A practical, number-by-number guide to decoding your A1C, fasting glucose, C-peptide, kidney panel, and lipid results — including the exact thresholds that separate normal from prediabetes from diabetes.
To read diabetes blood test results, match each value against its reference range: an A1C under 5.7%, fasting glucose under 100 mg/dL, and 2-hour OGTT under 140 mg/dL are normal. Prediabetes ranges are 5.7–6.4%, 100–125 mg/dL, and 140–199 mg/dL; diabetes begins at 6.5%, 126 mg/dL, and 200 mg/dL respectively.[1]
- The 9-point checklist for any diabetes lab report
- Reading your report line by line: a 7-step protocol
- Diagnostic thresholds at a glance (table)
- Monitoring targets once you have diabetes (table)
- What each individual test actually tells you
- Common reading mistakes that change the meaning of a number
- What "reading it correctly" looks like in practice
- When a result means call your doctor today
- Frequently asked questions
The 9-Point Checklist for Any Diabetes Lab Report
Roughly 38.4 million Americans live with diabetes, and the CDC estimates about 1 in 5 of them don't know it — often because a lab report full of numbers never got translated into plain language.[2] Another 97.6 million adults have prediabetes, which is where accurate reading of a report changes the trajectory of care.[2] Work through these nine items in order and you will extract nearly everything a clinician extracts from the same page.
One number tells you where you were standing. A row of numbers across several visits tells you which direction you're walking.
— The core principle of lab trend interpretation
Reading Your Report Line by Line: A 7-Step Protocol
Most people scan a lab report from top to bottom and stop at the first red flag. A more reliable method is to work in a fixed order, because context determines meaning. Use these seven steps on any printout or patient-portal PDF.
Diagnostic Thresholds at a Glance
This is the single table worth saving to your phone. Every value below is drawn from the American Diabetes Association's Standards of Care, which remain the reference standard for U.S. diagnosis and classification in 2026.[1]
| Test | What it reflects | Normal | Prediabetes | Diabetes |
|---|---|---|---|---|
| A1C | Average glucose over the previous 2–3 months | <5.7% | 5.7–6.4% | ≥6.5% |
| Fasting plasma glucose | Fasting baseline after ≥8 hours without calories | <100 mg/dL | 100–125 mg/dL | ≥126 mg/dL |
| 2-hour OGTT (75 g glucose) | How well you clear a glucose load | <140 mg/dL | 140–199 mg/dL | ≥200 mg/dL |
| Random glucose with classic symptoms | Any-time glucose when thirst, urination, weight loss present | — | — | ≥200 mg/dL (diagnostic alone) |
| Fasting C-peptide | How much insulin your pancreas makes | ~0.5–2.0 ng/mL | — | Low/undetectable suggests type 1 or advanced beta-cell failure[3] |
| eGFR | Kidney filtration rate | ≥90 mL/min/1.73 m² | — | <60 for ≥3 months defines chronic kidney disease[1] |
| Urine albumin-to-creatinine ratio | Early kidney and vascular damage | <30 mg/g | 30–299 mg/g (moderately increased) | ≥300 mg/g (severely increased)[1] |
Two practical notes on reading this table. First, the A1C is a population-level tool with individual blind spots — it is less reliable in people with anemia, hemoglobin variants such as sickle cell trait, recent blood loss, advanced kidney disease, or pregnancy.[3] In those situations, clinicians lean on glucose values or continuous glucose monitoring rather than the A1C alone. Second, prediabetes is not a soft category. An A1C of 6.0–6.4% carries a substantially elevated risk of progression to type 2 diabetes within a few years, which is precisely why it appears as a distinct band rather than a gray zone.[1]
Once You Have Diabetes: What the Numbers Should Be
Diagnostic thresholds and treatment targets are not the same set of numbers, and conflating them is one of the most common reading errors. A person diagnosed with diabetes is not trying to get back under 5.7% at any cost — they are trying to hit an individualized target that balances glucose control against hypoglycemia risk, treatment burden, and other health conditions.
| Measure | Typical target for most adults | When targets are relaxed or tightened |
|---|---|---|
| A1C | <7% | <6.5% if achievable without hypoglycemia; <8% with limited life expectancy, severe hypoglycemia history, or advanced complications[1] |
| Time in range (CGM) | >70% of readings between 70–180 mg/dL | Tightened in pregnancy; relaxed in older adults on insulin[1] |
| Blood pressure | <130/80 mm Hg | Individualized if orthostatic hypotension or frailty |
| LDL cholesterol | <70 mg/dL | <55 mg/dL for those with established atherosclerotic cardiovascular disease[1] |
| eGFR and UACR | Stable eGFR; UACR <30 mg/g | Checked at least annually, more often if already abnormal[1] |
| Vitamin B12 | Within lab reference range | Measured periodically in anyone on long-term metformin[1] |
If you wear a continuous glucose monitor, you will see three additional numbers that matter as much as the A1C. Time in range describes the percentage of the day spent between 70 and 180 mg/dL. Time below range should stay under 4% of readings, and under 1% below 54 mg/dL. Glucose variability, expressed as coefficient of variation, is generally considered stable when below 36%.[1] These metrics capture the swings that an averaged A1C hides — two people with identical A1C values can have very different day-to-day glucose patterns.
What Each Individual Test Actually Tells You
The same report can contain eight or ten different markers, and each answers a distinct question. Expand the entries below to see what a specific result means and what typically drives it up or down.
A1C (hemoglobin A1c, glycated hemoglobin)
The A1C measures the fraction of hemoglobin in your red blood cells that has glucose attached to it. Because red cells live roughly three months, the result reflects a weighted average of glucose over the preceding 8–12 weeks, with the most recent month contributing the heaviest share.[3]
Read it as a trend instrument, not a daily gauge. A 7% A1C can coexist with wide swings from 50 to 250 mg/dL, or with a flat line hovering near 154 mg/dL all day. The number cannot distinguish between those two very different situations.
Fasting plasma glucose
This is the morning baseline — the glucose your liver produces overnight without food input. A value of 100–125 mg/dL indicates impaired fasting glucose, and 126 mg/dL or higher on two occasions supports a diabetes diagnosis.[1]
Because it is a single time point, it is the most vulnerable to "white coat" effects: poor sleep, an argument, an early-morning commute, or a mild infection can push a fasting glucose up by 10–20 mg/dL without any change in underlying insulin sensitivity. That is why an isolated borderline fasting value should always be repeated before it changes your diagnosis.
2-hour oral glucose tolerance test (OGTT)
You drink a 75-gram glucose solution, and your blood is drawn two hours later. This test captures the entire dynamic response — insulin secretion and peripheral glucose uptake — and is the most sensitive of the three standard tests for detecting early dysfunction.
A 2-hour value of 140–199 mg/dL is impaired glucose tolerance; 200 mg/dL or higher is diagnostic of diabetes.[1] The OGTT is also the preferred screening test for gestational diabetes, though a different protocol and threshold set applies in pregnancy.
Random glucose (with or without symptoms)
A random glucose taken at any time of day is usually reassuring when it is normal, but it is generally uninterpretable when mildly elevated — a reading of 150 mg/dL means something entirely different one hour after lunch than it does at 7 a.m. after an overnight fast.
There is one clear exception: a random glucose of 200 mg/dL or higher in someone with classic symptoms of hyperglycemia (excessive thirst, frequent urination, unexplained weight loss) is diagnostic of diabetes without a second confirmatory test.[1]
C-peptide and fasting insulin
C-peptide is released from the pancreas in equal amounts to insulin, but it is not cleared by the liver, so it gives a more stable picture of how much insulin your body is actually producing. A normal fasting C-peptide is roughly 0.5–2.0 ng/mL.[3]
Interpretation is straightforward in principle. A low or undetectable C-peptide points toward type 1 diabetes or long-standing beta-cell failure. A high C-peptide in someone with elevated glucose points toward insulin resistance — the pancreas is working hard but the signal is not getting through. Fasting insulin levels are less reproducible and are not standardized across labs, so they are best interpreted alongside C-peptide and glucose rather than on their own.
Continuous glucose monitoring metrics: GMI, time in range, variability
If you use a CGM, your report will feature a glucose management indicator (GMI) — an A1C estimate derived from sensor data. The GMI and your lab A1C should be within about 0.3 percentage points of each other; a wider gap usually means your glucose is more variable than the A1C suggests, or that a condition is distorting the A1C.
Time in range above 70%, time below 70 mg/dL under 4%, and coefficient of variation under 36% are the three benchmarks most clinicians use to judge control quality.[1] Since the FDA cleared the first over-the-counter continuous glucose monitor in 2024, many people without a diabetes diagnosis now see these numbers too — and they should be read as trend direction, not as a diagnosis.[4]
Kidney panel: eGFR and urine albumin-to-creatinine ratio
These two numbers determine whether diabetes is affecting your kidneys, and they can move in opposite directions. The eGFR estimates filtration; a value below 60 mL/min/1.73 m² sustained for three months or more defines chronic kidney disease regardless of the cause.[1]
The urine albumin-to-creatinine ratio catches damage earlier. An ACR of 30 mg/g or higher signals albuminuria — the first sign of diabetic kidney injury — and it often appears while eGFR is still perfectly normal. On a report, an ACR between 30 and 299 mg/g is moderately increased; 300 mg/g and above is severely increased.[1] Anyone with diabetes should have both checked at least once a year.
Lipid panel, liver enzymes, and vitamin B12
Diabetes doubles cardiovascular risk, so the lipid panel on your report is not incidental — it drives statin decisions. LDL cholesterol below 70 mg/dL is the target for most adults with diabetes, and below 55 mg/dL for those with established cardiovascular disease.[1] Triglycerides above 150 mg/dL and HDL below 40 mg/dL (men) or 50 mg/dL (women) are common in insulin resistance.
ALT elevation may point toward metabolic-associated fatty liver disease, which co-occurs with type 2 diabetes frequently enough that screening is recommended. Vitamin B12 deserves a specific look if you take metformin, because long-term use lowers B12 absorption and can cause a neuropathy that mimics diabetic nerve damage.[1]
Common Reading Mistakes That Change the Meaning of a Number
Most misinterpretations of diabetes labs come from a small set of predictable errors. None of them require medical training to avoid — they require knowing where the traps are.
Treating the flag column as a diagnosis. Labs flag values outside the reference range of a healthy population. An A1C of 5.8% may sit unflagged on some reports even though it is squarely in the prediabetes band.[1]
Mixing up mg/dL and mmol/L. A fasting glucose of "5.6" is normal in mmol/L and catastrophic in mg/dL. If the number looks impossibly low, check the unit before you panic or relax.
Judging a fasting test you didn't actually fast for. Coffee with cream, a morning smoothie, or even a few crackers invalidates a fasting glucose. The lab will still run it, and the result will still print — it just won't mean what the label says.
Trusting the A1C unconditionally. Anemia, pregnancy, recent transfusion, advanced kidney disease, and hemoglobin variants all shift the A1C away from true average glucose. When the A1C and your CGM or meter disagree consistently, that gap is itself clinically meaningful.[3]
Reading glucose and ignoring everything else. The kidney, lipid, and B12 results on the same page frequently change management more than a 0.2% shift in A1C. A report is a package, not a single line item.
Panicking over one abnormal value. Pain, infection, poor sleep, corticosteroids, and acute stress can all transiently raise glucose. Confirm before you reclassify yourself.[5]
What "Reading It Correctly" Looks Like in Practice
Correct interpretation is less about memorizing cutoffs and more about building a habit: same context, same comparison, same questions, every time. The patterns below distinguish people who genuinely understand their results from those who simply react to them.
You keep a one-page tracker listing each test, the date, the value, and the unit — so any new result can be dropped into a trend line in seconds.
You know your personal targets, not just the population ranges: your A1C goal, your blood pressure goal, your LDL goal, and whether your clinician has individualized any of them for your age or other conditions.
You read the fasting glucose, the A1C, and any post-meal or CGM data as three separate signals that can disagree with each other — and you know that disagreement is information, not error.
You scan the non-glucose panels every time, especially eGFR, urine albumin-to-creatinine ratio, and LDL cholesterol, because those drive long-term outcomes as much as glucose does.[1]
You bring questions written down rather than trying to recall them in a fifteen-minute appointment, and you ask specifically what a change in the number would trigger — a medication adjustment, a repeat test, or a referral.
One habit ties all of these together: annotate your results at the moment you receive them. Write the date, what you ate the night before if it was a fasting test, whether you were sick, and any medication changes in the preceding three months. Six months later, when a value looks different, that annotation is often the explanation — and without it, you'll be guessing.[5]
When a Result Means Call Your Doctor Today
Some numbers are safe to discuss at your next scheduled visit. Others need a phone call within hours. The distinction usually comes down to how far outside the range the value sits and whether symptoms are present.
There is also a quieter category that does not require an urgent call but should not be shelved: a consistently rising A1C across three visits, a new vitamin B12 deficiency in someone on long-term metformin, or a triglyceride level above 500 mg/dL. Each of these represents a trajectory rather than an event, and the earlier it is addressed, the more options remain available.[5]
Frequently Asked Questions
These are the questions readers most often ask after opening a diabetes lab report for the first time.
What is a normal A1C, and when does it mean diabetes?
A normal A1C is below 5.7%. A value from 5.7% to 6.4% indicates prediabetes, and 6.5% or higher meets the diagnostic threshold for diabetes when confirmed.[1] The A1C is calculated from the percentage of glycated hemoglobin and reflects average glucose over roughly the previous two to three months.
Can I have a normal fasting glucose but still have diabetes?
Yes. Fasting glucose can remain under 100 mg/dL while post-meal glucose or the A1C is already in the diabetes range, because impaired fasting glucose and impaired glucose tolerance reflect different underlying defects. If your fasting glucose is normal but your A1C is 6.5% or higher, that combination still supports a diabetes diagnosis and often points to post-meal spikes as the dominant problem.[1]
What is the difference between mg/dL and mmol/L on a glucose result?
They are two units for the same measurement. The United States uses mg/dL; most of the world uses mmol/L. To convert mg/dL to mmol/L, divide by 18. So 126 mg/dL equals 7.0 mmol/L, and 100 mg/dL equals 5.6 mmol/L. Always confirm which unit your lab reports before comparing a value to any threshold you read online.
Does one high reading mean I have diabetes?
Usually not. Diagnosis generally requires two abnormal results, either the same test repeated on a different day or two different abnormal tests — with one exception: a random glucose of 200 mg/dL or higher accompanied by classic symptoms such as excessive thirst and urination is diagnostic on its own.[1] Single elevated readings from stress, illness, or medications such as steroids are common and should be confirmed.
What is a good A1C if I already have diabetes?
For most non-pregnant adults, the target is below 7%. A goal below 6.5% may be appropriate for those who can reach it without hypoglycemia, and a more relaxed goal around 8% is often safer for people with limited life expectancy, a history of severe hypoglycemia, or advanced complications.[1] The right target is a decision your clinician should make with you, not a universal number.
Why does my CGM reading differ from my blood glucose meter?
Continuous glucose monitors measure glucose in interstitial fluid, which lags behind blood glucose by roughly 5 to 15 minutes. The gap is widest when glucose is changing quickly — after a meal, during exercise, or while treating a low. A more persistent discrepancy between your GMI and your lab A1C, however, is worth raising with your clinician, because it can signal that the A1C is being distorted by another condition.[1]
Which tests should I have checked every year with diabetes?
At minimum: A1C (typically every 3 to 6 months depending on control), blood pressure at every visit, a lipid panel, eGFR and urine albumin-to-creatinine ratio, a dilated eye exam, a comprehensive foot exam, and vitamin B12 if you take metformin long term.[1] These are the components of standard annual diabetes care, and each one appears somewhere on a typical lab and visit schedule.
- Match each value to its own reference range: A1C under 5.7%, fasting glucose under 100 mg/dL, and 2-hour OGTT under 140 mg/dL are normal; diabetes begins at 6.5%, 126 mg/dL, and 200 mg/dL.[1]
- Diagnostic thresholds and treatment targets differ — most adults with diabetes aim for an A1C below 7%, not below 5.7%.
- A1C reflects a 2–3 month average and can be falsely high or low in anemia, pregnancy, kidney disease, and hemoglobin variants; confirm with glucose data when results disagree.
- Read the entire report, not just glucose — eGFR, urine albumin-to-creatinine ratio, LDL cholesterol, and vitamin B12 frequently change management more than a small A1C shift.[1]
- One abnormal value rarely diagnoses diabetes; confirmation on a second test is standard, except when a random glucose of 200 mg/dL or higher occurs with classic symptoms.
- Track trends across visits and annotate each result with context — fasting status, illness, and medication changes — so future comparisons remain meaningful.
- American Diabetes Association. Standards of Care in Diabetes—2026. Diabetes Care. diabetes.org
- Centers for Disease Control and Prevention. National Diabetes Statistics Report. cdc.gov
- MedlinePlus (U.S. National Library of Medicine). Hemoglobin A1c (HbA1c) Test; C-Peptide Test. medlineplus.gov
- U.S. Food and Drug Administration. Clearance of the first over-the-counter continuous glucose monitoring system (2024). fda.gov
- National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Diabetes Tests & Diagnosis. niddk.nih.gov
