That is not bad news. It means the rules you learned in earlier chapters travel with you. In Chapter 5 you saw that both the type and the amount of carbohydrate matter. In Chapter 6 you saw that the carbohydrate quarter of your plate keeps the amount in check. This chapter answers the practical follow-up questions: what does that quarter contain in your kitchen, how much carbohydrate is actually in it, and how do you keep the foods, celebrations, and fasts that make up your culture without giving up control of your glucose?

We start with the staples themselves and what the research says about them. We then turn to measurement: how to judge portions with household measures and simple food tables. We finish with the moments when food is about more than food: festivals, shared meals, and fasting.

7.1 The World's Staples

Six families of carbohydrate foods

Most of the world's carbohydrate comes from a small number of food families. Rice, wheat (bread, roti, noodles, pasta), and maize (tortillas, ugali, polenta, grits) are the three grain giants. Roots, tubers, and plantain (potato, sweet potato, cassava, yam, taro, green plantain) fill the same role in much of Africa, the Pacific, and Latin America. Millets, sorghum, and teff are older grains that anchor cooking in the Sahel, East Africa, and India. And pulses (lentils, beans, chickpeas, peas) sit at the border between carbohydrate and protein, which is why the plate in Chapter 6 gave them a special note.

No family is forbidden, and none is magical. What separates them for someone with diabetes is how much carbohydrate a portion contains, how much fibre comes with it, how refined it is, and how much of it you eat. The research is worth knowing for each.

Rice: the largest staple, and a careful reading

Rice is the daily staple of billions of people, and it is the food people most often ask about. A meta-analysis of seven prospective cohort studies, with 352,384 participants and 13,284 new cases of type 2 diabetes, found that higher white rice intake was associated with a higher risk of type 2 diabetes: 11 percent higher for each additional daily serving. The association was stronger in Asian populations, who eat far more rice (a relative risk of 1.55 for the highest versus the lowest intake) than in Western populations (1.12, with a confidence interval that included no effect).[@hu2012]

Read that carefully. These are observational studies, and they show association, not proof that rice caused diabetes. Independent experts pointed out that people who eat a lot of rice may differ in body size, wealth, and other habits, and cautioned against concluding that switching to brown rice would prevent diabetes.[@smc2012] What the data do support is a practical message: in populations where rice is eaten several times a day, it is worth watching the size of the rice portion and the variety and preparation of the rice (Chapter 5), and pairing it with plenty of vegetables and protein (Chapter 6). You do not need to give up rice.

Whole grains: the strongest consistent signal

Across foods, a pattern that keeps appearing is that whole (intact, minimally milled) grains do better than refined ones. A dose-response meta-analysis of 16 cohort studies found that each three servings a day of whole grains was associated with a lower risk of type 2 diabetes (relative risk 0.68), whereas refined grains showed no protective association (0.95).[@aune2013] These, too, are observational data, but they agree with the European guidance we saw in Chapters 5 and 6, which favours intact whole grains over finely milled ones and asks for at least 35 grams of fibre a day.[@easd2023] A useful rule across cultures: choose the least processed version of the grain your family already eats, such as whole wheat or whole-meal flour for roti and bread, hand-pounded or less-polished rice where available, whole maize meal, and whole-grain millet, sorghum, or oats.

Millets, sorghum, and teff

Traditional grains that are now less common than rice and wheat are getting fresh attention. A systematic review and meta-analysis of studies on millets, drawing on 65 studies with roughly 1,000 participants in total, reported that people with diabetes who ate millets had 12 to 15 percent lower fasting and after-meal glucose, and that the average glycemic index of the millets studied was about 52.7, roughly a third lower than milled rice and refined wheat.[@anitha2021] A later meta-analysis of 12 randomised trials likewise found lower fasting and post-meal glucose with millets, but no significant effect on HbA1c (only two short studies had measured it), and it noted limitations: 11 of the 12 studies were in India or Sri Lanka, samples were small, and there was some risk of publication bias.[@anitha2024] So millets are a promising, low-glycemic swap to try, especially for those who already eat them, but the evidence is early and geographically narrow. This chapter does not cite comparable trial evidence for teff and sorghum; enjoy them as whole grains, and measure the portion.

Roots, tubers, and plantain

Potato, sweet potato, yam, cassava, taro, and green plantain are starchy vegetables, and in the plate method they belong in the carbohydrate quarter, not the vegetable half.[@adaplate] Because they are watery, a cooked cup contains less carbohydrate than the same weight of a dry grain, but not so little that you can treat them as free foods. Their carbohydrate varies, as the table below shows, and cooking method and variety change their glycemic effect (Chapter 5). What you can say is that a boiled root eaten as a modest portion, with vegetables and protein, is a very different meal from a large mound of fried roots.

Maize

Maize is eaten as flatbreads and stiff or soft porridges under many names: tortillas, arepas, ugali, nshima, sadza, pap, polenta, grits. The same rules apply: measure the portion, choose whole-grain or minimally milled forms where you can, and add vegetables and protein. In the USDA data, a 28-gram corn tortilla contains about 12.7 grams of carbohydrate and 1.8 grams of fibre, so two small tortillas provide about 25 grams of carbohydrate.[@usda2026] For the thick porridges, which are dense and easy to over-serve, use a measuring cup or a small bowl for a while until your eye learns the size.

Pulses

Beans, lentils, chickpeas, and peas contain carbohydrate, but they come packaged with a lot of fibre and protein. A meta-analysis of 41 randomised trials with 1,674 participants found that pulses reduced fasting blood glucose and fasting insulin when eaten on their own, and that in trials of pulses within low-glycemic-index or high-fibre diets, a combined measure of longer-term glucose control (HbA1c or fructosamine) improved by a small amount. The results varied considerably between studies.[@sievenpiper2009] Pulses are one of the most useful additions in any cuisine: dal, ful, feijão, frijoles, hummus, misir wot, and bean soups all count.

What the numbers look like

The table below shows the carbohydrate and fibre in 100 grams of some cooked staples, from the USDA's food composition database, and, in the last column, how much cooked weight it takes to provide 30 grams of carbohydrate. The final column was calculated from the carbohydrate values and rounded to the nearest 5 grams.

Values are from the USDA Standard Release reference data for the foods named, weighed cooked. Your own rice, yam, or lentils may differ with variety, cooking time, and water. USDA "carbohydrate" includes fibre.[@usda2026]

{{fig:F7-A}}

Science Corner: Why cooked and raw weights are so different Dry grains and pulses soak up water when they cook, and roughly double or triple in weight. The carbohydrate does not change, but it is spread through much more weight. That is why 100 grams of dry rice contains far more carbohydrate than 100 grams of cooked rice, and why nutrition tables always say which state a food is in. If you weigh your food, be consistent: always weigh it cooked, or always weigh it dry, and use the table entry that matches. Cooked-weight values, as above, are usually the easiest for a home kitchen.

{{fig:F7-B}}

7.2 Measuring Your Quarter

Three levels of precision

You do not need the same precision for every meal, or from every person. Think of three levels, and choose the one that fits your situation.

Level 1: the plate. Fill half the plate with vegetables and no more than a quarter with carbohydrate foods, as in Chapter 6. This is enough for many people, and in a trial in China it worked at least as well as carbohydrate counting.[@zhang2022]

Level 2: household measures. Use tools you already own to make the quarter consistent from day to day: a measuring cup or a small bowl for rice and porridge, a serving spoon or a ladle for stews and dal, the number of tortillas, rotis, or slices of bread. Use the same bowl every time. Within a week, you will know what your usual portion looks like, and you will notice when it drifts.

Level 3: grams of carbohydrate. Weigh or measure the food, look up its carbohydrate content, and add it up. This is carbohydrate counting. Some people, especially those who take insulin at meals, are taught to do it because the dose may depend on it; if that is you, your care team or dietitian will show you how. Others use it for a few weeks to calibrate their eyes, then return to level 1 or 2.

Where to look up numbers

Good food composition tables exist for most of the world. The USDA's FoodData Central is the largest, and the FAO's International Network of Food Data Systems (INFOODS) lists national and regional tables for Africa, Asia (including separate tables for Indonesia, India, China, Japan, and many other countries), Latin America, the Middle East, Europe, and North America.[@infoods2026] Where possible, use a table from your own country, because it will include your local staples and dishes. The Indonesian composition table (TKPI) is one such example. A dietitian can also help you build a short personal list of your ten most-eaten foods and their portions, which is far more useful than a general table.

Ways to keep it practical

Three habits make this manageable. First, learn your top ten foods rather than a database of thousands. Second, remember that mixed dishes vary, since the carbohydrate in a stew, a fried rice, or a filled dumpling depends on the recipe; treat any number for them as an estimate and check how your glucose responds. Third, remember that tables give averages: your rice will not match the table's rice exactly, and that is fine. The goal is consistency, not perfection.

The table in section 7.1 also helps with swapping. If a meal uses about 105 grams of cooked rice (about 30 grams of carbohydrate), the same 30 grams in lentils would be about 150 grams, and it would come with about 12 grams of fibre instead of about 0.4. Trading part of a starch portion for pulses is one of the easiest ways to change quality without changing the amount of carbohydrate.

Science Corner: Do I need to count carbohydrate? It depends on your treatment. For many people with type 2 diabetes who are managing with food and possibly tablets, the plate and household measures are enough, and a plate-method trial found results at least as good as counting.[@zhang2022] For people who use mealtime insulin, counting the grams of carbohydrate is often central to dosing, and needs to be taught by a diabetes team. If you are unsure which group you are in, ask your care team what level of precision they want from you, and whether your plan includes a carbohydrate target.

7.3 Food, Culture, and Fasting

Keeping your food culture

Food is identity, memory, and belonging. People with diabetes can feel pressure to abandon their traditional foods and eat a bland imported diet, which is neither necessary nor sustainable. Many traditional cuisines already contain what this book recommends: vegetables, legumes, whole grains, fermented foods, and fish. What needs to change is often the proportion (more vegetables, a smaller starch mound), the refining of the starch, the sugar and frying, and the size of the portion.

Some simple adaptations work in nearly any cuisine. Serve a larger share of vegetables and pulses alongside the staple. Choose the whole-grain or less-milled version of your usual starch when you can. Use less sugar in drinks, sweets, and sauces. Bake, steam, grill, or braise more often than you deep-fry. Keep the celebratory dish, but let it be one part of the meal and not the whole plate.

Celebrations and shared meals

Weddings, Eid, Diwali, Lunar New Year, Christmas, funerals, harvest festivals, and family Sundays revolve around food, and refusing it can feel like refusing the people. A few habits help. Eat a normal meal earlier in the day and do not arrive starving. Start with vegetables, salad, or soup. Take a small serving of the special dishes you most look forward to instead of a little of everything. Drink water between sweet drinks. Walk after the meal if you can. If you use insulin or medicines that can cause lows, check your glucose as you usually would and keep your usual treatment plan; ask your care team beforehand about big feasts.

Fasting, and Ramadan in particular

Many traditions include fasting, and Ramadan is the largest. The International Diabetes Federation and the Diabetes and Ramadan (DAR) International Alliance publish practical guidelines for people with diabetes. They are clear that fasting is not recommended for everyone: people who are ill or have certain medical conditions, including some people with diabetes, can be exempted. Those at particular risk include people with poorly controlled diabetes, other conditions such as cardiovascular or kidney disease, pregnant women with diabetes, older people, and children. The guidelines recommend a medical assessment six to eight weeks before Ramadan begins, so that you and your team can decide whether to fast and adjust your plan.[@idfdar2021]

Food matters here too. The guidelines note that iftar, the meal that breaks the fast, often turns into a feast with large amounts of sugar and carbohydrate, and that skipping suhoor, the pre-dawn meal, depletes glycogen stores and brings on ketosis earlier in the fasting day.[@idfdar2021] The plate and the portion principles of this and the previous chapters apply directly: break the fast gently, put vegetables, protein, and a measured portion of a less-refined carbohydrate on the plate, and do not skip the pre-dawn meal. Any change to medicines during fasting must be planned with your team in advance.

Similar cautions apply to other fasting traditions. Whatever the tradition, the principle is the same: talk to your care team early, decide together, and have a plan for lows and highs.

Key Takeaways

• Every staple, whether rice, roti, injera, tortilla, ugali, potato, or plantain, delivers starch that becomes glucose. The rules of quality and quantity apply everywhere.
• White rice is associated with higher type 2 diabetes risk in observational data, especially where it is eaten several times a day; the message is to watch the portion and the preparation, not to abandon rice.
• Whole grains are consistently associated with lower risk; pulses and millets show promising, though still limited, evidence for glucose control.
• Roots, tubers, and plantain belong in the carbohydrate quarter. By weight, cooked pulses provide about the same carbohydrate as many roots but far more fibre.
• Measure in steps: the plate first, household measures next, and grams of carbohydrate only if you need it. Always weigh food in the same state, cooked or dry.
• Culture, celebration, and fasting can stay in your life with planning. For Ramadan and other fasts, see your care team well ahead of time.

Action Points

1. List your top ten carbohydrate foods. Write down what you eat most often, and your usual portion of each, using the bowl or spoon you really use.
2. Look up three of them in a food composition table for your country, or in the USDA database, and write the carbohydrate per portion.
3. Try one swap. Replace part of a refined starch with a whole-grain, pulse, or less-milled alternative, and keep the total carbohydrate similar.
4. Plan one celebration. Choose in advance what you will eat, what you will leave, and how you will bring your usual routine back afterwards.
5. If you fast, book your assessment with your care team six to eight weeks before the fast begins.

This book is intended for education and does not replace personal medical advice. If you have diabetes or take glucose-lowering medication, please consult your healthcare team before changing your diet, exercise, or treatment.