I have to admit, I have a knee-jerk negative reaction to phrases like “food is medicine” because of how much this concept has been abused within the alternative medicine sphere. The concept, for example, of “neutraceuticals” – using vitamins and supplements as if they were drugs – is misleading. Alternative practitioners also often behave as if every medical problem is nutritional, and directly lobby to blur the lines between nutrition and medicine. The Dietary Supplement Health and Education Act (DSHEA) of 1994 did this explicitly, regulating herbal products, which are drugs, as if they were food.
But the “Food Is Medicine” (FIM) movement is not this. “Medicine” in this term does not refer to concepts like “neutraceuticals” but rather to the health effects of nutrition. “Medicine” is sort-of a metaphor – meaning that nutrition can be a proper medical intervention, not that it should be thought of like a drug. And they are clear – they are talking about the health effects of good and bad nutrition.
A recent report (two US-based case studies) produced by the FIM group at Tufts University is a good example of the kinds of interventions they are talking about, and what they mean by “medicine”. They studied the potential impact of programs that fall into two basic groups – medically tailored meals (MTMs) and nutritional support. The ultimate point of the data is that public spending on these types of programs is cost effective. They will save more money than the programs cost. They note:
The economic costs of suboptimal diets due to health care spending and lost productivity are estimated at $1.1 trillion each year.
This is partly due to lack of nutrition, but also due to poor diets in general. This counts obesity and diabetes, for example, as chronic conditions impacted by diet. What kinds of interventions are they talking about? Three types of programs address tailoring food to specific chronic illnesses – medically tailored meals, medically tailored groceries, and produce prescription programs. One intervention addresses poor nutrition – nutrition security programs. And then they include population-level interventions dealing with food policies and programs.
Addressing nutrition as part of dealing with a chronic illness is nothing new. What is new about such programs is taking such interventions to the next level – treating it as a prescribed medical intervention (a literal prescription), one that would be covered by insurance and supported by specific programs.
This concept is not entirely new either. It is possible, for example, for physicians to document that a patient has a medical condition that warrants something, like air conditioning, that might otherwise be deemed a luxury. For certain diseases, air conditioning is not a luxury, it is a medical necessity. Often there are state programs or even regulations that force companies to provide services to customers if they are medically necessary. The FIM movement essentially wants to formalize a similar approach to nutrition.
It is pointless to consider whether the government “should” be paying for such things. The government is already paying for health care, which is the largest part of most government’s spending (even the US), and rising. The real question is – are such programs cost effective? These two case studies show that they potentially are.
The first study looks at a national medically tailored meal program. They estimate this will save $38.7 billion in direct healthcare costs, with a net savings of $13.6 billion. Even when just considering direct healthcare costs, such a program would more than pay for itself. This does not even consider the secondary savings from avoiding lost productivity and other government support programs that come along with that. For a net savings, this would also improve the health and quality of life for millions of people.
The second study looked at produce prescription programs. They estimated the total cost of the program at $44.3 billion with direct healthcare savings of $39.6 billion and productivity savings of $4.8 billion. Even though this does not produce a net positive in direct healthcare costs, it is essentially neutral when adding lost productivity. Without adding in lost productivity, it is still a highly cost-effective program, when considering the health outcomes per dollar spent.
Of course, these are estimates. For the first study:
For this case study, a population-level, cohort policy simulation model was created based on nationally representative samples from the 2019 Medical Expenditure Panel Survey (MEPS).
These are standard methods and are validated, but they are still estimates based on models. The next step is to actually implement such programs and measure outcomes. In the US the healthcare system is setup to easily allow for 50 experimental regions – the states. Such programs are implemented at the state level in any case, even if they receive federal funding. This allows for easy experimentation, to see which programs work best and then to replicate them on a national level. At the very least these studies show the potential of such programs – if implemented well, they are likely to be highly cost effective, and even net positive.
The broader concept here, that I think is legitimate, is that there are many potential interventions that have a huge impact on health that are not traditional medical interventions. Many of these we already take for granted, like poison control, public safety, and sanitation (the latter being arguably the greatest health intervention in human history). We should add to this list things like optimal nutrition, gun safety, and addressing health misinformation. And as always, such interventions need to be as evidence and science-based as possible.