You go to bed with a stable reading, eat nothing overnight, and wake up higher than when you went to sleep.
It’s one of the most frustrating patterns in diabetes management, partly because it seems to defy arithmetic. No food went in, yet glucose went up.
There are three main explanations, and they require opposite responses. Two of them look identical on a morning reading, which is why getting the diagnosis right matters more than any single adjustment you might make.
Quick Fact: The dawn phenomenon affects more than 50% of people with type 1 or type 2 diabetes — making it far and away the most common cause of morning highs, and considerably more common than the Somogyi effect it’s often confused with.
Last reviewed: October 2026
First: Is Your Number Actually High?
Before diagnosing a pattern, it helps to know what counts as elevated.
Fasting Glucose Reference
| Fasting reading | Category |
|---|---|
| Below 100 mg/dL | Normal |
| 100–125 mg/dL | Prediabetes range |
| 126 mg/dL or above | Diabetes range (on repeat testing) |
| 80–130 mg/dL | Common pre-meal target for people already diagnosed |
Two things worth noting. A person already diagnosed with diabetes has a treatment target, not a diagnostic threshold — 115 may be perfectly acceptable for them and a reason for screening in someone undiagnosed. And a single morning reading proves very little either way; it’s the multi-day pattern that carries information.
Our guide to normal blood sugar levels covers the full set of ranges.
The Three Causes
1. Dawn phenomenon. In the early morning hours — typically between 2 a.m. and 8 a.m. — your body releases cortisol, growth hormone, and catecholamines to prepare you for waking. These hormones signal your liver to release glucose for energy. In someone without diabetes, the pancreas releases matching insulin and levels stay stable. With diabetes, that compensation falls short, and glucose climbs.
This is a normal physiological process producing an abnormal result. Nothing has gone wrong overnight; the body simply can’t cover a routine hormonal surge.
2. Waning insulin. Longer-acting insulin wearing off before morning, or basal pump settings supplying too little overnight. Not a hormonal event — simply not enough insulin present during the hours it’s needed.
3. Somogyi effect (rebound hyperglycemia). Blood sugar drops too low overnight, and the body responds by releasing the same counterregulatory hormones — adrenaline, cortisol, growth hormone, glucagon — which overcorrect and drive glucose up. The morning reading is high because it was low a few hours earlier.
Comparing the Three
| Cause | What’s happening overnight | Correct response | Applies to |
|---|---|---|---|
| Dawn phenomenon | Hormone surge raises glucose | May need more overnight coverage | Everyone with diabetes |
| Waning insulin | Insulin runs out too early | Adjust timing or dose | Insulin users only |
| Somogyi effect | Glucose drops too low, rebounds | May need less insulin | Insulin or sulfonylurea users |
That last column matters more than it looks. If you don’t use insulin or a sulfonylurea, two of these three causes can’t apply to you — which narrows the question considerably before you test anything.
Why the Distinction Actually Matters
Here’s the part that makes this genuinely important rather than academic.
Dawn phenomenon and Somogyi effect can produce identical morning numbers — and the appropriate correction is the exact opposite in each case.
If it’s the dawn phenomenon, you may need more overnight insulin coverage. If it’s the Somogyi effect, more insulin would worsen the overnight low that caused the rebound in the first place, potentially dangerously.
Acting on the morning number alone means guessing between two options where one could be harmful. This is why the diagnostic step below isn’t optional.
How to Tell Them Apart
The method is straightforward: check what’s happening at 2–3 a.m.
- Glucose high or rising at 3 a.m. → dawn phenomenon or waning insulin
- Glucose low at 3 a.m. → suspect the Somogyi effect
Do this over several consecutive nights, not one. A single night can mislead — overnight patterns vary with the previous day’s food, activity, stress, and medication timing.
A continuous glucose monitor makes this considerably easier, since it captures the entire overnight curve rather than a single snapshot. If you have access to one, it answers the question in a night or two rather than a week of alarm-setting.
The Night Log
Record these five columns for three to five consecutive nights. The pattern is usually obvious once it’s written down.
| Night | Bedtime reading | 3 a.m. reading | Waking reading | Evening notes |
|---|---|---|---|---|
| 1 | Dinner time, carbs, exercise, alcohol, meds | |||
| 2 | ||||
| 3 | ||||
| 4 | ||||
| 5 |
The last column is the one people skip and the one that most often explains the outlier night.
The Somogyi Debate Worth Knowing About
Honest coverage requires this: the Somogyi effect is scientifically contested.
Named after Dr. Michael Somogyi, who described it in the 1930s, the theory has been challenged by more recent continuous glucose monitoring research. Studies using CGM have disputed the mechanism, and clinicians have observed that patients with early-morning hyperglycemia tend to show high overnight glucose readings rather than low ones.
The American Diabetes Association describes the Somogyi effect as a “much rarer” cause than the dawn phenomenon and waning insulin.
Why include it at all? Because it still appears in clinical guidance, it’s still checked for, and rebound hyperglycemia after a genuine overnight low does occur. But if you’re weighing probabilities before testing, the dawn phenomenon is by far the more likely explanation.
If You Don’t Use Insulin
Most coverage of morning highs is written for insulin users, which leaves out a large share of the people asking the question. If you have type 2 diabetes managed with diet, metformin, or other non-insulin medication, the dawn phenomenon still applies to you — your options are simply different ones.
What’s Available, By Situation
| Approach | Non-insulin users | Insulin users |
|---|---|---|
| Evening walk or light activity after dinner | Yes — improves overnight insulin sensitivity | Yes |
| Earlier dinner, more hours before bed | Yes | Yes |
| Lower-carbohydrate evening meal | Yes | Yes, with dose considerations |
| Limiting evening alcohol | Yes — it disrupts overnight glucose in both directions | Yes |
| Improving sleep duration and consistency | Yes | Yes |
| Reviewing medication timing with your provider | Yes — some medications can be shifted | Yes |
| Adjusting long-acting insulin timing | — | Provider-directed |
| Basal rate changes for pump users | — | Provider-directed |
| Reducing evening insulin for suspected Somogyi | — | Provider-directed |
The first five rows are available to everyone and require no prescription. Evening activity in particular has a reasonable rationale behind it — our guide to exercise and blood sugar covers how post-meal movement works, with the caveat that intense late exercise can cause delayed overnight lows in insulin users.
Sleep deserves its own mention. Short or fragmented sleep raises cortisol and amplifies the same hormonal surge that drives the dawn phenomenon — so a bad night makes the morning reading worse, independent of anything you ate. Our guide to sleep and blood sugar covers the mechanism.
What Actually Helps — Medication Side
For dawn phenomenon: adjusting the timing of long-acting insulin rather than only the dose; basal rate adjustments for pump users during early morning hours specifically; reviewing evening carbohydrate intake and timing.
For waning insulin: later dosing of long-acting insulin; discussing a longer-duration formulation with your provider; pump basal profile adjustment.
For suspected Somogyi effect: reducing evening insulin rather than increasing it; reviewing whether a bedtime snack is appropriate; avoiding intense exercise immediately before bed.
Every one of these is a medication or regimen change. None should be made independently — they require your provider’s involvement, because getting the direction wrong carries real risk.
What People Actually Run Into
The most common instinct is exactly the wrong one: seeing a high morning number and adding insulin.
This is understandable and often correct for the dawn phenomenon. But applied to a Somogyi pattern, it deepens the overnight low that’s driving the rebound. People frequently describe a cycle of escalating evening doses with morning numbers that stay stubbornly high — because they’ve been treating the symptom while worsening the cause.
The second pattern is frustration at the apparent unfairness. Eating nothing and waking higher feels like a failure of effort, and people often assume they’ve done something wrong. The dawn phenomenon is not a behavioral failure — it’s a hormonal event that happens on schedule regardless of what you did the previous evening.
Third: many people never do the 3 a.m. check because it’s genuinely unpleasant. This is understandable, but it’s also why the pattern often goes unresolved for months. A CGM removes this obstacle entirely, and it’s worth raising with your provider specifically for this purpose if the morning highs are persistent.
And a fourth, specific to type 2: people not on insulin often read an article like this, find only insulin advice, and conclude there’s nothing to be done. The dawn phenomenon affects them too, and evening timing, activity, alcohol and sleep are all genuinely available levers.
Common Mistakes
Increasing evening insulin without checking overnight levels. If the cause is a rebound from an overnight low, more insulin makes it worse. Test first.
Judging from a single morning reading. One high morning is normal variation and has little effect on A1C. A consistent multi-day pattern is what warrants investigation.
Checking only once at 3 a.m. Overnight patterns vary night to night. Several consecutive nights give a real picture; one gives a guess.
Assuming you ate something wrong. The dawn phenomenon is hormonal and occurs independent of the previous evening’s meal.
Adjusting your own medication. These corrections move in opposite directions depending on cause, and the wrong direction can cause dangerous hypoglycemia.
Ignoring persistent morning highs as “just mornings.” Occasional highs matter little, but consistent ones raise A1C meaningfully over time.
Assuming nothing applies if you don’t use insulin. Two of the three causes require insulin or sulfonylureas — but the most common cause does not.
Practical Tips
Log bedtime, 3 a.m., and waking readings for three to five consecutive nights before drawing any conclusion. Use the table above.
Record evening context alongside the numbers: dinner timing, carbohydrate amount, exercise, alcohol, and medication time. Patterns usually emerge from the context, not the readings alone.
Ask your provider about a CGM specifically for overnight pattern detection if morning highs are persistent — it removes the need for alarm-set testing.
Bring the full log to your appointment rather than describing the pattern from memory.
Ask specifically about medication timing, not just dose — for waning insulin, when you take it matters as much as how much.
If you don’t use insulin, start with the two changes that cost nothing: finish dinner earlier, and walk for ten to fifteen minutes after it.
When Should You See a Doctor?
Contact your provider if morning highs persist for several consecutive days, since consistent elevation raises A1C over time even when individual readings seem manageable. Bring your overnight log — the 3 a.m. data is what allows an accurate adjustment rather than a guess.
Seek prompt guidance if you experience symptoms of overnight lows: night sweats, nightmares, waking with a headache, or unusual morning fatigue. These can indicate nocturnal hypoglycemia, which carries its own risks independent of the morning reading.
See a doctor about screening if you don’t have a diabetes diagnosis and your fasting readings are consistently above 100 mg/dL — morning highs in an undiagnosed person are a reason for proper testing, not self-management.
Do not adjust insulin doses independently. The correct adjustment differs by cause, and moving in the wrong direction can cause serious hypoglycemia.
Frequently Asked Questions
Why is my blood sugar higher in the morning than before bed? Most commonly the dawn phenomenon — a normal early-morning hormone surge that signals the liver to release glucose. In diabetes, insulin can’t fully compensate, so levels rise despite eating nothing.
How do I know if it’s the dawn phenomenon or the Somogyi effect? Check your glucose at 2–3 a.m. across several nights. High or rising suggests dawn phenomenon; low suggests a Somogyi rebound. A CGM captures this automatically. If you don’t use insulin or a sulfonylurea, a Somogyi rebound is very unlikely.
What can I do about morning highs if I don’t take insulin? The dawn phenomenon still applies. Finishing dinner earlier, walking after the evening meal, limiting evening alcohol, improving sleep consistency, and reviewing medication timing with your provider are all available and don’t require a prescription.
Does the dawn phenomenon affect people without diabetes? The hormonal surge occurs in everyone, but people without diabetes produce matching insulin, so glucose stays stable. The rise only becomes visible when insulin response is impaired.
Is one high morning reading a problem? Generally not. An occasional morning high has little impact on A1C. Consistent highs across multiple days are what warrant investigation and adjustment.
The Bottom Line
Morning highs have three common explanations that produce the same number and demand different responses — sometimes opposite ones. The dawn phenomenon is by far the most likely, affecting over half of people with diabetes, and it isn’t caused by anything you did the night before.
The single most useful step isn’t an adjustment at all. It’s writing down what your glucose is actually doing at 3 a.m., across several nights, before changing anything — and if you don’t use insulin, starting with the two things that cost nothing: an earlier dinner and a walk after it.
This article is for general informational purposes and is not a substitute for personalized medical advice. Never adjust insulin or diabetes medication without consulting your healthcare provider.
