AI and you: Is the promise of personalized nutrition apps worth the hype?
As a type 2 diabetic, Michael Snyder has long been interested in how blood sugar levels vary from one person to another in response to the same food, and whether a more personalized approach to nutrition could help tackle the rapidly cascading levels of diabetes and obesity in much of the western world.
Eight years ago, Snyder, who directs the Center for Genomics and Personalized Medicine at Stanford University, decided to put his theories to the test. In the 2000s continuous glucose monitoring, or CGM, had begun to revolutionize the lives of diabetics, both type 1 and type 2. Using spherical sensors which sit on the upper arm or abdomen – with tiny wires that pierce the skin – the technology allowed patients to gain real-time updates on their blood sugar levels, transmitted directly to their phone.
It gave Snyder an idea for his research at Stanford. Applying the same technology to a group of apparently healthy people, and looking for ‘spikes’ or sudden surges in blood sugar known as hyperglycemia, could provide a means of observing how their bodies reacted to an array of foods.
“We discovered that different foods spike people differently,” he says. “Some people spike to pasta, others to bread, others to bananas, and so on. It’s very personalized and our feeling was that building programs around these devices could be extremely powerful for better managing people’s glucose.”
Unbeknown to Snyder at the time, thousands of miles away, a group of Israeli scientists at the Weizmann Institute of Science were doing exactly the same experiments. In 2015, they published a landmark paper which used CGM to track the blood sugar levels of 800 people over several days, showing that the biological response to identical foods can vary wildly. Like Snyder, they theorized that giving people a greater understanding of their own glucose responses, so they spend more time in the normal range, may reduce the prevalence of type 2 diabetes.
The commercial potential of such apps is clear, but the underlying science continues to generate intriguing findings.
“At the moment 33 percent of the U.S. population is pre-diabetic, and 70 percent of those pre-diabetics will become diabetic,” says Snyder. “Those numbers are going up, so it’s pretty clear we need to do something about it.”
Fast forward to 2022,and both teams have converted their ideas into subscription-based dietary apps which use artificial intelligence to offer data-informed nutritional and lifestyle recommendations. Snyder’s spinoff, January AI, combines CGM information with heart rate, sleep, and activity data to advise on foods to avoid and the best times to exercise. DayTwo–a start-up which utilizes the findings of Weizmann Institute of Science–obtains microbiome information by sequencing stool samples, and combines this with blood glucose data to rate ‘good’ and ‘bad’ foods for a particular person.
“CGMs can be used to devise personalized diets,” says Eran Elinav, an immunology professor and microbiota researcher at the Weizmann Institute of Science in addition to serving as a scientific consultant for DayTwo. “However, this process can be cumbersome. Therefore, in our lab we created an algorithm, based on data acquired from a big cohort of people, which can accurately predict post-meal glucose responses on a personal basis.”
The commercial potential of such apps is clear. DayTwo, who market their product to corporate employers and health insurers rather than individual consumers, recently raised $37 million in funding. But the underlying science continues to generate intriguing findings.
Last year, Elinav and colleagues published a study on 225 individuals with pre-diabetes which found that they achieved better blood sugar control when they followed a personalized diet based on DayTwo’s recommendations, compared to a Mediterranean diet. The journal Cell just released a new paper from Snyder’s group which shows that different types of fibre benefit people in different ways.
“The idea is you hear different fibres are good for you,” says Snyder. “But if you look at fibres they’re all over the map—it’s like saying all animals are the same. The responses are very individual. For a lot of people [a type of fibre called] arabinoxylan clearly reduced cholesterol while the fibre inulin had no effect. But in some people, it was the complete opposite.”
Eight years ago, Stanford's Michael Snyder began studying how continuous glucose monitors could be used by patients to gain real-time updates on their blood sugar levels, transmitted directly to their phone.
The Snyder Lab, Stanford Medicine
Because of studies like these, interest in precision nutrition approaches has exploded in recent years. In January, the National Institutes of Health announced that they are spending $170 million on a five year, multi-center initiative which aims to develop algorithms based on a whole range of data sources from blood sugar to sleep, exercise, stress, microbiome and even genomic information which can help predict which diets are most suitable for a particular individual.
“There's so many different factors which influence what you put into your mouth but also what happens to different types of nutrients and how that ultimately affects your health, which means you can’t have a one-size-fits-all set of nutritional guidelines for everyone,” says Bruce Y. Lee, professor of health policy and management at the City University of New York Graduate School of Public Health.
With the falling costs of genomic sequencing, other precision nutrition clinical trials are choosing to look at whether our genomes alone can yield key information about what our diets should look like, an emerging field of research known as nutrigenomics.
The ASPIRE-DNA clinical trial at Imperial College London is aiming to see whether particular genetic variants can be used to classify individuals into two groups, those who are more glucose sensitive to fat and those who are more sensitive to carbohydrates. By following a tailored diet based on these sensitivities, the trial aims to see whether it can prevent people with pre-diabetes from developing the disease.
But while much hope is riding on these trials, even precision nutrition advocates caution that the field remains in the very earliest of stages. Lars-Oliver Klotz, professor of nutrigenomics at Friedrich-Schiller-University in Jena, Germany, says that while the overall goal is to identify means of avoiding nutrition-related diseases, genomic data alone is unlikely to be sufficient to prevent obesity and type 2 diabetes.
“Genome data is rather simple to acquire these days as sequencing techniques have dramatically advanced in recent years,” he says. “However, the predictive value of just genome sequencing is too low in the case of obesity and prediabetes.”
Others say that while genomic data can yield useful information in terms of how different people metabolize different types of fat and specific nutrients such as B vitamins, there is a need for more research before it can be utilized in an algorithm for making dietary recommendations.
“I think it’s a little early,” says Eileen Gibney, a professor at University College Dublin. “We’ve identified a limited number of gene-nutrient interactions so far, but we need more randomized control trials of people with different genetic profiles on the same diet, to see whether they respond differently, and if that can be explained by their genetic differences.”
Some start-ups have already come unstuck for promising too much, or pushing recommendations which are not based on scientifically rigorous trials. The world of precision nutrition apps was dubbed a ‘Wild West’ by some commentators after the founders of uBiome – a start-up which offered nutritional recommendations based on information obtained from sequencing stool samples –were charged with fraud last year. The weight-loss app Noom, which was valued at $3.7 billion in May 2021, has been criticized on Twitter by a number of users who claimed that its recommendations have led to them developed eating disorders.
With precision nutrition apps marketing their technology at healthy individuals, question marks have also been raised about the value which can be gained through non-diabetics monitoring their blood sugar through CGM. While some small studies have found that wearing a CGM can make overweight or obese individuals more motivated to exercise, there is still a lack of conclusive evidence showing that this translates to improved health.
However, independent researchers remain intrigued by the technology, and say that the wealth of data generated through such apps could be used to help further stratify the different types of people who become at risk of developing type 2 diabetes.
“CGM not only enables a longer sampling time for capturing glucose levels, but will also capture lifestyle factors,” says Robert Wagner, a diabetes researcher at University Hospital Düsseldorf. “It is probable that it can be used to identify many clusters of prediabetic metabolism and predict the risk of diabetes and its complications, but maybe also specific cardiometabolic risk constellations. However, we still don’t know which forms of diabetes can be prevented by such approaches and how feasible and long-lasting such self-feedback dietary modifications are.”
Snyder himself has now been wearing a CGM for eight years, and he credits the insights it provides with helping him to manage his own diabetes. “My CGM still gives me novel insights into what foods and behaviors affect my glucose levels,” he says.
He is now looking to run clinical trials with his group at Stanford to see whether following a precision nutrition approach based on CGM and microbiome data, combined with other health information, can be used to reverse signs of pre-diabetes. If it proves successful, January AI may look to incorporate microbiome data in future.
“Ultimately, what I want to do is be able take people’s poop samples, maybe a blood draw, and say, ‘Alright, based on these parameters, this is what I think is going to spike you,’ and then have a CGM to test that out,” he says. “Getting very predictive about this, so right from the get go, you can have people better manage their health and then use the glucose monitor to help follow that.”
Two years, six million deaths and still counting, scientists are searching for answers to prevent another COVID-19-like tragedy from ever occurring again. And it’s a gargantuan task.
Our disturbed ecosystems are creating more favorable conditions for the spread of infectious disease. Global warming, deforestation, rising sea levels and flooding have contributed to a rise in mosquito-borne infections and longer tick seasons. Disease-carrying animals are in closer range to other species and humans as they migrate to escape the heat. Bats are thought to have carried the SARS-CoV-2 virus to Wuhan, either directly or through another host animal, but thousands of novel viruses are lurking within other wild creatures.
Understanding how climate change contributes to the spread of disease is critical in predicting and thwarting future calamities. But the problem is that predictive models aren’t yet where they need to be for forecasting with certainty beyond the next year, as we could for weather, for instance.
The association between climate and infectious disease is poorly understood, says Irina Tezaur, a computational scientist at Sandia National Laboratories. “Correlations have been observed but it’s not known if these correlations translate to causal relationships.”
To make accurate longer-term predictions, scientists need more empirical data, multiple datasets specific to locations and diseases, and the ability to calculate risks that depend on unpredictable nature and human behavior. Another obstacle is that climate scientists and epidemiologists are not collaborating effectively, so some researchers are calling for a multidisciplinary approach, a new field called Outbreak Science.
Climate scientists are far ahead of epidemiologists in gathering essential data.
Earth System Models—combining the interactions of atmosphere, ocean, land, ice and biosphere—have been in place for two decades to monitor the effects of global climate change. These models must be combined with epidemiological and human model research, areas that are easily skewed by unpredictable elements, from extreme weather events to public environmental policy shifts.
“There is never just one driver in tracking the impact of climate on infectious disease,” says Joacim Rocklöv, a professor at the Heidelberg Institute of Global Health & Heidelberg Interdisciplinary Centre for Scientific Computing in Germany. Rocklöv has studied how climate affects vector-borne diseases—those transmitted to humans by mosquitoes, ticks or fleas. “You need to disentangle the variables to find out how much difference climate makes to the outcome and how much is other factors.” Determinants from deforestation to population density to lack of healthcare access influence the spread of disease.
Even though climate change is not the primary driver of infectious disease today, it poses a major threat to public health in the future, says Rocklöv.
The promise of predictive modeling
“Models are simplifications of a system we’re trying to understand,” says Jeremy Hess, who directs the Center for Health and the Global Environment at University of Washington in Seattle. “They’re tools for learning that improve over time with new observations.”
Accurate predictions depend on high-quality, long-term observational data but models must start with assumptions. “It’s not possible to apply an evidence-based approach for the next 40 years,” says Rocklöv. “Using models to experiment and learn is the only way to figure out what climate means for infectious disease. We collect data and analyze what already happened. What we do today will not make a difference for several decades.”
To improve accuracy, scientists develop and draw on thousands of models to cover as many scenarios as possible. One model may capture the dynamics of disease transmission while another focuses on immunity data or ocean influences or seasonal components of a virus. Further, each model needs to be disease-specific and often location-specific to be useful.
“All models have biases so it’s important to use a suite of models,” Tezaur stresses.
The modeling scientist chooses the drivers of change and parameters based on the question explored. The drivers could be increased precipitation, poverty or mosquito prevalence, for instance. Later, the scientist may need to isolate the effect of one driver so that will require another model.
There have been some related successes, such as the latest models for mosquito-borne diseases like Dengue, Zika and malaria as well as those for flu and tick-borne diseases, says Hess.
Rocklöv was part of a research team that used test data from 2018 and 2019 to identify regions at risk for West Nile virus outbreaks. Using AI, scientists were able to forecast outbreaks of the virus for the entire transmission season in Europe. “In the end, we want data-driven models; that’s what AI can accomplish,” says Rocklöv. Other researchers are making an important headway in creating a framework to predict novel host–parasite interactions.
Modeling studies can run months, years or decades. “The scientist is working with layers of data. The challenge is how to transform and couple different models together on a planetary scale,” says Jeanne Fair, a scientist at Los Alamos National Laboratory, Biosecurity and Public Health, in New Mexico.
Disease forecasting will require a significant investment into the infrastructure needed to collect data about the environment, vectors, and hosts a tall spatial and temporal resolutions.
And it’s a constantly changing picture. A modeling study in an April 2022 issue of Nature predicted that thousands of animals will migrate to cooler locales as temperatures rise. This means that various species will come into closer contact with people and other mammals for the first time. This is likely to increase the risk of emerging infectious disease transmitted from animals to humans, especially in Africa and Asia.
Other things can happen too. Global warming could precipitate viral mutations or new infectious diseases that don’t respond to antimicrobial treatments. Insecticide-resistant mosquitoes could evolve. Weather-related food insecurity could increase malnutrition and weaken people’s immune systems. And the impact of an epidemic will be worse if it co-occurs during a heatwave, flood, or drought, says Hess.
The devil is in the climate variables
Solid predictions about the future of climate and disease are not possible with so many uncertainties. Difficult-to-measure drivers must be added to the empirical model mix, such as land and water use, ecosystem changes or the public’s willingness to accept a vaccine or practice social distancing. Nor is there any precedent for calculating the effect of climate changes that are accelerating at a faster speed than ever before.
The most critical climate variables thought to influence disease spread are temperature, precipitation, humidity, sunshine and wind, according to Tezaur’s research. And then there are variables within variables. Influenza scientists, for example, found that warm winters were predictors of the most severe flu seasons in the following year.
The human factor may be the most challenging determinant. To what degree will people curtail greenhouse gas emissions, if at all? The swift development of effective COVID-19 vaccines was a game-changer, but will scientists be able to repeat it during the next pandemic? Plus, no model could predict the amount of internet-fueled COVID-19 misinformation, Fair noted. To tackle this issue, infectious disease teams are looking to include more sociologists and political scientists in their modeling.
Addressing the gaps
Currently, researchers are focusing on the near future, predicting for next year, says Fair. “When it comes to long-term, that’s where we have the most work to do.” While scientists cannot foresee how political influences and misinformation spread will affect models, they are positioned to make headway in collecting and assessing new data streams that have never been merged.
Disease forecasting will require a significant investment into the infrastructure needed to collect data about the environment, vectors, and hosts at all spatial and temporal resolutions, Fair and her co-authors stated in their recent study. For example real-time data on mosquito prevalence and diversity in various settings and times is limited or non-existent. Fair also would like to see standards set in mosquito data collection in every country. “Standardizing across the US would be a huge accomplishment,” she says.
Understanding how climate change contributes to the spread of disease is critical for thwarting future calamities.
Jeanne Fair
Hess points to a dearth of data in local and regional datasets about how extreme weather events play out in different geographic locations. His research indicates that Africa and the Middle East experienced substantial climate shifts, for example, but are unrepresented in the evidentiary database, which limits conclusions. “A model for dengue may be good in Singapore but not necessarily in Port-au-Prince,” Hess explains. And, he adds, scientists need a way of evaluating models for how effective they are.
The hope, Rocklöv says, is that in the future we will have data-driven models rather than theoretical ones. In turn, sharper statistical analyses can inform resource allocation and intervention strategies to prevent outbreaks.
Most of all, experts emphasize that epidemiologists and climate scientists must stop working in silos. If scientists can successfully merge epidemiological data with climatic, biological, environmental, ecological and demographic data, they will make better predictions about complex disease patterns. Modeling “cross talk” and among disciplines and, in some cases, refusal to release data between countries is hindering discovery and advances.
It’s time for bold transdisciplinary action, says Hess. He points to initiatives that need funding in disease surveillance and control; developing and testing interventions; community education and social mobilization; decision-support analytics to predict when and where infections will emerge; advanced methodologies to improve modeling; training scientists in data management and integrated surveillance.
Establishing a new field of Outbreak Science to coordinate collaboration would accelerate progress. Investment in decision-support modeling tools for public health teams, policy makers, and other long-term planning stakeholders is imperative, too. We need to invest in programs that encourage people from climate modeling and epidemiology to work together in a cohesive fashion, says Tezaur. Joining forces is the only way to solve the formidable challenges ahead.
This article originally appeared in One Health/One Planet, a single-issue magazine that explores how climate change and other environmental shifts are increasing vulnerabilities to infectious diseases by land and by sea. The magazine probes how scientists are making progress with leaders in other fields toward solutions that embrace diverse perspectives and the interconnectedness of all lifeforms and the planet.
Scientists use AI to predict how hospital stays will go
The Friday Five covers five stories in research that you may have missed this week. There are plenty of controversies and troubling ethical issues in science – and we get into many of them in our online magazine – but this news roundup focuses on scientific creativity and progress to give you a therapeutic dose of inspiration headed into the weekend.
Here are the promising studies covered in this week's Friday Five:
- The problem with bedtime munching
- Scientists use AI to predict how stays in hospitals will go
- How to armor the shields of our livers against cancer
- One big step to save the world: turn one kind of plastic into another
- The perfect recipe for tiny brains
And an honorable mention this week: Bigger is better when it comes to super neurons in super agers