The problem: your meter says one unit, your doctor's chart says the other
You bought a glucose meter in the US and it reads mg/dL. Your doctor in London — or the research paper you're reading, or the LibreLink app your friend uses — speaks in mmol/L. The numbers are not close to each other: 90 mg/dL is 5.0 mmol/L, and 200 mg/dL is 11.1 mmol/L. You can't eyeball the conversion because the factor isn't 10 or 100 — it's 18.0182, a number that comes from the molar mass of glucose and that nobody memorizes. The honest move is one multiplication or one division, and a tool that also tells you which clinical range the result falls into, because a number without its range is just a number.
Fastest path
Open the Blood Sugar Converter, type the value, pick the unit, read both.
Input: 90 mg/dL
→ mmol/L: 5.00
→ Category: Normal (Fasting 70–99 mg/dL)
Input: 7.2 mmol/L
→ mg/dL: 129.7
→ Category: Pre-Diabetic (Fasting 100–125 mg/dL)
The tool converts through the factor 18.0182 and labels the result with its ADA clinical category — normal, pre-diabetic, diabetic, or low (hypoglycemia). The rest of this guide is why the factor is what it is, what the ranges mean, and how the HbA1c estimator works.
The substance: one factor, two units, six ranges
The factor: 18.0182
1 mmol/L = 18.0182 mg/dL. The number comes from the molar mass of glucose (C₆H₁₂O₆ = 180.156 g/mol) and the volume conversion between liters and deciliters (1 dL = 0.1 L). One millimole of glucose is 180.156 mg. Dissolved in a liter, that's 180.156 mg/L = 18.0156 mg/dL. The tool uses 18.0182, a commonly cited rounded value; the difference from 18.0156 is 0.014%, which is below the precision of any home glucose meter.
The conversion is one operation:
mg/dL → mmol/L: divide by 18.0182
mmol/L → mg/dL: multiply by 18.0182
People who use 18 as a shortcut get close (5.0 mmol/L × 18 = 90 mg/dL, vs the exact 90.1), but 18 introduces a 0.1% error per conversion. That error compounds if you're logging hundreds of readings — use the factor, not the shortcut.
Why two units exist
mg/dL (milligrams per deciliter) is mass concentration — it measures the mass of glucose in a volume of blood. It's used in the United States, Japan, and a handful of other countries.
mmol/L (millimoles per liter) is molar concentration — it measures the number of glucose molecules in a volume of blood. It's used in Canada, the UK, Australia, and most of Europe and the world.
The difference is mass vs molecule count, but because glucose has a known molar mass, the conversion is a single constant. The US held onto mg/dL because the medical system standardized on it before the SI shift; the rest of the world moved to mmol/L as part of metrication. Neither is more correct — they measure the same thing in different units.
The six reference ranges (ADA guidelines)
| Category | mg/dL | mmol/L | When |
|---|---|---|---|
| Normal | 70 – 99 | 3.9 – 5.5 | Fasting |
| Pre-diabetic | 100 – 125 | 5.6 – 6.9 | Fasting |
| Diabetic | ≥ 126 | ≥ 7.0 | Fasting |
| Normal | < 140 | < 7.8 | 2h post-meal |
| Pre-diabetic | 140 – 199 | 7.8 – 11.0 | 2h post-meal |
| Diabetic | ≥ 200 | ≥ 11.1 | 2h post-meal |
The fasting ranges are the diagnostic ones — a fasting glucose of 126 mg/dL (7.0 mmol/L) on two separate tests is the ADA threshold for type 2 diabetes. The 2-hour post-meal ranges are for the oral glucose tolerance test (OGTT), where you drink 75g of glucose and your blood is checked two hours later.
The tool also flags hypoglycemia (below 70 mg/dL / 3.9 mmol/L) — the "Low" category. This is the one to act on, not just convert. Below 54 mg/dL (3.0 mmol/L) is clinically significant hypoglycemia and requires immediate carbohydrate intake.
The HbA1c estimator (Nathan formula)
The tool's second tab estimates HbA1c from average glucose using the Nathan formula:
HbA1c (%) = (avg_mgdl + 46.7) / 28.7
This is the Nathan et al. 2008 formula, derived from the A1c-Derived Average Glucose (ADAG) study. HbA1c (glycated hemoglobin) measures the percentage of your hemoglobin that has glucose attached — it reflects average blood glucose over the past 2-3 months, because red blood cells live about 120 days. The formula lets you estimate what your next lab HbA1c might be from your CGM (continuous glucose monitor) average.
The HbA1c thresholds:
| Category | HbA1c |
|---|---|
| Normal | < 5.7% |
| Pre-diabetic | 5.7 – 6.4% |
| Diabetic | ≥ 6.5% |
The estimate is ±0.5 percentage points for most people, wider for those with hemoglobin variants (sickle cell trait, thalassemia) or conditions that affect red blood cell lifespan. The formula assumes normal red blood cell turnover — if your RBCs live shorter or longer than 120 days, your actual HbA1c will diverge from the estimate. A lab test is always more accurate.
Gotchas
- Use 18.0182, not 18. The shortcut introduces a 0.1% error per conversion. Small per reading, visible across a log of 300. The tool uses the precise factor.
- The direction matters. mg/dL → mmol/L is division. mmol/L → mg/dL is multiplication. Getting it backwards gives you 1620 instead of 90 — an order of magnitude off, and clinically nonsensical. The tool handles the direction from the unit toggle.
- Fasting vs post-meal ranges are different. 140 mg/dL is normal 2 hours after eating but pre-diabetic if fasting. The tool labels the category from the fasting thresholds; if you're checking post-meal, read the reference table, not just the label.
- Hypoglycemia is the one to act on. Below 70 mg/dL (3.9 mmol/L) the tool flags "Low" — this is not a number to convert and move on. Below 54 mg/dL (3.0 mmol/L) is clinically significant and needs carbohydrates now.
- The HbA1c estimate is ±0.5%, wider with hemoglobin variants. Sickle cell trait, thalassemia, and any condition that changes red blood cell lifespan will make your actual HbA1c diverge from the Nathan formula estimate. Trust the lab, not the formula, for diagnosis.
- The ADA and WHO thresholds differ slightly. The tool uses ADA guidelines (fasting ≥126 mg/dL = diabetic). The WHO uses the same fasting threshold but different impaired fasting glucose cutoffs (WHO: 110-125 mg/dL, ADA: 100-125 mg/dL). If you're outside the US, check which guideline your clinician uses.
- Home meters have ±15% accuracy. ISO 15197:2013 allows 95% of readings to be within ±15 mg/dL at ≥100 mg/dL and ±15% at <100 mg/dL. A 90 mg/dL reading could actually be 77-103. The tool gives you a precise conversion of an imprecise input — don't over-interpret single readings.
- This is informational, not diagnostic. The ranges and the HbA1c estimate are for understanding your numbers, not for diagnosing diabetes. A diagnosis requires two abnormal lab tests on separate days. Consult your healthcare provider.
Summary
- The conversion factor is 18.0182. mg/dL ÷ 18.0182 = mmol/L; mmol/L × 18.0182 = mg/dL. It comes from the molar mass of glucose (180.156 g/mol). Don't use 18 as a shortcut — the 0.1% error compounds across a log.
- Two units because the US uses mass (mg/dL), the rest of the world uses molecules (mmol/L). They measure the same glucose, different concentration scales. The single factor converts between them.
- Six ADA reference ranges: fasting normal 70-99 mg/dL (3.9-5.5 mmol/L), pre-diabetic 100-125 (5.6-6.9), diabetic ≥126 (≥7.0); post-meal normal <140 (<7.8), pre-diabetic 140-199 (7.8-11.0), diabetic ≥200 (≥11.1). Hypoglycemia is <70 mg/dL (3.9 mmol/L) — act on it, don't just convert it.
- HbA1c from average glucose: the Nathan formula
A1c = (avg_mgdl + 46.7) / 28.7estimates your 2-3 month average. Normal <5.7%, pre-diabetic 5.7-6.4%, diabetic ≥6.5%. The estimate is ±0.5% and wider with hemoglobin variants — trust the lab for diagnosis. - Home meters are ±15% accurate. A precise conversion of an imprecise reading is still imprecise. Don't over-interpret single values; look at trends.
- Convert at the Blood Sugar Converter; pair with BMI Calculator for weight-class context, Cholesterol Checker for the lipid panel, and Blood Pressure Analyzer for the cardiovascular picture.