Data & method

Diagnostic thresholds

Two sets, published separately, shown separately. The American Diabetes Association values (fasting 100–125 for prediabetes, 126 and above for the diabetes range; two-hour 140–199 and 200 and above; HbA1c 5.7–6.4% and 6.5% and above) are taken from the ADA Standards of Care and cross-checked against the NIDDK's public restatement of them. The WHO values come from the 2006 WHO/IDF consultation, with the HbA1c position from the 2011 WHO consultation.

The two bodies agree on every diabetes threshold and disagree on exactly one number: the lower edge of impaired fasting glucose, which the ADA places at 100 mg/dL and WHO at 110 mg/dL. This site shows both rather than picking one, because a reader whose fasting result is 105 mg/dL deserves to know that the answer depends on which rulebook their laboratory follows.

Source: American Diabetes Association, Standards of Care in Diabetes — Diagnosis and Classification of Diabetes. Diabetes Care 2025;48(Suppl. 1):S27–S49. Link
Source: National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institutes of Health — Diabetes Tests & Diagnosis. Link
Source: World Health Organization / International Diabetes Federation. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: report of a WHO/IDF consultation. Geneva: WHO, 2006. Link
Source: World Health Organization. Use of glycated haemoglobin (HbA1c) in the diagnosis of diabetes mellitus: abbreviated report of a WHO consultation. Geneva: WHO, 2011. Link

Pregnancy

Gestational diabetes thresholds are from the WHO 2013 guideline on hyperglycaemia first detected in pregnancy: fasting 5.1–6.9 mmol/L (92–125 mg/dL), one hour 10.0 mmol/L (180 mg/dL) or above, two hours 8.5–11.0 mmol/L (153–199 mg/dL).

Source: World Health Organization. Diagnostic criteria and classification of hyperglycaemia first detected in pregnancy. Geneva: WHO, 2013. Link

HbA1c conversion

eAG (mg/dL) = 28.7 × HbA1c(%) − 46.7 · mmol/L = mg/dL ÷ 18.016 · mmol/mol = (HbA1c(%) − 2.15) × 10.929

The first is the linear regression published by the A1c-Derived Average Glucose study group in Diabetes Care in 2008, which is the equation the ADA uses; it reported r² = 0.84 across 507 participants. The third is the NGSP master equation. Every value on the HbA1c chart is computed at build time from these equations rather than transcribed, which is why the whole-percentage rows match the NGSP's own published table exactly (5% → 97 mg/dL, 7% → 154, 12% → 298).

Source: Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, for the A1c-Derived Average Glucose (ADAG) Study Group. Translating the A1C assay into estimated average glucose values. Diabetes Care 2008;31(8):1473–1478. Link
Source: National Glycohemoglobin Standardization Program (NGSP) — HbA1c and estimated average glucose (eAG). Link

Unit conversion

1 mmol/L of glucose = 18.016 mg/dL, from the molar mass of glucose (180.16 g/mol). The universal clinical shorthand of "divide by 18" is accurate to within a rounding error across the whole reporting range.

Glycemic index

Glycemic index cannot be calculated. It is measured by feeding volunteers a portion of a food containing 50 g of available carbohydrate and comparing their two-hour glucose response against pure glucose. Consequently this site publishes a glycemic index only where a peer-reviewed study measured one. There are currently 15 such values, from three studies:

• Twelve traditional Indian preparations from the Indian Journal of Medical Research, 2022 — millet rotis, idli, dosa, paratha, upma and two sundals, all against a glucose reference.
• Two values from a Journal of Food Science and Technology study of South Indian breakfast foods, 2022. That paper published glycemic index per food but not carbohydrate per food, so this site leaves the glycemic load blank for those two rather than borrowing a carbohydrate figure from a different dataset.
• Two multigrain roti formulations from Preventive Nutrition and Food Science, 2020, published here as a single entry that states both measured values.

Glycemic load is computed as GI × available carbohydrate in the stated portion ÷ 100, using the available-carbohydrate figure reported by the same study. Note that the source papers also report a "glycemic load" computed on the 50 g test portion; that figure is a property of the test protocol rather than of a meal, and this site does not reproduce it as if it were a serving.

Source: Shobana S, Geetha G, Ramya Bai M, et al. Carbohydrate profiling & glycaemic indices of selected traditional Indian foods. Indian J Med Res 2022;155(1):56–65. Link
Source: Shakappa D, Naik R, Sobhana PP. Glycemic carbohydrates, glycemic index, and glycemic load of commonly consumed South Indian breakfast foods. J Food Sci Technol 2022;59(9):3619–3626. Link
Source: Nagaraju R, Sobhana PP, Thappatla D, Epparapalli S, Kandlakunta B, Korrapati D. Glycemic index and sensory evaluation of whole grain based multigrain Indian breads (rotis). Prev Nutr Food Sci 2020;25(2):194–202. Link

Rice

Published GI values for rice span most of the scale — 64 to 93 for white polished rice, 39 to 79 for parboiled, 50 to 96 for basmati, 51 to 87 for brown — so this site publishes the ranges from a systematic review and refuses to print a single number for any rice.

Source: Boers HM, Seijen ten Hoorn J, Mela DJ. A systematic review of the influence of rice characteristics and processing methods on postprandial glycaemic and insulinaemic responses. Br J Nutr 2015;114(7):1035–1045. Link

Food composition

Carbohydrate, fibre, protein and energy per serving come from the Indian Food Composition Tables 2017 (ICMR–National Institute of Nutrition) via the dataset shared with our sister site ThaliFacts, with standard household serving sizes. The dataset's carbohydrate figure is treated as available carbohydrate; if a particular entry were total carbohydrate instead, the true available figure would be slightly lower.

Source: Longvah T, Ananthan R, Bhaskarachary K, Venkaiah K. Indian Food Composition Tables 2017. National Institute of Nutrition, ICMR, Hyderabad. Link

Rounding

Glucose to whole mg/dL and one decimal in mmol/L, matching how laboratories report. HbA1c to one decimal, mmol/mol to whole numbers. Glycemic index and glycemic load to one decimal, which is the precision the source papers use.

What this site deliberately does not publish

• No diagnosis, and no diagnostic language. Thresholds are described as ranges results fall into, never as verdicts.
• No treatment targets. Guidelines do publish glycemic targets for people already diagnosed, and they are individualised by a doctor. Printing them beside a public converter would invite people to apply someone else's target to themselves, so this site prints none at all.
• No medicines, no doses, no supplements, no herbal remedies, no "reverse your diabetes".
• No claim that any food treats, prevents or cures diabetes.
• No age-banded normal ranges, because no health authority publishes any.
• No estimated glycemic index values. A guessed GI is a fabricated measurement.
• No medical reviewer. This site has not been reviewed by a clinician, so it does not display a reviewer, a credential or a "medically reviewed" badge. Several sites in this subject display them without substance; the absence here is deliberate and is the point.

Corrections

If a figure here disagrees with the primary source, the primary source wins and the page is fixed. Send the URL.