Food Is Not Medicine but Maybe It's Healthcare

This transcript has been edited for clarity.

Robert A. Harrington, MD: Hi. I'm Bob Harrington from Weill Cornell Medicine, and I'm here for theheart.org | Medscape Cardiology at the American College of Cardiology (ACC) meeting in Atlanta. I love to use these opportunities at the big meetings to catch up with friends and colleagues, and to talk about things they're doing that might be of interest to the Medscape Cardiology listening audience.

I think we're going to have a terrific conversation for this one. We're going to explore the topic of "food is medicine." That topic has been talked about often online. There have been some recent papers out on the topic. What exactly does it mean? How are people thinking about it? How are patients understanding it?

I can't imagine a better person to join me today than my friend Mitch Elkind. Mitch is the chief clinical science officer at the American Heart Association (AHA) and is part of the large team at the AHA working on a "food is medicine" project.

Mitchell S.V. Elkind, MD, MS, MPhil: First of all, Bob, thank you so much for having me on the show today. Our initiative at the AHA is called Health Care by Food, which is healthcarexfood.org. People can find info online.

Harrington: That's fantastic. I want to start broad.

Elkind: Okay.

'Food Is Medicine' Is a Misnomer

Harrington: What's the concept of "food is medicine"?

Elkind: The concept of "food is medicine" is to provide nutritious food to people who have diet-related conditions. We at the AHA, the ACC, and many organizations have dietary guidelines and nutritional recommendations. We've been in the space for a long time, I think, of telling people what healthy foods are, what people should eat to prevent hypertension, diabetes, heart failure, and so forth.

"Food is medicine" really goes a step further. It's about actually providing that healthy food to people so they can eat it and consume it and try to prevent and manage cardiovascular and other conditions.

Harrington: Making the assumption that if you do that, they will have better outcomes.

Elkind: Correct.

Harrington: That's what we don't know yet.

Elkind: That's what we don't really know yet. That's right. The literature so far is limited and it's suggestive, but doesn't yet prove the point, for example, the way we might want to prove it for a medication, that providing meals and groceries or produce to people works.

Harrington: As I understand it, having been involved with this through the AHA, part of the challenge is that food means many things. There's the nutritional element to it. There's the social element to it. There's a cultural element to it. How does one think about all of that and try to improve health outcomes?

In some ways, you could say, well, don't we know what we're supposed to eat? Hasn't the AHA talked about this for years? In fact, there's still so much we don't know, isn't there?

Elkind: You're right. We know what the healthy and nutritious foods are, for the most part. There is certainly more work that could be done in that area of nutrition. For the most part, we know what people should be eating and we put out those recommendations. The problem is that people are essentially swimming upstream, right? It's very difficult in our modern world for people to easily consume the foods that are best for them. There are many pressures against them in terms of the easy availability of unhealthy foods. They're more affordable and so forth.

"Food is medicine" would be a way to provide those healthy foods directly to people, especially people who are food- or nutrition insecure, meaning they don't have that kind of access.

I think the term "food is medicine" is in some ways really a misnomer because food is not medicine. When we think of medicine, we think of a pill that we might use for heart failure, such as one of your four pillars of heart failure treatments.

Harrington: It's more than that now, but that's okay.

Elkind: I can't keep up. I'm a neurologist so I'm a little bit behind in the cardiology space. The medications are intentionally designed and manufactured to be identical from one pill to the next. You know the exact dose and you're giving it to a person in that same shape and form, regardless of their religion, their background, their cultural preferences, their personal tastes, and so forth.

Food is the exact opposite of that. Food means what your family likes to eat, what your personal tastes and preferences are, maybe what allergies you have, what your religious affiliation is, your background, what people eat in your neighborhood, your region of the country — all of these things.

Harrington: Also, what time of day you work and building food around that.

Elkind: You take medicine on your own in your bathroom maybe, looking in the mirror. You eat food with your family, maybe with friends or an extended group of people. It's much more complicated than medication for that reason, and that's one of the challenges in studying it and proving that this kind of an approach works.

Harrington: The other thing about medicine is, at least in cardiovascular medicine and neurology, we test them as a community. You have a medicine that you think improves somebody's outcome with heart failure, and we do the randomized clinical trial to see if that medicine adds value to everything else that that patient is doing. With food, that's been tried, and those are tough studies to do.

Elkind: They're very tough studies to do. The studies that have been done so far have been a bit mixed. Where the evidence is strongest, perhaps, is in a pre/post design where people have been under a certain regimen and then a "food is medicine" intervention is introduced. By the way, when we're talking about "food is medicine" interventions, there are three general categories that people refer to.

First, there are medically tailored meals, where a meal is delivered to the person's home. A patient goes home from the hospital with, say, heart failure. They have a high risk for recidivism, or getting readmitted within 30 days, so you actually deliver 10 meals a week, let's say, to the person at home.

By the way, we don't know what the right dose is for that so it's, again, exactly opposite of medication. Should it be to give people 10 meals a week? What do they do for their other meals for the week? Are they eating even more unhealthily during the other time? There are many tricky issues there.

There are medically tailored meals; medically tailored groceries, where the groceries are delivered and then people still have to cook it as opposed to a [prepared] meal; and then produce prescriptions, where you essentially give people vouchers that they can bring to the farmer's market or grocery store to get healthy fruits and vegetables.

Harrington: The second two really allow people to put the food into their social context.

Quality Research Lacking

Elkind: It gives them more flexibility, perhaps. The question is whether people will use that as much or not. As you say, there have been some randomized trials, but they've been limited. The randomized trials that exist tend to be small in size. They tend to use variable definitions of dose and duration and these other issues. How long should it go on for? Should it be 3 months, 6 months, or longer than that? That obviously becomes expensive, too.

There are many issues that arise in the research that's been published. So far, we haven't reached the level of a high-quality, rigorous randomized trial that would prove to and ensure that they should cover this the way they might cover a medication.

Harrington: I was going to bring this up as the next point that, as a clinical researcher and clinical trialist, I became aware of this when Nancy Brown, the CEO of the AHA, told me that we really need some traditional trialists to take a look at this space and help us determine the level of evidence to be in a guideline — what would be the level of evidence to convey to an insurer that this evidence is good enough to show this is providing health benefits and you should cover it as part of your insurance plan. That's how you got involved.

Elkind: That's how I got involved, too, as a trialist — not as a food expert or as a nutrition expert, but as somebody who's done trials in the stroke space, like you in the cardiology space. Thank you for being on our team at the AHA and looking into this as well.

We were looking at how we could take that next step to provide high-quality evidence. I think you ask a good question about what kind of evidence will convince policymakers — let's say, the federal government — to cover this under traditional programs like Medicare and Medicaid or private insurers.

Harrington: What would it take for us at the AHA and the ACC to put it in our guidelines?

Elkind: Those may even be different kinds of questions. In other words, the threshold for what one would consider adequate evidence, I think, might differ between a professional society and a payer, what they're going to want. I think we don't really have a great answer to that question.

Going back to this issue of how variable "food as medicine" is, there are all the different issues about the dose and duration and so forth. There are also many different conditions we're talking about. There's heart failure, which may be a prime example of where this could work. There's also hypertension, diabetes, and the outpatient space for prevention. There are different populations, like pregnancy and the postpartum state, where we think this could have a big impact. There are children — we have an obesity epidemic among kids; what can we do there?

Immediately, it becomes many different trials. You're going to need many different studies and different groups of people, really trying to get the evidence — and the best evidence we can — in these different areas that will help policymakers and payers make these kinds of decisions. We're not looking for the one definitive, phase 3 trial, like the stuff presented at ACC on AEGIS-II. I think this is really getting a mass of different types of data that will inform us in making these kinds of decisions, because it's just that complex.

Harrington: To use the phrase "the totality of the evidence," what does that totality look like that would make people comfortable that this is a good thing to do, that, as a society, this is what we should be investing in? Talk about how the AHA got involved with this.

Elkind: The AHA's Health Care by Food Initiative is primarily focused on research. It came about through conversations between the AHA CEO, Nancy Brown, and the CEO of the Rockefeller Foundation, Raj Shah. They recognized that the Rockefeller Foundation has been interested in "food is medicine" for a long time, but they really wanted to go to that next level to get that high-grade clinical trial evidence that would demonstrate to insurers the cost-effectiveness of these types of programs. They recognized that the AHA, thanks to leaders like you and others, has that kind of experience and expertise to conduct those trials. That was the genesis of the initiative.

As we've built it, we've brought in other tremendous folks. Our leader is Kevin Volpp, who is a general internist and behavioral economist with a PhD in economics from the Wharton School at the University of Pennsylvania, who's on faculty at the University of Pennsylvania now. He's been leading this initiative. Kevin has brought tremendous knowledge and insight into this whole process.

Harrington: He's done some very interesting studies over the years, with his background in behavioral economics, of how one influences behavior. He's done very rigorous randomized trials looking at different behavior modifiers. Now he's being asked to put it to this task.

Elkind: That's exactly right. Kevin is also a trialist, but he does a different kind of trial. These trials investigate whether you can pay people to quit smoking or pay people to exercise more. That has really inflected our approach to these kinds of trials. We're not doing those huge, definitive trials yet. What we're focusing on initially is addressing some of these initial behavioral types of questions, so that when we do the big trials, we have the highest likelihood of succeeding.

We're trying to address questions around how to make sure doctors refer patients for "food is medicine" programs. How do you get people to enroll in them? How do you ensure that they adhere to a regimen and that they fulfill their benefits, for example.

Something I learned through doing this is that WIC, the Women, Infants, and Children program, is one of the most successful "food is medicine" programs out there currently. It turns out that only about 50% of the moms who are eligible for that benefit take advantage of it. There are literally billions of dollars left on the table not getting used.

If you try to do a big program in a health insurance program or Medicaid or Medicare, and half of the people aren't even using it, then it's not going to work. We're trying to address some of those questions: How do you make sure that people take those vouchers and go to the farmer's market and bring home the healthy food?

Harrington: And then actually eat it.

Elkind: And eat it. If a basket of vegetables is delivered to the front door and people don't know how to cook them or use them, then that doesn't help either. This goes back to what you were saying before, that you have to include some coaching and nutritional education, culinary education, and so forth for people who take advantage of this.

Harrington: I've been amazed at the diversity of expertise that this project has required. It's not just from, I'll call the science research policy community, because there are some amazing people working on this, but also from industry: the people who deliver food as part of their business model, people who sell food.

I was amazed that Kroger was interested in this. I remember being at the one of the first meetings at Rockefeller; I was sitting next to somebody from Instacart and somebody from Apple, and I was sitting around the table with the CEO of Kroger. That's amazing that everybody wants to see this really evaluated properly.

Elkind: I think that's right. I think people are very excited by it and interested in it. To your point, it really requires many diverse stakeholders and expertise. The way the AHA goes about these things, the research is at the core, but it's surrounded by an envelope of other systems that are helping to facilitate this. Our advocacy team is involved, as are our patient communications team, our health equity team, communications, and marketing. All of these are going to play an important role.

If you get the science done and we prove that this works, it doesn't do any good if it doesn't get implemented on the other side. We have a very diverse team of experts, including experts in cost-effectiveness analysis, experts in implementation science, experts in behavioral science other than Kevin, experts in user experience.

Technology, as you mentioned, is an important part. When you think about food being personalized for people because of all those issues we talked about, such as the different cultural factors involved in food, that means that you can't give the identical meal. You have to come up with different meal plans for different people. Artificial intelligence and other technological advances can play a role in coming up with recipes that are attractive to people but provide the same nutritional value. It's a very complex endeavor, but many people are working on it.

Harrington: You've learned a ton, obviously, being involved in this, because, as you say, it's bringing in people and experiences that you and I don't ordinarily think about. That part's been fun. What have you learned so far about this?

Elkind: We have launched our initiative with a total of about 20 clinical trials that we've started. These are small pilot trials on the order of 18 months in duration to try to answer some of those early.

Harrington: Let's think about that. That's pretty amazing.

Elkind: It's pretty amazing within the first year of this initiative to get 20 trials off the ground. Several of them have already started enrolling patients. We don't have the results from these studies yet, but that's accruing.

Harrington: Give us an example of one or two of the trial questions that are being asked.

Elkind: Some of the questions have to do with coaching, for example. What is the value of coaching on top of providing people with food? If you give people food for 3 months or 6 months and then you pull the plug, then what happens? We're trying to come up with ways of providing that coaching.

One question is whether it should be done in person or virtually. With COVID-19 and telehealth, obviously we have the opportunity to do that. Is a telehealth approach or a virtual, online cooking class, or something along those lines, adequate? Or is individual attention important?

I mentioned the issue of people fulfilling their benefits. There's a question of whether we can increase the likelihood that people fulfill benefits by texting them regularly, for example. It's a simple intervention, but can we improve that? People have shown, for example, that texting increases vaccination rates.

Harrington: I was going to bring that up.

Elkind: It's a similar kind of approach.

Harrington: That's a very public health approach.

Elkind: It's a public health approach. It's a behavioral medicine approach. These are the kinds of things that Kevin has brought into this as well. His center is called the Penn Medicine Nudge Unit. It's all about how to nudge people toward healthy behaviors.

We know what people should eat. Going back to what you said, yes, we know what the nutrition is. We're not thinking so much about what is healthy. We have dietitians and nutritionists who can tell us what people should eat, given their health status. The question is, how do you get people to do that? It's these nudges that we're working on. Those are just some examples of how we do that.

Harrington: This is fascinating. Give the website again if they want to take a look.

Elkind: The website is healthcarexfood.org. We expect that there will be other opportunities for grant funding. Hopefully our research community will be involved. There are many people who are cardiologists who do this kind of research already. This is a well-developed field. There are companies out there that provide medically tailored meals and health systems that provide it for their members. The issue is that it's not yet proven to work so it's not covered broadly.

I'll tell you one other thing. As a neurologist, this is not my bread and butter, so to speak.

Harrington: You're not supposed to be eating bread and butter.

Elkind: Exactly. My bread and olive oil, if you will. When I think about what we're doing with "food is medicine," it's about the behavioral economics, the behavioral science. And that, frankly, is decision neuroscience.

I got into this because I joined the staff of the AHA, this was a big initiative, and because I do clinical trials and so forth. I see it as a way to engage what we're doing at the AHA around brain health, too, because to change behavior, we're talking brain health.

Harrington: That's exactly right. What amazes me about this is the power of the AHA to be able to, as you said, have research at its core, but also everything that goes around it that few organizations can do. It really does allow us, with advocacy and all the other pieces of the AHA, to think about being bold enough to launch this. Kudos to you and the team. It's a terrific idea.

This has been a fabulous conversation. I hope you've enjoyed it. Food is medicine. I've been here with Mitch Elkind, the chief clinical science officer from the AHA. Check out the website and see how you might be able to get involved from the research community perspective.

Mitch, thanks for joining us.

Elkind: Bob, thank you so much for having me.

Robert A. Harrington, MD, is the Stephen and Suzanne Weiss Dean of Weill Cornell Medicine and provost for medical affairs of Cornell University, as well as a former president of the American Heart Association. He cares deeply about the generation of evidence to guide clinical practice. When not focusing on medicine, Harrington dreams of being a radio commentator for the Boston Red Sox fan.

© 2024  WebMD, LLC

Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.

Cite this: Food Is Not Medicine but Maybe It's Healthcare - Medscape - Jun 10, 2024.