Reprinted from the Washington Post.

The UNC Nutrition Research Institute is one of the sites conducting the study described in this article. Anyone 18 or older who is interested in participating can join by enrolling first in the national All of Us research program then opting into the Nutrition for Precision Health study. For additional information call 800-561-0315 or write to


The federal government wants you — yes, you — to join a large and ambitious diet study that could change the way we think about the best foods to eat for optimal health.

For the most part, nutrition experts know what constitutes a healthy diet. The government’s Dietary Guidelines for Americans encourages people to eat fruits, vegetables, beans, nuts, seafood, low-fat dairy and lean meat and whole grains, while limiting things such as alcohol, sodium, refined grains and added sugars.

But every person is unique, and so is the way that we metabolize food. Scientists have found that our genes, sex, gut microbiomes, sleep, exercise, stress levels and other factors can influence how our bodies respond to food. Even identical twins can have different metabolic responses to a banana, a cookie, a slice of bread or a bowl of oatmeal.

To sort it all out, the National Institutes of Health is spending $189 million over five years and recruiting 10,000 adults. The goal of the study, called Nutrition for Precision Health, is to find out how different people metabolize and respond to various diets. The investigators say they plan to use this data to develop machine-learning algorithms that can offer people personalized diet plans to improve their health.

Unhealthy diets are one of the leading causes of death and chronic disease around the world. Yet health authorities have taken a largely one-size-fits-all approach to nutrition, said Sai Krupa Das, a nutrition scientist at Tufts University who is involved in the study. “We know what’s healthy,” she said. “But not specifically for each subgroup of the population.”

The agency is recruiting people of different racial and ethnic backgrounds, people with obesity, disabilities, chronic health conditions and more. The investigators say they need diversity to understand the factors that determine how we metabolize food.

“We like to say that variation is our friend,” said Diana Thomas, a math professor at the U.S. Military Academy who is involved in the study. “We’re recruiting everybody we possibly can.”

The nutritional holy grail

The study is embedded in a larger federal research program called All of Us, which was created in 2015 under President Barack Obama. The goal of All of Us is to advance precision medicine — and ultimately tailor health care for individuals — by collecting health and genetic data from a million volunteers around the country.

Some experts say that developing an algorithm that can offer every individual a tailored diet based on their unique needs and physiology is the holy grail of nutrition.

A handful of companies, such as Zoe and DayTwo, have created algorithms to sell personalized diet plans. But they’ve typically relied on limited personal data, such as blood glucose tests and microbiome analyses, said Eric Topol a cardiologist and executive vice president of Scripps Research who wrote about personalized nutrition in his book “Deep Medicine.”

Topol said that to tailor dietary advice, more data is needed — such as information about a person’s genome, preexisting medical conditions, and sleep, exercise and stress levels. He said one service he tried recommended foods high in oxalates, such as nuts and strawberries. Topol has a history of kidney stones, which high-oxalate foods can exacerbate — something the service didn’t take into account. “They recommended that I eat nuts, which are the absolute worst food for me,” he said.

Topol and his colleagues are leading a precision nutrition study at Scripps to see whether they can predict how different foods affect blood sugar response and whether avoiding blood sugar spikes can prevent health problems.

He said the NIH study is so large, well-funded and comprehensive that it will probably yield promising results. “I think it’s going to play out to be useful in people who want to learn what is an optimal diet — not for the human species, but for them personally,” he added.

Sure, but what’s in it for me?

Enrollment in the Nutrition for Precision Health study is open online. The investigators say people who join will have the satisfaction of contributing to landmark research, and they’ll be given free analyses of their gut microbiomes, daily blood sugar fluctuations, insulin, gut and satiety hormones, and other intricacies of their metabolic health.

“People will get a lot of great information about their health,” said Holly Nicastro, a program director at the NIH and coordinator for the study.

Another selling point: Volunteers will be paid. Participants in the first phase can earn up to $300. Those who continue on to the second phase can earn another $2,000. And people who volunteer for the third, which requires living at a clinic site for weeks at a time, will earn up to $6,200.

Eye glasses that monitor your eating

But taking part in the study won’t be a piece of cake. In the first module, participants will be asked to provide access to their electronic health records so investigators know their medical histories and medications. They’ll have to make at least two trips to a clinic site for grip strength and body composition checks. They’ll provide a stool sample, and for about 10 days they’ll wear an activity tracker and glucose monitor as they go about their daily lives.

During that time, they’ll be required to document meals or take pictures of their food. Others will wear an automatic ingestion monitor — a small camera sensor that attaches to eyeglasses. Those who don’t wear glasses will be given fake ones. The camera sensor detects chewing and takes pictures to track food intake.

Nicastro said participants can remove the device for privacy. “You always have the option if you’re going to a doctor’s visit or viewing sensitive materials for work to take it off,” she said.

During the second phase of the study, a subset of volunteers will consume three different test diets for two weeks at a time. The participants will be provided all their meals.

Nicastro said the diets contain varying amounts of fats, carbs, protein, sugar, fiber and processed foods but won’t be named. “There’s no diet that is labeled a healthy diet or anything that’s going to cause a reaction in the participants,” she added. “We don’t want the participants to expect that one diet is going to be good for them or another diet is going to be bad.”

One diet for example contains a lot of plant foods, moderate amounts of dairy, meat, fish and nuts, and very low amounts of sugary drinks and desserts. Another diet has high amounts of refined grains, meat, sugary drinks and processed foods and low amounts of fruits, vegetables, and whole grains. And the third diet has moderate to high amounts of vegetables, meat, fish, nuts and oils, low amounts of fruit and dairy, and very little sugar or grains.

Nicastro said there needed to be enough “sunlight” between the diets to help the investigators understand the factors that cause different metabolic responses. “Our goal was to design three diets that would elicit very different responses in the participants so that we could study those differences,” said Nicastro.

The third and most intense phase will include 500 people who will live at a clinic site for six weeks. During that time, the investigators will provide all meals, track every morsel of food the participants eat, and closely monitor their weight, sleep, physical activity, glucose levels and the precise number of calories they consume and burn.

Okay, so how do you join?

Almost anyone in the United States who is 18 or older can join. Start by enrolling at the All of Us study site. Then you’ll have the option to sign up online for Nutrition for Precision Health, which is being carried out at 14 different research centers in six states — Alabama, California, Illinois, Louisiana, Massachusetts and North Carolina.

Topol said that developing better algorithms for personalized diet advice is a “long haul” effort that could take years, but it’s sorely needed. “Each of us has a different metabolism, microbiome, genome and physiology,” he said. “The fact that we have these general recommendations for nutrition for all humans — we can be more intelligent than that.”