Which Atta Is Best for Diabetes? What the Evidence Says
We first published this in January 2023. That version stated that whole wheat atta has a low glycemic index. That was not accurate, and we have corrected it.
Published GI values for whole wheat roti range from roughly 45 to 70 depending on the study, which spans the entire low, medium, and high classification. Calling it "low" picked the friendliest end of a very wide range and presented it as settled.
The rewritten article below explains what the evidence actually supports, what it does not, and why almost every number you will read on this subject disagrees with every other number. Sources are listed at the foot.
Nutrition · Evidence review
There is no best atta for diabetes, and the honest reason is uncomfortable: the measurement everyone is arguing about is not precise enough to settle the argument.
Search this question and you will find confident answers giving whole wheat roti a glycemic index of 45, or 55, or 62, or 70. Those are not small differences. They straddle the boundaries that decide whether a food is called low, medium, or high. Someone is wrong, and it is worth understanding why before you change what you eat.
And the part that no atta brand will tell you: switching flour is not a treatment for diabetes. It is a small adjustment inside a much larger picture of portion, the rest of the plate, medication, activity, and sleep.
- What glycemic index actually measures
- Why every atta number online disagrees
- The reproducibility problem
- Why glycemic load matters more
- What the evidence does support
- What ICMR-NIN actually recommends
- The only response that counts is yours
- When this is a doctor's question
- Frequently asked questions
- Sources
What Glycemic Index Actually Measures
Worth being precise, because most of the confusion starts here.
GI ranks a food by the blood glucose response it produces relative to a reference, usually pure glucose set at 100. Under the international standard, ISO 26642:2010, testing requires at least ten healthy adults, and the test portion must contain either 25 or 50 grams of available carbohydrate, with blood sampled at set intervals over two hours [1].
Two things follow from that definition, and both matter enormously.
- The portion is fixed, and it is large. To get 50 grams of available carbohydrate from atta you need roughly four to five medium rotis, eaten alone, on an empty stomach. That is not how anyone eats.
- It is tested in healthy adults, not people with diabetes. The standard specifies subjects with normal glucose tolerance. So the number was not generated in the population most interested in it.
- It is a single food, eaten alone. Not roti with dal, sabzi, and curd, which is what actually reaches your table.
The classification thresholds in common use are 55 or below for low, 56 to 69 for medium, and 70 or above for high.
Why Every Atta Number Online Disagrees
Here is the actual state of the published evidence for whole wheat roti.
| Reported GI | Source | Classification it implies |
|---|---|---|
| 45.1 | Peer-reviewed trial, 18 healthy adults, GI testing laboratory at the Madras Diabetes Research Foundation, published in the British Journal of Nutrition [2] | Low |
| Around 62 | The figure most commonly repeated across Indian health and clinical websites | Medium |
| Around 70 | Cited by companies selling alternative flours | High |
That is a 25-point spread on a 100-point scale, for the same food, crossing every category boundary there is.
The Reproducibility Problem
This is the section that actually answers your question, and it is why we will not give you a ranking.
Researchers at the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University ran a controlled feeding trial in 63 healthy adults, a sample far larger than the ten the standard method requires. They found the GI value of a given food varied by an average of about 20 percent within the same individual and about 25 percent between individuals. The authors concluded that GI has limited usefulness for predicting how a food affects an individual's blood glucose [3].
An earlier study in Diabetes Care tested commercial white bread in the same people on three separate occasions. The three GI determinations came out at 78, 60, and 75. Within-individual variation was around 43 percent [4].
To be fair to GI, this is contested. Other work, including analysis of a database of over 1,000 foods, has reported substantially lower within-subject variability, around 17 percent, and argues GI ranks foods usefully [5]. The honest summary is that GI is a real and useful concept at the level of broad categories, and unreliable at the level of small differences between similar foods.
Why Glycemic Load Matters More
Even if the GI number were perfectly reliable, it would not answer your question, because it deliberately ignores portion.
Glycemic load accounts for both the GI and the amount of carbohydrate actually in the serving you eat. It is the more practical of the two, and it points somewhere unglamorous: how many rotis, not which flour.
- Six rotis of a lower-GI atta can produce a larger glucose load than three rotis of ordinary atta. The flour change is undone by the portion change, and portion is the variable you control most easily.
- A ten-point GI difference is small next to a doubled portion. If you are choosing between optimising your flour and optimising your serving size, the serving size is not close.
- Switching atta and eating more of it is a common and understandable trap. Healthier-sounding food invites larger portions.
What the Evidence Does Support
Having pulled the ground out from under the question, here is what is genuinely well established.
Whole grain beats refined
This part of the original article was right. Whole wheat atta retains the bran and germ, and therefore fibre and micronutrients, which refined flour has had removed. The direction of that comparison is not seriously disputed. Whole grain over maida is a sound default.
What is not supported is the precision. "Whole wheat is better than maida" is defensible. "Whole wheat is low GI" is not.
What you eat with it changes the response
The glycemic response to a meal is not the response to its highest-GI component. Protein, fat, and fibre eaten alongside carbohydrate slow gastric emptying and blunt the rise. Roti with dal, sabzi, and curd is a different metabolic event from roti alone, and it is how most Indian meals are already eaten.
This is one reason single-food GI values transfer poorly to real plates.
Total carbohydrate and total energy matter
Across the population level, high consumption of refined cereals is associated with type 2 diabetes in Indians, and white rice intake above 450 grams a day has been linked with increased prevalence [6]. The signal is about quantity and refinement, not about which brand of atta is in the container.
Chakki versus commercial, multigrain versus wheat
You will find confident claims that coarser chakki atta has a lower GI than finely milled commercial atta, and that multigrain beats wheat. There is a plausible mechanism, since particle size and fibre content affect digestion rate, and some studies do report lower GI for multigrain rotis than whole wheat rotis [7].
But the differences claimed are usually within the noise described above, and much of what circulates is not from testing that meets the ISO standard. We are not going to give you numbers we cannot stand behind. If a brand claims a specific GI, ask which laboratory tested it, in how many subjects, and whether the method met ISO 26642.
What ICMR-NIN Actually Recommends
India's authoritative guidance is the Dietary Guidelines for Indians, published by the ICMR National Institute of Nutrition in May 2024, the first revision in thirteen years [8].
It does not tell you which atta to buy. It recommends drawing food from a minimum of eight food groups, with vegetables, fruits, greens, roots, and tubers making up roughly half the plate, and cereals and millets forming the other major portion, followed by pulses, nuts, oilseeds, and milk or curd. It restricts added sugar to under 5 percent of daily energy, and cautions against ultra-processed foods.
The Only Response That Counts Is Yours
Here is the practical consequence of everything above. Since GI is a population average with substantial variation between people, the average tells you very little about you specifically.
The way to find out how your body handles your roti is to measure your own response. A continuous glucose monitor does exactly that, and for someone with pre-diabetes or type 2 diabetes it will answer this question more usefully in two weeks than any table of GI values ever will. It will also tell you things a table cannot: how your response changes with portion, with what you eat alongside it, with the time of day, and with how you slept.
When This Is a Doctor's Question
We are coaches, not doctors or dietitians, and diabetes is one of the conditions where nutrition is genuinely part of medical treatment rather than general lifestyle advice.
- You are on insulin or any glucose-lowering medication. Changing your carbohydrate intake can change your medication requirement, and that adjustment belongs to your doctor.
- You have kidney involvement. Diabetic kidney disease changes the nutrition rules substantially, and general dietary advice can be actively wrong.
- You have not had bloodwork recently. HbA1c reflects roughly two to three months of average glucose and is the number worth tracking. Guessing from a flour label is not a substitute.
- You want a plan rather than a principle. A registered dietitian who can see your reports will do more for you than any article, including this one.
Frequently Asked Questions
Which atta is best for diabetes?
No atta is best, and the honest reason is that glycemic index is not precise enough to rank flours that differ by a few points. Published GI values for whole wheat roti range from roughly 45 to 70 depending on the study, which crosses every classification boundary. What is well supported is that whole grain beats refined flour, that portion size matters more than flour type, and that what you eat alongside the roti changes the response substantially. Speak to your doctor or a registered dietitian about your own situation.
Is whole wheat atta low glycemic index?
Not reliably. One peer-reviewed Indian trial measured whole wheat flour roti at 45.1, which is low. The figure most commonly repeated across Indian health sites is around 62, which is medium. Companies selling alternative flours often cite around 70, which is high. All three cannot be right, and the spread reflects genuine variability in how GI is measured rather than one source simply being dishonest.
Is chakki atta better than commercial atta for diabetics?
There is a plausible mechanism, since coarser particle size and higher bran content can slow digestion, and you will find confident claims to this effect. But the differences typically claimed are smaller than the measurement noise in GI testing, and much of what circulates has not been tested to the ISO 26642 standard. If a brand claims a specific GI, ask which laboratory tested it, in how many subjects, and by what method.
Is multigrain atta good for diabetes?
Some studies do report lower glycemic responses for multigrain rotis compared with whole wheat rotis. The direction is plausible and the evidence is reasonable, but the size of the benefit is frequently overstated in marketing, and "multigrain" is not a regulated term, so the composition varies enormously between products. Read the ingredient list rather than the front of the bag.
Why do different websites give different GI values for roti?
Because GI is genuinely variable. A controlled trial in 63 adults at Tufts found the GI of a given food varied by about 20 percent within the same person and 25 percent between people. An earlier Diabetes Care study testing white bread three times in the same people returned 78, then 60, then 75. Add real differences in atta type, particle size, and cooking, and a 45 to 70 spread stops being surprising.
How many rotis can a diabetic eat?
That depends on your medication, your activity, your total daily carbohydrate, and your own glucose response, which is why no article can answer it responsibly. It is a question for your doctor or a registered dietitian who can see your reports. What is worth knowing is that portion is a larger lever than flour type: six rotis of a lower-GI atta can produce a bigger glucose load than three rotis of ordinary atta.
What actually helps more than changing atta?
Portion size, what you eat alongside the roti, total daily carbohydrate, and everything outside the plate: medication taken as prescribed, physical activity, sleep, and regular bloodwork. India's own dietary guidance from ICMR-NIN does not name a best flour. It describes a plate, with vegetables and fruits making up roughly half of it. That is where the evidence is strong.
Sources
- ISO 26642:2010. Food products, determination of the glycaemic index (GI) and recommendation for food classification. International Organization for Standardization.
- Radhika G, Sumathi C, Ganesan A, Sudha V, Henry CJK, Mohan V. Glycaemic index of Indian flatbreads (rotis) prepared using whole wheat flour and 'atta mix'-added whole wheat flour. British Journal of Nutrition. 2010. GI of whole wheat flour roti reported as 45.1 (SEM 3.5) in 18 healthy non-diabetic subjects. PubMed
- Matthan NR, Ausman LM, Meng H, Tighiouart H, Lichtenstein AH. Estimating the reliability of glycemic index values and potential sources of methodological and biological variability. American Journal of Clinical Nutrition. 2016. Controlled feeding trial, 63 healthy adults. AJCN
- Vega-López S, Ausman LM, Griffith JL, Lichtenstein AH. Interindividual variability and intra-individual reproducibility of glycemic index values for commercial white bread. Diabetes Care. 2007;30(6):1412-1417. Diabetes Care
- Brand-Miller J, Stockmann K, Atkinson F, Petocz P, Denyer G. Glycemic index, postprandial glycemia, and the shape of the curve in healthy subjects: analysis of a database of more than 1000 foods. American Journal of Clinical Nutrition. 2009. Reports lower within-subject variability of around 17%.
- Association between high white rice consumption and type 2 diabetes prevalence in Indian populations, as discussed in analyses of the Dietary Guidelines for Indians 2024.
- Glycemic index and sensory evaluation of whole grain based multigrain Indian breads (rotis). Journal of Food Science and Technology. Reviews evidence that multigrain rotis show lower glycemic responses than whole wheat rotis.
- ICMR-National Institute of Nutrition. Dietary Guidelines for Indians. Released May 2024. ICMR-NIN
- Atkinson FS, Brand-Miller JC, Foster-Powell K, Buyken AE, Goletzke J. International tables of glycemic index and glycemic load values 2021: a systematic review. American Journal of Clinical Nutrition. 2021;114(5):1625-1632. The standard reference compilation of GI values.
About the Author
Niraj Kumar Borah
Founder and head coach of Fitness Bootcamp, a four-week residential health transformation programme in Rishikesh, India. Since 2020 he has guided more than 4,600 guests through structured, fully supported transformations.
His coaching is biomarker-driven, built from bloodwork, body composition, and recovery data. He is certified in nutrition coaching through Precision Nutrition, and is a coach rather than a doctor or dietitian, which is why this article ends by telling you to see one.
- Nutrition: Precision Nutrition Level 1.
- HYROX: HYROX Academy Level 1 certified, Creating Athletes. Affiliated HYROX Performance Coach, listed in the public directory.
- Conditioning and running: Bioforce Conditioning Coach, VDOT Certified Running Coach.
- Heart rate: NESTA Certified Heart Rate Performance Specialist.
Want to know how your body handles your roti?
Not from a table of averages. From your own glucose data, alongside your doctor's guidance. Tell us where you are starting from.
Message the team on WhatsAppThis article is general educational information about the evidence on glycemic index and is not medical or dietetic advice. It is not a recommendation to change your diet, your medication, or your management of any condition. Diabetes is a medical condition in which nutrition forms part of treatment, and dietary changes can affect medication requirements. Please speak with your doctor or a registered dietitian before making changes. Figures quoted are from the sources listed above and were checked in July 2026. Where the evidence is contested, we have said so rather than choosing the most convenient number. If you believe any figure here is inaccurate, please contact us with the source and we will correct it.
