Meals and blood sugar: what the research has found

People often ask whether changing what they eat can change their blood sugar. Researchers have asked the same question for decades, and some of the answers are unusually clear for nutrition science. Others are smaller, shorter or less certain than headlines suggest.

This article walks through five well-known studies: who took part, what they did, what was measured, and what each one cannot tell you. It reports research. It is not a plan to follow, and it says nothing about your own results.

Two prevention trials: people at high risk of type 2 diabetes

The two best-known trials did not test food alone. They tested a package of lifestyle changes in people whose glucose tests were already above normal but below the diabetes range.

The Finnish Diabetes Prevention Study [1] randomly assigned 522 middle-aged, overweight adults with impaired glucose tolerance to individual counselling or to a control group. The counselling aimed at losing weight, eating less total fat and less saturated fat, eating more fibre, and moving more. After four years, 11 percent of the counselling group had developed diabetes, against 23 percent of the control group. Over the trial, the risk was 58 percent lower in the counselling group.

The US Diabetes Prevention Program [2] randomly assigned 3,234 adults with raised fasting and after-meal glucose to a lifestyle programme, to the medicine metformin, or to a placebo. The lifestyle goals were losing at least 7 percent of body weight and at least 150 minutes of activity a week. Over an average of 2.8 years, diabetes developed in 4.8 people per 100 per year in the lifestyle group, 7.8 with metformin and 11.0 with placebo: 58 percent fewer cases with the lifestyle programme, and 31 percent fewer with metformin.

The same people were followed for about 15 years [3]. The difference narrowed over time but did not disappear: diabetes was 27 percent less likely in the original lifestyle group than in the placebo group. By year 15, 55 percent of the lifestyle group had developed diabetes, against 62 percent of the placebo group.

What these trials cannot tell you. Food was one part of each programme, alongside activity, weight loss and regular support, so the trials cannot say how much of the effect came from meals alone. The participants were adults already at high risk, not everyone. And the long follow-up shows that the gap shrinks over the years.

A trial in people who already had type 2 diabetes

DiRECT [4], in Scotland and the north of England, tested something far more intensive. 306 adults who had been diagnosed with type 2 diabetes within the previous six years, were not using insulin and had a body mass index of 27 to 45 were assigned, by GP practice, to a weight management programme or to usual care.

The programme was not an everyday diet. Diabetes and blood pressure medicines were stopped under supervision; participants then lived on a formula diet of about 825 to 853 kcal a day for three to five months, followed by a stepped return to food and support to keep the weight off.

After 12 months, 46 percent of the programme group were in remission, defined as an HbA1c below 6.5 percent (48 mmol/mol) after at least two months without any diabetes medicine, against 4 percent of the usual-care group. Remission was more common the more weight people lost: 86 percent of those who lost 15 kg or more were in remission, and none of those who gained weight. That last comparison is an observation within the trial, not a randomised result.

What it cannot tell you. This was a medically supervised programme with a formula diet, not a change in ordinary meals. Everyone knew which group they were in. The paper reports 12 months; it does not show what happens years later. Its participants had a recent diagnosis and were not on insulin.

A small study on the order of food within a meal

A 2015 pilot study [5] gave 11 adults with type 2 diabetes, all treated with metformin, the same 628 kcal meal on two days a week apart: chicken, salad and broccoli, plus bread and orange juice. On one day they ate the bread and juice first and the rest 15 minutes later; on the other day, the reverse.

When the protein and vegetables came first, blood glucose was 29 percent lower at 30 minutes, 37 percent lower at 60 minutes and 17 percent lower at 120 minutes than when the carbohydrate came first. The total rise over two hours was 73 percent smaller.

What it cannot tell you. Eleven people, one meal each way, two hours of measurements. The authors call it a pilot and say longer studies are needed. It shows what happened after one meal in a lab, not what happens to HbA1c or to health over months.

A large review of fibre and whole grains

A series of systematic reviews published in The Lancet in 2019 [6] pooled about 185 long-term observational studies and 58 clinical trials. In the observational studies, people who ate the most fibre had a 15 to 30 percent lower risk of several conditions, type 2 diabetes among them, than people who ate the least. The largest differences were seen at about 25 to 29 g of fibre a day. Whole grains showed similar patterns. Comparing diets with a low and a high glycaemic index showed smaller or no differences.

The reviewers rated the certainty of the evidence moderate for fibre, low to moderate for whole grains, and low to very low for glycaemic index.

What it cannot tell you. Most of these findings come from observational studies. They show that people who ate more fibre had fewer of these conditions; they cannot prove the fibre caused the difference. The reviews left out studies of people who already had a chronic disease.

What the research, taken together, does and does not say

• The strongest results come from structured programmes with support, not from a single food or a single trick.
• In people at high risk, programmes that combined eating changes, activity and weight loss were followed by fewer new cases of type 2 diabetes, and the gap narrowed over 15 years.
• The most dramatic result, DiRECT, came from a medically supervised formula diet, not from ordinary meals.
• Smaller studies, like the one on food order, describe what happens after a meal; they do not show long-term outcomes.
• None of these studies can predict what will happen to any one person.

Where a food log fits

Most of the programmes above asked people to pay attention to what they ate over months. A food log is one way to keep that record. Senumi lets you log meals by photo, barcode, typed description or database search, and keep readings such as fasting blood sugar next to them, typed in or read from a lab report, so you have the record in one place when you talk to your doctor.

Senumi does not change blood sugar, does not give medical advice, and its food and portion figures are estimates. It is a notebook, not a treatment.

Sources

1. Tuomilehto J, Lindström J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med. 2001;344(18):1343-50.
2. Diabetes Prevention Program Research Group (Knowler WC, et al.). Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393-403.
3. Diabetes Prevention Program Research Group. Long-term effects of lifestyle intervention or metformin on diabetes development and microvascular complications over 15-year follow-up: the Diabetes Prevention Program Outcomes Study. Lancet Diabetes Endocrinol. 2015;3(11):866-75.
4. Lean MEJ, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT): an open-label, cluster-randomised trial. Lancet. 2018;391(10120):541-51.
5. Shukla AP, Iliescu RG, Thomas CE, Aronne LJ. Food order has a significant impact on postprandial glucose and insulin levels. Diabetes Care. 2015;38(7):e98-9.
6. Reynolds A, Mann J, Cummings J, et al. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet. 2019;393(10170):434-45.