then take it to your care team.
References
Evidence grades A (strongest) to D (editorial or working analysis).
- Battelino T, et al. Clinical Targets for Continuous Glucose Monitoring Data Interpretation (ATTD / IHSG consensus). Diabetes Care. 2019;42(8):1593-1603, DOI 10.2337/dci19-0028; and American Diabetes Association. Standards of Care in Diabetes 2024, Section 6 (Glycemic Targets). Diabetes Care. 2024;47(Suppl 1). The Level 1 / Level 2 / Level 3 framework and the time-below-range targets. A
- The CGM accuracy margin underpinning the 3.9 mmol/L (70 mg/dL) treatment threshold (a true ~3.2 mmol/L can read up to ~4.0 mmol/L within the accepted accuracy band). GNL working synthesis drawn from CGM accuracy literature; see the GNL CGM guide. D
- Bergenstal RM, et al; ASPIRE In-Home Study Group. Threshold-based insulin-pump interruption for reduction of hypoglycemia. N Engl J Med. 2013;369(3):224-232, DOI 10.1056/NEJMoa1303576. Nocturnal hypoglycaemia AUC reduced ~37.5%, HbA1c unchanged. A
- Buckingham BA, et al. Predictive low-glucose insulin suspension reduces duration of nocturnal hypoglycemia in children. Diabetes Care. 2015;38(7):1197-1204, DOI 10.2337/dc14-3053. 50 to 54% reduction in time below 3.9 mmol/L over 3,420 nights; no severe hypo, no ketone rise. A
- Gold AE, MacLeod KM, Frier BM. Frequency of severe hypoglycemia in patients with type 1 diabetes with impaired awareness of hypoglycemia. Diabetes Care. 1994;17(7):697-703 (Gold score; ~6-fold severe-hypo risk); and Geddes J, et al. Prevalence of impaired awareness of hypoglycaemia in adults with type 1 diabetes. Diabet Med. 2008;25(4):501-504 (UK prevalence ~19.5% Gold, ~25% Clarke). B
- Cryer PE. Mechanisms of hypoglycemia-associated autonomic failure in diabetes. N Engl J Med. 2013;369(4):362-372, DOI 10.1056/NEJMra1215228. The HAAF framework. A
- Fanelli CG, et al. Meticulous prevention of hypoglycemia normalizes the glycemic thresholds and magnitude of most of neuroendocrine responses to, symptoms of, and cognitive function during hypoglycemia in intensively treated patients with short-term IDDM. Diabetes. 1993;42(11):1683-1689; and Cranston I, et al. Restoration of hypoglycaemia awareness in patients with long-duration insulin-dependent diabetes. Lancet. 1994;344(8918):283-287. B
- DCCT Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329(14):977-986. Severe hypoglycaemia ~61 vs ~19 events per 100 patient-years, an approximately threefold cost of intensive control. A
- Deshmukh H, et al; Association of British Clinical Diabetologists (ABCD) audit. Time below range and impaired awareness as independent markers of severe hypoglycaemia in type 1 diabetes (n=15,777). Diabetes Care. 2024. Time-below-range ~4% over 14 days excludes severe-hypo risk at ~97% negative predictive value. A
One page, three voices: Jude, Grace, John. Population-average, not personalised.