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.”
Story by Big Think
For most of history, artificial intelligence (AI) has been relegated almost entirely to the realm of science fiction. Then, in late 2022, it burst into reality — seemingly out of nowhere — with the popular launch of ChatGPT, the generative AI chatbot that solves tricky problems, designs rockets, has deep conversations with users, and even aces the Bar exam.
But the truth is that before ChatGPT nabbed the public’s attention, AI was already here, and it was doing more important things than writing essays for lazy college students. Case in point: It was key to saving the lives of tens of millions of people.
AI-designed mRNA vaccines
As Dave Johnson, chief data and AI officer at Moderna, told MIT Technology Review‘s In Machines We Trust podcast in 2022, AI was integral to creating the company’s highly effective mRNA vaccine against COVID. Moderna and Pfizer/BioNTech’s mRNA vaccines collectively saved between 15 and 20 million lives, according to one estimate from 2022.
Johnson described how AI was hard at work at Moderna, well before COVID arose to infect billions. The pharmaceutical company focuses on finding mRNA therapies to fight off infectious disease, treat cancer, or thwart genetic illness, among other medical applications. Messenger RNA molecules are essentially molecular instructions for cells that tell them how to create specific proteins, which do everything from fighting infection, to catalyzing reactions, to relaying cellular messages.
Johnson and his team put AI and automated robots to work making lots of different mRNAs for scientists to experiment with. Moderna quickly went from making about 30 per month to more than one thousand. They then created AI algorithms to optimize mRNA to maximize protein production in the body — more bang for the biological buck.
For Johnson and his team’s next trick, they used AI to automate science, itself. Once Moderna’s scientists have an mRNA to experiment with, they do pre-clinical tests in the lab. They then pore over reams of data to see which mRNAs could progress to the next stage: animal trials. This process is long, repetitive, and soul-sucking — ill-suited to a creative scientist but great for a mindless AI algorithm. With scientists’ input, models were made to automate this tedious process.
“We don’t think about AI in the context of replacing humans,” says Dave Johnson, chief data and AI officer at Moderna. “We always think about it in terms of this human-machine collaboration, because they’re good at different things. Humans are really good at creativity and flexibility and insight, whereas machines are really good at precision and giving the exact same result every single time and doing it at scale and speed.”
All these AI systems were in put in place over the past decade. Then COVID showed up. So when the genome sequence of the coronavirus was made public in January 2020, Moderna was off to the races pumping out and testing mRNAs that would tell cells how to manufacture the coronavirus’s spike protein so that the body’s immune system would recognize and destroy it. Within 42 days, the company had an mRNA vaccine ready to be tested in humans. It eventually went into hundreds of millions of arms.
Biotech harnesses the power of AI
Moderna is now turning its attention to other ailments that could be solved with mRNA, and the company is continuing to lean on AI. Scientists are still coming to Johnson with automation requests, which he happily obliges.
“We don’t think about AI in the context of replacing humans,” he told the Me, Myself, and AI podcast. “We always think about it in terms of this human-machine collaboration, because they’re good at different things. Humans are really good at creativity and flexibility and insight, whereas machines are really good at precision and giving the exact same result every single time and doing it at scale and speed.”
Moderna, which was founded as a “digital biotech,” is undoubtedly the poster child of AI use in mRNA vaccines. Moderna recently signed a deal with IBM to use the company’s quantum computers as well as its proprietary generative AI, MoLFormer.
Moderna’s success is encouraging other companies to follow its example. In January, BioNTech, which partnered with Pfizer to make the other highly effective mRNA vaccine against COVID, acquired the company InstaDeep for $440 million to implement its machine learning AI across its mRNA medicine platform. And in May, Chinese technology giant Baidu announced an AI tool that designs super-optimized mRNA sequences in minutes. A nearly countless number of mRNA molecules can code for the same protein, but some are more stable and result in the production of more proteins. Baidu’s AI, called “LinearDesign,” finds these mRNAs. The company licensed the tool to French pharmaceutical company Sanofi.
Writing in the journal Accounts of Chemical Research in late 2021, Sebastian M. Castillo-Hair and Georg Seelig, computer engineers who focus on synthetic biology at the University of Washington, forecast that AI machine learning models will further accelerate the biotechnology research process, putting mRNA medicine into overdrive to the benefit of all.
This article originally appeared on Big Think, home of the brightest minds and biggest ideas of all time.
Opioid prescription policies may hurt those in chronic pain
Tinu Abayomi-Paul works as a writer and activist, plus one unwanted job: Trying to fill her opioid prescription. She says that some pharmacists laugh and tell her that no one needs the amount of pain medication that she is seeking. Another pharmacist near her home in Venus, Tex., refused to fill more than seven days of a 30-day prescription.
To get a new prescription—partially filled opioid prescriptions can’t be dispensed later—Abayomi-Paul needed to return to her doctor’s office. But without her medication, she was having too much pain to travel there, much less return to the pharmacy. She rationed out the pills over several weeks, an agonizing compromise that left her unable to work, interact with her children, sleep restfully, or leave the house. “Don’t I deserve to do more than survive?” she says.
Abayomi-Paul’s pain results from a degenerative spine disorder, chronic lymphocytic leukemia, and more than a dozen other diagnoses and disabilities. She is part of a growing group of people with chronic pain who have been negatively impacted by the fallout from efforts to prevent opioid overdose deaths.
Guidelines for dispensing these pills are complicated because many opioids, like codeine, oxycodone, and morphine, are prescribed legally for pain. Yet, deaths from opioids have increased rapidly since 1999 and become a national emergency. Many of them, such as heroin, are used illegally. The CDC identified three surges in opioid use: an increase in opioid prescriptions in the ‘90s, a surge of heroin around 2010, and an influx of fentanyl and other powerful synthetic opioids in 2013.
As overdose deaths grew, so did public calls to address them, prompting the CDC to change its prescription guidelines in 2016. The new guidelines suggested limiting medication for acute pain to a seven-day supply, capping daily doses of morphine, and other restrictions. Some statistics suggest that these policies have worked; from 2016 to 2019, prescriptions for opiates fell 44 percent. Physicians also started progressively lowering opioid doses for patients, a practice called tapering. A study tracking nearly 100,000 Medicare subscribers on opioids found that about 13 percent of patients were tapering in 2012, and that number increased to about 23 percent by 2017.
But some physicians may be too aggressive with this tapering strategy. About one in four people had doses reduced by more than 10 percent per week, a rate faster than the CDC recommends. The approach left people like Abayomi-Paul without the medication they needed. Every year, Abayomi-Paul says, her prescriptions are harder to fill. David Brushwood, a pharmacy professor who specializes in policy and outcomes at the University of Florida in Gainesville, says opioid dosing isn’t one-size-fits-all. “Patients need to be taken care of individually, not based on what some government agency says they need,” he says.
‘This is not survivable’
Health policy and disability rights attorney Erin Gilmer advocated for people with pain, using her own experience with chronic pain and a host of medical conditions as a guidepost. She launched an advocacy website, Healthcare as a Human Right, and shared her struggles on Twitter: “This pain is more than anything I've endured before and I've already been through too much. Yet because it's not simply identified no one believes it's as bad as it is. This is not survivable.”
When her pain dramatically worsened midway through 2021, Gilmer’s posts grew ominous: “I keep thinking it can't possibly get worse but somehow every day is worse than the last.”
The CDC revised its guidelines in 2022 after criticisms that people with chronic pain were being undertreated, enduring dangerous withdrawal symptoms, and suffering psychological distress. (Long-term opioid use can cause physical dependency, an adaptive reaction that is different than the compulsive misuse associated with a substance use disorder.) It was too late for Gilmer. On July 7, 2021, the 38-year-old died by suicide.
Last August, an Ohio district court ruling set forth a new requirement for Walgreens, Walmart, and CVS pharmacists in two counties. These pharmacists must now document opioid prescriptions that are turned down, even for customers who have no previous purchases at that pharmacy, and they’re required to share this information with other locations in the same chain. None of the three pharmacies responded to an interview request from Leaps.org.
In a practice called red flagging, pharmacists may label a prescription suspicious for a variety of reasons, such as if a pharmacist observes an unusually high dose, a long distance from the patient’s home to the pharmacy, or cash payment. Pharmacists may question patients or prescribers to resolve red flags but, regardless of the explanation, they’re free to refuse to fill a prescription.
As the risk of litigation has grown, so has finger-pointing, says Seth Whitelaw, a compliance consultant at Whitelaw Compliance Group in West Chester, PA, who advises drug, medical device, and biotech companies. Drugmakers accused in National Prescription Opioid Litigation (NPOL), a complex set of thousands of cases on opioid epidemic deaths, which includes the Ohio district case, have argued that they shouldn’t be responsible for the large supply of opiates and overdose deaths. Yet, prosecutors alleged that these pharmaceutical companies hid addiction and overdose risks when labeling opioids, while distributors and pharmacists failed to identify suspicious orders or scripts.
Patients and pharmacists fear red flags
The requirements that pharmacists document prescriptions they refuse to fill so far only apply to two counties in Ohio. But Brushwood fears they will spread because of this precedent, and because there’s no way for pharmacists to predict what new legislation is on the way. “There is no definition of a red flag, there are no lists of red flags. There is no instruction on what to do when a red flag is detected. There’s no guidance on how to document red flags. It is a standardless responsibility,” Brushwood says. This adds trepidation for pharmacists—and more hoops to jump through for patients.
“I went into the doctor one day here and she said, ‘I'm going to stop prescribing opioids to all my patients effective immediately,” Nicolson says.
“We now have about a dozen studies that show that actually ripping somebody off their medication increases their risk of overdose and suicide by three to five times, destabilizes their health and mental health, often requires some hospitalization or emergency care, and can cause heart attacks,” says Kate Nicolson, founder of the National Pain Advocacy Center based in Boulder, Colorado. “It can kill people.” Nicolson was in pain for decades due to a surgical injury to the nerves leading to her spinal cord before surgeries fixed the problem.
Another issue is that primary care offices may view opioid use as a reason to turn down new patients. In a 2021 study, secret shoppers called primary care clinics in nine states, identifying themselves as long-term opioid users. When callers said their opioids were discontinued because their former physician retired, as opposed to an unspecified reason, they were more likely to be offered an appointment. Even so, more than 40 percent were refused an appointment. The study authors say their findings suggest that some physicians may try to avoid treating people who use opioids.
Abayomi-Paul says red flagging has changed how she fills prescriptions. “Once I go to one place, I try to [continue] going to that same place because of the amount of records that I have and making sure my medications don’t conflict,” Abayomi-Paul says.
Nicolson moved to Colorado from Washington D.C. in 2015, before the CDC issued its 2016 guidelines. When the guidelines came out, she found the change to be shockingly abrupt. “I went into the doctor one day here and she said, ‘I'm going to stop prescribing opioids to all my patients effective immediately.’” Since then, she’s spoken with dozens of patients who have been red-flagged or simply haven’t been able to access pain medication.
Despite her expertise, Nicolson isn’t positive she could successfully fill an opioid prescription today even if she needed one. At this point, she’s not sure exactly what various pharmacies would view as a red flag. And she’s not confident that these red flags even work. “You can have very legitimate reasons for being 50 miles away or having to go to multiple pharmacies, given that there are drug shortages now, as well as someone refusing to fill [a prescription.] It doesn't mean that you’re necessarily ‘drug seeking.’”
While there’s no easy solution. Whitelaw says clarifying the role of pharmacists and physicians in patient access to opioids could help people get the medication they need. He is seeking policy changes that focus on the needs of people in pain more than the number of prescriptions filled. He also advocates standardizing the definition of red flags and procedures for resolving them. Still, there will never be a single policy that can be applied to all people, explains Brushwood, the University of Florida professor. “You have to make a decision about each individual prescription.”