What Low GI Atta Actually Means on Indian Labels

Last updated: 1 September 2026

Indian law permits a pack to say "Low GI" only where the measured glycemic index is below 55. That threshold is not advisory. It sits in Schedule I of the Food Safety and Standards (Advertising and Claims) Regulations, 2018, and it is the only glycemic claim the schedule defines.

Set that against India's own measurement. ICMR-NIN puts wheat chapatti at 65.66, comfortably above the line, so ordinary wheat atta does not qualify on the national nutrition authority's own figures. What pulls the number down is what gets blended into the dough.

What does a low GI claim legally require?

The 2018 claims regulations list permitted nutrition claims in a closed table. Item 16 reads, in full: glycemic index, low GI, GI value below 55.

That is the entire permission. There is no defined claim for medium GI, reduced GI, lower GI than a comparator, or very low GI. A pack making any of those is outside the schedule.

One detail deserves attention. The regulation defines GI against a reference food and lists four: glucose powder, dextrose, an oral glucose tolerance solution, or white bread. Glucose and white bread are not the same scale, and a value measured against white bread runs higher. The regulation permits either yet fixes one threshold without saying which scale the 55 refers to.

Where does ordinary wheat atta actually measure?

ICMR-NIN's own table appears in Dietary Guidelines for Indians. Each value is the mean of ten participants, calculated for 50 g of available carbohydrate.

Food	Glycemic index	Glycemic load	Source
Rice	78.23	49.38	ICMR-NIN, Annexure III
Wheat chapatti	65.66	32.83	ICMR-NIN, Annexure III
Mixed dhal	43.64	21.82	ICMR-NIN, Annexure III
Wheat with chana dhal and barley	39.27	19.63	ICMR-NIN, Annexure III
Wheat with chana dhal	32.37	16.18	ICMR-NIN, Annexure III

Plain wheat chapatti sits above the 55 threshold. Published chapati glycemic index values for wheat with chana dhal sit at half that, comfortably below.

ICMR-NIN attaches its own caution, noting the values are indicative and may differ with ingredient ratios and preparation. That caveat is part of the finding, not a disclaimer to skip.

What actually moves the number?

Blending moves it most, and by a wide margin. A trial in the British Journal of Nutrition measured whole wheat flour roti at a GI of 45.1 against 27.3 for roti made with an atta mix of roasted Bengal gram flour, psyllium husk and fenugreek, across eighteen subjects against a glucose reference.

Soluble fibre does the same. Chapatis made with barley beta-glucan, in a 2009 Nutrition Research trial, measured 54 without it and 30 and 29 at 4 g and 8 g, though the same study's 6 g arm came in at 45, so the dose-response is not clean.

Wheat variety is the smallest lever measured. Across seventeen named bread wheat varieties, glycemic index ranged from about 63 to 71, a spread of eight points, while blending and preparation move it by fifty.

Context matters as much. A study in the Indian Journal of Endocrinology and Metabolism found wheat chapatti in a North Indian mixed meal measured 85.5, statistically no different from the same meal built on rice.

Who checks a low GI claim before it reaches a shelf?

Nobody, in the sense most buyers assume. Because low GI is a defined claim in Schedule I, it is self-certified: prior approval is required only for claims the regulations do not already define.

The obligation is real but general. Regulation 4(6) requires a claim to be scientifically substantiated by validated methods of characterising or quantifying the substance behind it, but does not prescribe how. We searched the 2018 gazette and the current compendium for any testing standard, minimum subject count or accredited-laboratory requirement in a glycemic context, and found none.

Indian law fixes the threshold and the calculation basis while leaving the method open. That is the gap that makes a low GI claim on an Indian pack weakly policed rather than false.

Does making the claim change anything else on the pack?

It does, and this is the most useful consequence for a buyer. Under the Labelling and Display Regulations 2020, unprocessed single-ingredient products are exempt from nutritional labelling, but that exemption lifts the moment a nutrition or health claim appears.

A plain single-ingredient atta may therefore carry no nutrition panel. The same atta claiming low GI must carry the full panel: energy, protein, carbohydrate including sugars, fats and sodium. A pack claiming low GI while showing no nutrition panel is not following the labelling rules.

There is also a line the claim must not cross. Regulation 10 prohibits any claim about a food's suitability for preventing, alleviating, treating or curing a disease unless separately permitted. Low GI as a bare nutrient content claim is lawful; presenting it as suitability for diabetes is not.

How reliable is a single GI number anyway?

Less than two-digit precision suggests. Measuring white bread in 63 healthy volunteers across repeat challenges, the mean came out at 62 with a standard deviation of 15, and individual results ranged from 35 to 103. Within-person and between-person variation ran at 20% and 25%, and the authors concluded GI is unlikely to be a good approach to guiding food choices.

Measured Indian whole wheat chapatti values collected in a compendium of non-Western foods span roughly 45 to 84. A good part of that spread is measurement noise and human physiology rather than flour.

ICMR-NIN says as much about labels generally, noting on page 109 that although the food authority has strict norms, the information presented on labels could be misleading.

What about heritage and traditional wheat?

Here the measurement does not exist. We found no published human glycemic index value for khapli or emmer atta, none for Indian dwarf wheat, and none in the major GI databases.

The study usually cited for emmer measured no glycemic index at all. It tracked 22 diabetic subjects over six weeks and found no significant difference in fasting or post-prandial blood glucose, though it did find lipid changes. A heritage-wheat low GI claim is an assertion rather than a measurement.

SatRas publishes no glycemic index figure for its Sona Moti whole wheat atta, and on the evidence above no one selling an Indian heritage wheat could publish one honestly. What it does publish is covered in our piece on what Sona Moti wheat is, alongside choosing an atta for daily rotis. The range sits in the whole wheat atta collection.

What do buyers most often get wrong?

Four readings of the label do not hold.

• Reading "low GI" as diabetic-friendly. Indian regulations prohibit claiming a food is suitable for preventing, alleviating or treating a disease. Low GI describes one measured property, nothing more.
• Assuming the number is on the pack. A pack may lawfully say low GI without printing the value or naming the reference food.
• Choosing on wheat variety. Variety moves measured GI by about eight points; blending and preparation move it by fifty or more.
• Comparing figures across brands. Two makers may test against different permitted reference foods, so the numbers are not on one scale.

What this does not settle

The threshold question stays open. Indian law fixes 55 without resolving which reference food it applies to, and that ambiguity sits in the regulation, not in any one product.

Two other limits are worth stating. The ICMR-NIN values rest on ten participants each, normal for this work but small, and the variability research above suggests any single figure carries wide error. And no study we could locate measures one wheat prepared three ways under a single protocol.

This article is educational and is not medical advice. The glycemic index is one characteristic of food but alone does not determine its suitability or unsuitability for an individual. If you have diabetes or any other condition, it is always recommended that you consult a qualified physician or a registered dietitian.

Frequently asked questions

Can an atta pack legally say low GI in India?

Yes, provided the measured glycemic index is below 55. That value is set in Schedule I of India's advertising and claims regulations, and it is the only glycemic claim the schedule defines. Anything phrased as medium GI, reduced GI or lower GI than a rival product falls outside the permitted list entirely.

Is a low GI claim the same as diabetic friendly?

No, and treating them as equivalent crosses a legal line. Indian regulations prohibit claims about a food's suitability for preventing, alleviating, treating or curing a disease unless separately permitted. Low GI is a nutrient content claim describing one measured property. It is not a statement that a food suits any medical condition.

Does a low GI pack have to print the number?

Not under the nutrient content claim rules. A pack may lawfully say low GI without disclosing the measured value or the reference food used. What it must carry, once any nutrition claim appears, is the full nutrition panel, since making a claim removes the exemption a plain single-ingredient flour would otherwise have.

Why might two brands report different GI for similar flour?

Because the method is not fixed. The regulations permit several reference foods, including glucose and white bread, which produce different numbers for the same food, and they prescribe no subject count, protocol or accredited laboratory. Published research also shows wide variation between individuals, so two honest tests can genuinely disagree.

Does adding chana dal flour to atta lower the glycemic index?

Published Indian measurements point that way. National nutrition tables list wheat with chana dhal at 32.37 against 65.66 for plain wheat chapatti, and a separate trial measured roti made with an added legume and fibre mix at 27.3 against 45.1 for plain whole wheat. Preparation and ratios change the result.

Is heritage or khapli atta lower GI than regular wheat?

There is no published measurement to support that. No human glycemic index value appears for khapli, emmer or Indian dwarf wheat in the major databases or in the compendium covering non-Western foods. The trial usually cited measured no glycemic index and found no significant blood glucose difference over six weeks.