Lots of Individual Difference in Blood Glucose Response to Food

variety of foods
cafotodigital/E+/Getty Image

Capsule Summary: The effect different foods have on blood sugar, gut bacteria, and other factors varies, sometimes a lot, from person to person. Knowing their blood glucose response to foods could be helpful to people, even if they are not diabetic or prediabetic. This article will talk about some of the ways people differ that may affect diet, and discuss a November 2015 study that looked at this in detail.

The Search for "The Best Diet for Everyone" is a Fool's Errand

Whether it's the Dietary Guidelines for Americans, or recommendations from health organizations such as the American Heart Association, it seems that everyone wants to give advice about what everyone should eat. But is this really a good idea?

At this point, there have been lots of studies comparing diets of various types. Usually, they take groups of people who are randomly chosen and give them the different diets, and then measure various outcomes, be it weight, blood pressure, risk factors for heart disease, or a myriad of others. Many of these studies do not show big differences between the diet groups, possibly because people were reacting in opposite ways to some of the diets, making the average similar.

In contrast, relatively little research has been done that tries to determine which people do better on which diets. This, in spite of the fact that we have lots of evidence that there is quite a lot of individual variation when it comes to food.

We do have some hints. There is a fair amount of evidence that people with insulin resistance do better by reducing carbs , rather than fat or calories. There is also some research indicating that people who have large insulin responses to meals are, again, likely to respond best to limiting carbohydrate in the diet.

But there is not widespread knowledge or acceptance of this among nutrition and medical professionals, and we clearly need much more research of this kind.

Personalized Nutrition: We Are Different

In November 2015, a fascinating study called Personalized Nutrition by Prediction of Glycemic Responses was published in the journal Cell. It was a very well-controlled experiment that attempted to investigate several factors that influence health and explore how they may work together. It wasn't specifically about weight loss, but more about other indicators of good health.

Who Were The People Who Were Studied? The study took place in Isreal, and followed 800 people very closely. In looking at the stats on the participants, they were basically average people, health-wise. People who already had a diabetes diagnosis were excluded, although about 3% look like they probably were diabetics that hadn't been diagnosed, and 22% had blood glucose numbers consistent with prediabetes. The rest would be considered to have normal blood glucose, which is important when looking at what happened.

The average BMI for the group was 26.4, which is slightly overweight. About 22% were obese (BMI over 30).

Why Focus on Blood Glucose? The researchers decided to focus on blood glucose because they point out that higher blood glucose is associated with many negative health conditions: diabetes, heart disease, liver cirrhosis, obesity, and cancer.

The participants in the study were given continuous glucose monitors (these do pretty much what they sound like: constantly keep track of blood glucose). They were given the first meal of the day for seven days. There were three different standard meals, so they could compare how each person responded to the meals and also how different people might differ. The rest of the time, the participants ate what they wanted, but carefully measured and tracked what they ate on a mobile app, so that the blood glucose monitor could be matched to what they were eating, as well as to sleep and exercise, which were also tracked. Also, participants gave stool samples so that their gut bacteria could be monitored, and other measurements were taken through blood tests, weight, etc.

In all, the responses to about 47,000 meals were tracked.

Although the amount of total carbohydrate consumption in a meal is the best general indicator we have of how much blood sugar will rise, on average it doesn't explain as much of blood glucose as you'd think, and the results of this study bears this out. It turns out that in some people it's a great indicator, while in others, it's not as accurate a predictor. Moreover, people who had blood sugar spikes from a certain food (tomatoes has been given as an example), might not have much of a response to a different carb-containing food, and a different person could have the reverse responses to both. (This is one of the main reasons why the glycemic index does a fairly bad job of helping people with food choices.)

It turns out that the same goes for the amount of fat we eat. The standard thought about fat in a meal is that it will slow the blood glucose response of the meal, and on average it does. But guess what? For some people this is very true, but for others fat in a meal doesn't make much difference at all in terms of the amount blood glucose goes up or how long it takes for it to go up.

What's Going on in the Gut?

This, to me, is even MORE interesting than the blood sugar results. The researchers focused on a few different groups of gut bacteria that are associated either with diabetes (and poor blood glucose control), or with obesity. With some of the bacteria, we have hints of how they might be working to affect health, and with others we only know they are correlated with certain health factors. In the case of the people that were measured in this study, the bacteria present in their guts actually did line up with the strength of their blood glucose responses in general - people with higher blood glucose after meals had more of the "bad" bacteria associated with obesity and diabetes. But wait, there's more!

Putting the Data to Use: Predicting the Best Diet for the Person

The next phase of the research was to make an algorithm (basically the math behind a computer program) that would take all the measurements from the first phase and predict what diet would regulate blood glucose best for an individual. The researchers were also curious to find out whether different diets would affect people's gut bacteria in different ways.

In this phase, a new group of 100 people was run through the same week of data-gathering, and their diets predicted, either using the algorithm or a trained expert looking at the same data. Then a "good diet" and a "bad diet" were designed for each person. Note that some foods on the "good diet" for one person would be on the "bad diet" for another. Each person followed each of the diets for a week, with continued monitoring.

It worked! Not only did people have better blood glucose control on their "good" diets, but in general their gut bacteria actually responded positively as well! (There were a lot of individual differences in changing gut bacteria, but several types of the bacteria did change in the positive direction across most of the people.)

Reminder: At least 3/4 of these people would test out as glucose tolerant/having normal blood glucose.

What We Don't Know About the Study

The biggest thing that they don't tell us in the paper is what the people were eating! What were the "good diets" and the "bad diets"? I assume there weren't too many doughnuts in the "good diets", but what were the common foods? We know that people tended to eat similar amounts of protein, no matter how much other things in the diet varied, but how much carbohydrate did they eat? How much fat? It would be important to know!

What Can We Conclude?

Obviously, we have a long way to go in figuring out how individuals will respond to what they eat. What did we learn from this research that we can use today?

- Even for people with blood glucose in the normal range, variations in blood sugar can have an impact on health, and blood sugar spikes and valleys should be avoided. We already know that avoiding refined carbohydrates is the first step to more even blood sugar over the course of the day.

- Lower carbohydrate in the diet will result in better blood sugar control in general, but the details will vary from person to person.

- Regulating blood sugar may have a positive impact on the bacteria in our guts, which may create a positive feedback loop, improving our blood sugar even more, and possibly helping to prevent diabetes and obesity.

- It may be helpful for people to track their blood glucose response to learn how different foods affect them, even if they are not diabetic or prediabetic.


Zeevi, D. Korem T. et al. Personalized Nutrition by Prediction of Glycemic Responses.   Cell. 163:5. p1079–1094, 19 November 2015

Continue Reading