What samai gives you that rice does not
Samai is the small dull grain sold here as samai arisi, and as ചാമ അരി on the packets that cross from the Kerala side. It is not a Kanyakumari crop — this is paddy country — so it arrives from the interior districts, sown on rainfed land in Dharmapuri, Salem, Erode, Tirunelveli and Thoothukudi in June–July and again in September–October (TNAU Agritech). What it brings to a diabetic plate is measurable rather than magical. Weighed raw against the milled rice it replaces, samai carries 13 g less available carbohydrate, 2 g more protein and nearly three times the total dietary fibre (IFCT 2017). It also carries far more magnesium — 91 mg against rice’s 19 mg. Two things it does not carry are ragi’s calcium and ragi’s potassium, and the second of those turns out to be an advantage in the right patient.
| Per 100 g, raw grain | Samai (little millet) | Varagu (kodo) | Ragi (finger) | Milled white rice |
|---|---|---|---|---|
| Available carbohydrate | 65.6 g | 66.2 g | 66.8 g | 78.2 g |
| Total dietary fibre | 7.7 g | 6.4 g | 11.2 g | 2.8 g |
| Protein | 10.1 g | 8.9 g | 7.2 g | 7.9 g |
| Calcium | 16 mg | 15 mg | 364 mg | 7 mg |
| Potassium | 105 mg | 94 mg | 443 mg | 108 mg |
| Energy | 346 kcal | 332 kcal | 321 kcal | 356 kcal |
All figures IFCT 2017 (ICMR-NIN), food codes A016, A017, A010 and A015; energy converted from kilojoules. Compare the grains side by side on the millet glycemic index chart.
Is the glycemic index of samai actually low?
The packets say low. The best Indian measurement says otherwise, and a patient deserves to hear it. In 2020 the Madras Diabetes Research Foundation in Chennai fed twelve healthy volunteers a portion of cooked, unpolished little millet containing 50 g of available carbohydrate — 171 g of cooked grain — and measured a glycemic index of 88.6 ± 5.7 against a glucose reference. The foxtail millet tested alongside it came out the same. Their conclusion was that substituting millet for white rice may be of limited benefit on glycemic grounds alone (Indian J Med Res, 2020). Set against that, an older study of little-millet flakes reported a GI of 52.11 and a low glycemic load of 9.24 (J Food Sci Technol, 2015) — flakes, not cooked grain, which is why that figure and this one can both be true. And the largest pooled analysis of millet trials put millets as a class in the mid-50s against 71.7 for milled rice, with exactly one grain whose data were inconsistent: little millet (Anitha et al., Frontiers in Nutrition, 2021). Read together, samai is not a low-GI grain. It is a grain whose number depends almost entirely on what was done to it before it reached you.
How much samai can a sugar patient eat?
Do the arithmetic and the answer stops being controversial. In the Chennai test, 171 g of cooked little millet supplied 50 g of available carbohydrate. A normal cup of cooked samai, roughly 150 g, therefore carries about 44 g of carbohydrate — an estimate, but a sound one — which is very close to what a similar cup of cooked rice carries. That is the whole argument in one line: cup for cup, samai is not a smaller carbohydrate load than rice. So the portion rule for samai is the rice rule, not a relaxed version of it. One cup cooked at a meal for most adults; less if you are small, sedentary, or sitting above your HbA1c target. And it must replace the rice at that meal. The commonest mistake I see in clinic is the patient who eats samai upma at eight in the morning, then the usual full rice lunch and the usual rice dinner, and cannot understand why nothing has moved. The same trap catches people with red rice and brown rice.
What to say at the shop, and what to check on the packet
In a Nagercoil or Marthandam provision store, ask for it plainly: “பாலிஷ் இல்லாத சாமை இருக்கா?” — polish illaadha samai irukkaa? Across the border at Kaliakkavilai or Parassala: “പോളിഷ് ചെയ്യാത്ത ചാമ ഉണ്ടോ?” — polish cheyyaatha chama undo? Expect no more often than yes. When the Chennai market was surveyed properly across 100 outlets, 40 of 43 little-millet brands — 93% — were polished, and only three were not (Indian J Med Res, 2020).
So do the label test yourself, because the shopkeeper often does not know. Turn the packet over and read the dietary fibre line. Whole samai runs about 7.7 g per 100 g (IFCT 2017); that same survey found brands declaring 1–2 g, which is white rice wearing a millet’s name. Anything under roughly 4 g per 100 g has had its bran taken off. Then look at the grain itself: unpolished samai is dull and greyish-brown with the bran still on it, while polished samai is chalk-white, shiny, and cooks in the time rice does. A 2024 study from the same Chennai group with ICAR-IIMR Hyderabad found that debranning stripped fibre, fat, minerals and phytate from all five small millets while raising available carbohydrate, and its co-author put it plainly: polished millets are “fibre and nutrient depleted yet sold at a premium” (Shobana Shanmugam, in Discovery Food, 2024).
Samai sadam, samai upma — or samai payasam?
The forms that work here are the savoury ones, and this district already cooks all of them. Samai sadam — plain cooked samai eaten with sambar, rasam, a poriyal and curd — is the straight rice replacement, and the easiest switch for a family that will not tolerate a different breakfast. Samai upma built on real vegetable, drumstick, beans, carrot, with a spoon of roasted gram or peanut, spreads the load further. Samai idli and dosa, fermented overnight with urad dal, bring protein into the same plate and are the version most of my patients actually keep up for years. Samai thayir sadam is fine. What does not work is the sweet drift: samai payasam and samai sakkarai pongal are jaggery dishes with a millet in them, and it is the jaggery your meter is reading — the same arithmetic as sakkarai pongal. Thin samai kanji carries the problem of every thin kanji, liquid starch that empties fast (kanji and diabetes). Millet biscuits and millet health-mix powders are biscuits and powders.
Who should take extra care with samai
Three groups. First, anyone on insulin or a sulfonylurea who suddenly swaps a rice meal for a smaller samai one can run low — samai is not the cause, the reduced carbohydrate load is, and the answer is to test more often in the first week and tell whoever prescribes for you. Your doctor adjusts the dose; never adjust it yourself (recognising a low). Second, patients with reduced kidney function are usually warned off ragi, and rightly — 443 mg of potassium per 100 g. Samai sits at 105 mg, essentially the same as milled rice (IFCT 2017), which often makes it the easier millet in chronic kidney disease; still ask the doctor managing your kidneys before it becomes daily. Third, anyone who has been sold samai as a treatment. No grain treats diabetes. Millet eaters in the pooled analysis did see HbA1c fall from 6.65% to 5.67% over months, but those were mixed millets replacing refined staples across a whole changed diet, with weight loss and walking alongside — not one grain acting on its own. Test your own two-hour reading after your usual samai plate; that single number is worth more than any packet claim.
Sources
- IFCT 2017 (ICMR-NIN), Indian Food Composition Tables — food codes A016 (samai), A017 (varagu), A010 (ragi), A015 (milled rice). Every composition figure above comes from this book, energy converted from kilojoules. Note that IFCT still files samai under the older botanical name Panicum miliare.
- Malavika M, Shobana S, Vijayalakshmi P, et al., Indian J Med Res 2020;152(4):401–409 — the Chennai market survey and the glycemic index of cooked unpolished little and foxtail millet. The single most important paper on this page, and the one the marketing never quotes.
- Anitha S, et al., Frontiers in Nutrition 2021;8:687428 — the pooled analysis of millet trials. Useful for millets as a class, and honest that little millet’s own data were inconsistent.
- Patil KB, Chimmad BV, Itagi S., J Food Sci Technol 2015;52(9) — the GI of 52.11 that packet copy quotes; measured on flakes, not cooked grain.
- ICAR-IIMR / MDRF debranning study, Discovery Food 2024;4:136, reported by Nature Index — what polishing removes from five small millets.
- TNAU Agritech portal, samai crop production — sowing seasons, growing districts and the CO and Paiyur varieties.
Related questions
Is ragi good for diabetes? · Is varagu (kodo millet) good for diabetes? · Is thinai (foxtail millet) good for diabetes? · Is kambu (pearl millet) good for diabetes? · Millet glycemic index chart · Guess the GI quiz