AP NEWS

Personalized diet predicts blood sugar spikes better than carb-counting

February 19, 2019
Nelson

Mayo Clinic researchers published the “baseline study for proving that personalized nutrition has a future in medicine.”

A great many variables play into the body’s glycemic, or blood sugar response to different foods — something like 72 factors, including genes, exercise levels, gut bacteria, and age, said Dr. Heidi Nelson, a co-author of the study.

“People always think it’s about how much you eat, or calories, or carbohydrates,” Nelson said. “It’s not that simple. How do you explain somebody who eats broccoli and gets a glycemic response, and the same person eats ice cream and gets no glycemic response? That’s what happened with one of my colleagues.”

The research, published in JAMA Network Open on Feb. 8, showed that individuals had wide-ranging glycemic responses to simple meals. The researchers took it one step further, then, and developed a model for predicting blood sugar responses to foods, based on those 72 factors.

Some people, the study indicates, feel energized in response to things like bananas, but may see a blood sugar spike and then feel tired after eating a cookie, Nelson said. Others may have an inverse reaction to the fruit vs. the sugary treat.

That’s more reassuring than you think.

When Nelson sees her patients, she’s often advised them to get to know the impact that certain foods have on them, she said. That’s because traditional dietary advice lumps foods into categories, and encourages dieters to eliminate or cut down on things like fats or carbohydrates.

An individualized diet would allow patients to eat things in some of those “problem” categories, if the data showed they wouldn’t have a significant glycemic response.

“If you think about how important diet is — lots of studies have shown correlation between diet and certain conditions,” Nelson said. “But if you look at diets for people who are overweight, one person may thrive on a specific, low-carb diet. Another may go for the Mediterranean diet. There doesn’t seem to be one diet that works for everyone.”

How they did it

The researchers developed the model by genetically sequencing 327 healthy people with no conditions that affected their blood sugar levels (i.e. diabetes or prediabetes) for six days. The participants wore blood glucose monitors that tracked their glycemic responses, and ate bagels and cream cheese for breakfast each day. Then they kept diaries of their self-determined food intake the rest of the day.

The researchers found that their model accurately predicted the participants’ blood sugar response to foods 62 percent of the time.

Does that sound low? Maybe. But it’s better than predicting blood sugar response based on carbohydrate content (40 percent accuracy) or calorie count (32 percent).

The research was done in collaboration with DayTwo Inc., an organization that provides genetic nutrition advice based on a similar study conducted at the Weizmann Institute of Science in Israel. The Mayo Clinic has invested in DayTwo Inc.

Previously, Mayo Clinic put a glucose monitor on study participants to monitor their responses to foods. They were able to track people’s response to different foods. But the Mayo Clinic’s model and tools like DayTwo may be able to predict the body’s response to new foods.

The report shows first, that nutrition can – and should – be personalized, and second, that complex data can be used as a predictive tool, Nelson said.

“This is a window into how different one person is from another in their response to the same foods,” Nelson said. “The next step will be, ’Can this be used to prevent a condition developing, because you’re controlling glucose levels. … It’s going to take some time to really sort through all of the implications, and see if it matters in a condition like prediabetes or diabetes – to see if you could really help people by eating the foods that are right for them, and whether that could change their medication.”

A different study years from now may find long-term benefits in diabetic or pre-diabetic populations, Nelson said, or look into how food combinations can cancel each others’ glycemic effects.

“It’s not just the food that matters, it’s what’s inside of you,” Nelson said. “It’s your microbiome, it’s how active you are, it’s your body type. A number of variables go together to make up your metabolic response and help inform what’s going to happen. It’s not just about eating a certain kind of food.”