It was a tasty task: to create the healthiest bread possible.
First, the researchers from the Weizmann Institute of Science procured a superior grain: hard red bread wheat. They processed the seeds in a traditional mill, without machinery or industrial methods. The stone-milled whole wheat flour was then given to an artisanal bakery, where it was prepared with only water, salt, and sourdough starter; no additives, no preservatives. After baking in a stone hearth oven for an hour, the loaves were done. (Cue freshly baked bread smell.)
"This was, at least as far as we're concerned, the best bread you could make," Tal Korem, a computational biologist, says. "During the study, we had a continuous stream of fresh bread loaves coming to the lab each morning. I ate so much until my teeth would hurt."
Now the scientists would put the bread to the test, alongside a less-lovingly crafted competitor: white bread. The white bread was one of the most common brands in Israel (think, Wonderbread), made in an industrial factory with additives, preservatives, and leavened with an industrial yeast. The whole wheat bread would certainly prove to be healthier—or would it?
Korem works in the lab of Eran Segal—also a computational biologist—who for the past five years has been studying individual responses to all foods, both healthy and unhealthy. After a large study in 2015, published in Cell, the team found that even when eating identical meals, people respond very differently. Their findings raised questions about generalized nutritional recommendations and dietary suggestions—like the food pyramid we're so accustomed to following. What they discovered is that what's good for you, might not be good for me.
In 2015, before pitting the breads against each other, Segal and his group recruited 800 volunteers who self-reported almost 10 million calories over 47,000 meals. The main measurement they collected was each person's blood glucose response, along with their physical activities, overall lifestyle, and periodic stool samples to get data on the bacteria in their guts.
Glucose can provide key insight into the body's response to food and overall health. It's the brain's main energy source and we need it to function normally. But too much glucose can be a poison to the body. High glycemic responses are associated with obesity, liver and heart disease, and other oxidative stresses, inflammation, and chronic diseases.
Overall blood glucose levels are rising, as seen through the increase in prediabetes and impaired glucose intolerance. In the US, it affects 37 percent of the adult population. Being pre-diabetic greatly increases the risk of Type II diabetes (70 percent of pre-diabetics eventually get it) and it's linked to many other health issues, a cluster called the metabolic syndrome.
After each meal, blood sugar levels increase as your body absorbs nutrients, and your body takes action to reduce these levels back to normal. We're told that foods that score lower on the glycemic index have a slower release of glucose—that's why we stay away from sugary foods and carbs like white bread.
"The glycemic index says that the glucose response is a property of the food," Korem says. "If I eat a certain food, or you eat a certain food, or someone from India eats a certain food, and we compare the glucose, then it's the type of food that matters, and not the person that's eating."
What Segal and his group found was different. Their 800 people, even when eating the same things, had different blood glucose levels, sometimes even opposite responses. This means that some healthy foods caused blood sugar spikes in some, and other unhealthy foods had very low glycemic responses.
More From Tonic:
Using all their data, the team came up with an algorithm that could predict what kind of blood sugar levels an individual would have. Rather than just looking at nutritional composition or the glycemic index, they developed an even more accurate prediction model by including the kinds of bacteria in each person's microbiome.
Your microbiome makeup is influenced by what you eat, and it, in return, changes the way you react to and digest foods and nutrients. Since each person's microbiome varies, so do our personal responses to foods. Segal's group still doesn't know enough about specific microbes, or even how much of a role the microbes play in the glycemic response itself. At the moment, the microbiota is just a fingerprint, but one that tells us more about a person's overall response to a meal.
They validated their algorithm on a new group of 100 volunteers, to prove that it worked, and then put it into action. They found that if they used the algorithm to advise people on what to eat—eating foods that were individually best for them—they could lower that person's glycemic responses.
"We actually saw very significant differences between them," Korem says. "Most of the people in this study were prediabetic and in the week of eating 'good' meals for them, they had blood glucose levels that were almost completely normal."
Eran Elinav, a co-author and collaborator on the study, says that their findings reveal the need for a more personalized nutrition approach. When I ask him if he now views government issued dietary recommendations as inaccurate, he says: "Absolutely."
"For the vast majority of foods and the vast majority of individuals, unless you use quantitative measures, I don't see a rational way that you would be able to come up with a healthy diet per individual," Elinav says.
Here's an example: In a forthcoming study from the group, one of their volunteers is a diabetic man who was consistently eating something that caused his blood glucose level to sky rocket, and it would remain high for hours after. When they asked him, "Why are you eating that food," he said that his clinical dietician had told him to.
"I'm not saying that person did a poor job," Korem says. "I think that for 999/1000 people this recommendation would have been great. This type of food would have been healthy for them. But for this kind of person specifically, this was a very bad recommendation. He was eating something that is a whole food, organic, non-processed, with complex carbohydrates, high fiber quantity. It's considered very healthy, and is probably healthy for a majority of people. But not for this person."
So what about our artisanal whole wheat bread? After their 2015 study, the group wanted to test two foods, where one was widely accepted to be healthier: white and wheat bread. They recruited groups to eat one type or the other—about 25 percent of their participants' caloric intake was from either bread.
When they first looked at the results, it seemed as if there wasn't any difference. Korem says their data indicated either the breads were identical in effect (unlikely) or that the inter-personal differences varied so much, that when they took the average of the responses they lost the signal. And that turned out to be true for their glycemic responses.
"What we saw is that half of the people in the study had higher responses to white bread, while the other half of the people in the study had higher responses to sourdough bread," he says. "When you average them together, you see no difference between the breads, but when you look at each one, you see that there is a very significant difference."
When re-applying microbiome algorithm, they could again, accurately predict a person's blood sugar levels in response to the white or wheat bread.
They are still focusing on glycemic response, but Korem thinks that it could apply to a variety of other factors. "Obviously, glucose responses aren't the entire picture," he says. "And we're not claiming it to be the only thing to look at. What we're saying is that glucose responses are something we can measure and it tells us something. The glucose responses tell us very strongly that people respond very differently to identical foods. If that's the evidence you give me, I think the most plausible hypothesis is that it applies to other means of nutritional effects. And I think this is something that science should research and study."
I asked Jairam K. P. Vanamala, a professor of food science at Penn State University and faculty at the Penn State Hershey Cancer Institute, about the glucose findings, and how we could use this information to inform our own diets. How do I know if I should eat white or wheat bread without sequencing my gut microbiome?
Vanamala says that interpersonal variability is an important discovery, and will bring us closer to a food-based approach to countering chronic diseases, something that his lab focuses on. But in the meantime, it's probably best to assume you're not one of those outliers that has a low-glycemic response to white bread, and a high glycemic response to kale. The one thing we do know is good for glycemic responses and overall health is diversity in the microbiome. His advice to achieve this isn't anything you haven't heard before.
"While we are trying to understand all of these intricacies, there are some general guidelines that apply to most people," Vanamala says. "That's the reason why we say to eat a plant-based diet made of different types of plants, different parts of the plant and different colors of plants. This seems to be beneficial for not only promoting bacterial diversity, but also preventing chronic diseases.
"But at the same time, you are right on the money when you said that we cannot come up with the guidelines that are appropriate and accurate for each individual. As we move forward, that's the reason why we're interested in understanding how this works. We cannot come to that personalized nutrition if we don't understand fundamentally how food and gut bacteria and the host are interacting in health and disease."
As for bread, Korem says it wouldn't be responsible for him to claim that white bread is healthy, even after their study.
"But we now have evidence showing that it might be healthy for some people," he says. "What this study means is that the statement that whole wheat sourdough bread is always better than white bread—that statement is wrong."
Read This Next: More Evidence That You Should Be Eating Gluten