Trump wants RFK Jr. to Make America Healthy Again. This is what they need to know about ultraprocessed foods

© Daisy Daisy - stock.adobe.com

One of the first reading assignments students receive in my Introduction to Culinary Medicinecourse taught at the University of Montana is the economics classic “I, Pencil,” written by the master economist Leonard E. Read in 1958. It was immortalized by Nobel laureate economist Milton Friedman, who praised the essay for its clear depiction of the “invisible hand” of the market. Friedman even wrote an afterword for a reprint of the essay, emphasizing its enduring relevance. Although ostensibly about producing a pencil and how no individual has the raw materials and knowledge to make such a simple everyday tool, it teaches a much more profound lesson. And that is the presence of the “timeless and invisible hand” that weaves the interconnections that bind us in so many individual, societal and global ways.

Understanding these interconnections, or relationships, is crucial to unraveling the food-health equation. In our Culinary Medicine approach to solving this conundrum, that means considering the variables in terms of the relationships between them using a systems analysis. An important acknowledgment of the limitations of a systems approach is that it is always an approximation of the underlying structure and is critically dependent upon the boundary limits.

Michael S. Fenster, MD

As we do in our Introduction to Culinary Medicine course, we can begin with relevant economics. In 2023, health care spending in the U.S. rose 7.5%, outpacing the projected annual gross domestic product (GDP) growth rate of 6.1%. Health care spending in the U.S. is set to grow an average of 5.6% a year between now and 2032, outpacing the projected annual GDP growth rate of 4.3% during the same period. This will increase health expenditures to almost 20% of GDP by 2032 (see Figure 1). As a comparison, the USSR had been spending between 12% and 20% of GDP on military expenditures — generally agreed by experts to be a significant contributing factor in the collapse of the Soviet Union — prior to its dissolution at the end of 1991. Whether it is for tanks or tissues, that level of spending is not sustainable. This is especially true in light of our return on investment, particularly when compared with other countries with similar economies (see Figure 2).

Currently, over 90% of our annual $4.5 trillion health care expenditure in the United States is attributable to chronic disability and disease (CDD). Many, if not most, like cardiovascular disease (CVD) — which is the No. 1 cause of death in the United States and many other industrialized nations — are related to dietary choices. Over 90% of all cases of Type 2 diabetes (T2DM) — a potent risk factor for the development of CVD — are considered to be preventable with lifestyle modifications, and over 70% of T2DM cases are believed to arise from poor diet. Diets with high levels of ultraprocessed foods (UPFs) have been correlated to an increased risk of developing Type 2 diabetes by 30% to 74% (for details, see Appendix).

Figure 1: U.S. health care expenditures as a percentage of GDP actual and projected through 2032.

Figure 2: U.S. health care quality rankings versus spending as a percent of GDP compared with similar economies.

In Figure 3, the relationship between U.S. health care expenditures and the prevalence of Type 2 diabetes as a function of time can be appreciated. For the sake of simplicity, I will use Type 2 diabetes as a representative example of CDD throughout this op-ed. In this particular case, the timescale is focused on the last 50 years to acquire a broader perspective of the general trends. A wider perspective can aid in identifying and understanding specific drivers that influence outputs of interest. If we accept the very reasonable hypothesis that the rapid rise in the incidence and prevalence of chronic disabilities and diseases such as Type 2 diabetes is a significant driver of our ever-increasing health care expenditures, a natural point of query is what is driving the increase of CDD like Type 2 diabetes. In other words, what are the flows, stocks (for a further description of stocks, see Appendix), and feedback mechanisms of this particular system?

Multiple epidemiologic studies over the last decade have correlated increased UPF consumption with a significantly increased risk of developing Type 2 diabetes. Figure 3 also demonstrates the rise in the prevalence of Type 2 diabetes against the backdrop of increasing UPF consumption as well as U.S. health care expenditure. The plot displays UPF consumption and Type 2 diabetes prevalence as independent variables from 1970 to 2023. This timeline illustrates a clear upward trend in both UPF consumption and diabetes prevalence, with more pronounced increases beginning around the 1990s and early 2000s. This visualization supports observations that UPF consumption and diabetes prevalence have risen in parallel, especially in recent decades.

Figure 3: U.S. health care expenditure (as % of GDP), Type 2 diabetes prevalence (as % of the population) and ultraprocessed food consumption (as % of total calories). The data on UPF consumption prior to 2000 are based on the best assumption from indirect data sources, as direct measurements are lacking. See the Appendix for details.

In further defining the relationship between total U.S. health care expenditures and Type 2 diabetes, we can specifically examine pharmaceutical costs, which contribute roughly 20% to the yearly health care expenditure total, with antidiabetic medications accounting for between 5% and 10% of total pharmaceutical expenditures. It is important to note that different sources may report varying figures due to differences in data collection methodologies and definitions, so we are looking for consistency in the data on which to base reasonable assumptions. If we further drill down specifically at pharmaceutical revenues derived from the sale of antidiabetic drugs, we can begin to unravel potential positive feedback mechanisms. The time course and relationship between the prevalence of Type 2 diabetes and the amount of pharmaceutical revenues derived from the sale of antidiabetic medications can be seen in Figure 4. The data regarding the increasing consumption of UPF, growing prevalence of Type 2 diabetes and rising pharmaceutical revenues suggest an interdependence of the variables, a hallmark of systems dynamics.

Figure 4: The prevalence of Type 2 diabetes and revenues from antidiabetic pharmaceutical drugs over time.

One of the very interesting things about natural systems, including individual human behavior, human social systems, and systems surrounding food and health, is that they rarely operate in a linear, predictable, the-whole-equals-the-sum-of-the-parts way. Understanding such complexities from a systems perspective is in complete contradistinction to how we have approached food-health relationship over the last half-century and continue to do so. Our current method is grounded in the linear reductionism of nutritionism that became entrenched in the 1980s and continues to emphasize only the preferential consumption of selected nutrients and avoidance of others as the determinants of health and wellness or disability and disease.

An information problem

In 2009, Professor Carlos Monteiro, MD, PhD, and his colleagues at the University of São Paulo in Brazil introduced the NOVA classification. They proposed an examination of food and drink based not by nutrients or energy density, but by the amount of processing. Today it remains the gold standard in research to identify and classify UPFs (also known as NOVA classification group 4). Graduate students often remark that if the public were only made more aware of, or were educated at an earlier age about, the NOVA classification then there would be no crisis of diet-related CDD. Current U.S. guidelines and recommendations do not address the presence or absence of UPFs in the diet. Although the education of both health care providers and consumers is undoubtedly part of the solution, just focusing on education alone as a panacea also misses the mark.

To suggest that all we need to do is increase the budget for teaching and shout the message ignores the realities that impact dietary choices. The current nutritionism paradigm assumes that given a choice between hyper-palatable (and some might argue addictive), cheap and convenient comestibles and better health down the road, people will always make the rational choice: defer immediate gratification, bypass the drive-through and head homeward (although perhaps a bit hangrier in the process).

I believe it is safe to say that the folks waiting in the long line at the drive-through or ordering the triple meat lover’s pizza special of the day are under no illusions that they’re making a healthful choice. Human behavior is an imperfect variable and thus introduces defects into models that assume a stable equilibrium, like the classical model of economics.

We are all familiar with that prototypical American tourist who travels overseas to a quaint but remote village. After entering a particular establishment, they find the shopkeeper understands no English. Their solution is to simply speak louder. And that is what researchers and governmental agencies continue to do: repeat the same message but increase the volume to 11. But it is not a volume problem; it is an information problem.

Yet, much like the antiquated classical economic approach that tried to force fit perfectly rational human behavior within the model of a stable economic supply-demand equilibrium, modern medicine all too often uses the same template in addressing health and disease. Without jumping too far down the rabbit hole, in our contemporary Culinary Medicine approach, we use methods to analyze the food-health relationship analogous to those currently utilized in complex systems models of economics and other disciplines. One of those is understanding the nonlinear nature of various agents and events. Looking closely at Figure 5, we can see that the relationship between UPF consumption, pharmaceutical revenues from antidiabetic medications and the prevalence of Type 2 diabetes is nonlinear.

Figure 5: Ultraprocessed food consumption (as % of total calories), pharmaceutical revenues from antidiabetic drugs to treat Type 2 diabetes (in billions of dollars), and the prevalence of Type 2 diabetes (as a percent of the population). These relationships are clearly nonlinear.

Diagnosis

A power law fit is beneficial when analyzing relationships where one variable grows at a nonlinear rate relative to another. In economics and health care, small perturbations can lead to substantial impacts over time. For example, certain behaviors, e.g., increased UPF consumption, can lead to widespread and disproportionately large impacts, e.g., increased cases of Type 2 diabetes and medical costs, especially in larger populations. The curves, as shown in Figure 5, as real-world data often do, lend themselves to power law fitting (for details, see Appendix).

What’s to eat?

However, the system we have described to date is incomplete, lacking the positive and negative feedback loops that can influence flows and stocks. There is no doubt that UPF is big business on a global scale, and that is a potentially powerful positive feedback loop. To put it in perspective, let’s look at only one fast food purveyor. For that example, we will use the fast-food giant McDonald’s. In 2023, McDonald’s revenue was approximately $23 billion. If McDonald’s were a country, that revenue would place it between Iceland and Estonia in terms of gross domestic product, making it around the 100th largest economy in the world. And that is only one fast-food company.

If we look at international purveyors of UPFs, the largest is Nestlé, which has an annual revenue of over $100 billion. Applying the previous criteria, if Nestlé were a country, its revenue would place it between Kenya and Puerto Rico, making it around the 60th largest economy in the world. When such forces align, they are incredibly powerful in their global influence.

But fast food, junk food and UPFs in general also have significant adverse effects that should act to decrease their consumption despite being cheap, convenient and addictively tasty. These include the increased risks of developing CDD, e.g., Type 2 diabetes, associated with UPF consumption. Beyond the individual morbidity and mortality, on which it is difficult to place a cost on human suffering, there are measurable health care costs. Type 2 diabetes management alone costs approximately $327 billion annually. CDD increases costs through medication, long-term care and frequent medical visits. This impacts public health insurance programs, leading to higher premiums across the board as insurance companies compensate for higher claims by adjusting costs. Public health programs like Medicare and Medicaid must divert more funds to managing diet-related diseases, reducing resources available for other health care areas. This impacts overall health care quality and availability and puts budget strains on these systems.

In the classic film “The Ten Commandments,” starring Charlton Heston, there is a scene where Heston, as Moses, is being held accountable for allowing the slaves to get grain from the royal granary. As each accusation mounts, Yul Brynner, as Rameses, puts a golden weight on a scale. After the last accusation is levied and the scale sits askew, Moses puts a brick on the scale, causing it to crash to the table with a bang. Moses replies, “A city is built of brick, Pharaoh. The strong make many, the starving make few. The dead make none.” Health issues arising from high UPF intake contribute to increased absenteeism and presenteeism (working while sick). An ill and depleted workforce reduces economic productivity and economic output.

Chronic diseases like Type 2 diabetes are associated with a host of disabilities, including cardiovascular disease, chronic kidney disease, peripheral vascular disease, stroke, blindness and many others. These disabilities can force people into early retirement or long-term disability programs. This can shrink the labor force and increase the financial burden on social welfare systems. The shift in spending toward health care costs reduces consumers’ disposable income for other goods and services, potentially reducing economic growth in other sectors.

High UPF consumption has reshaped the economics of the food industry. While UPF producers benefit from high UPF consumption, it may harm producers of wholesome foods like fruits and vegetables, impacting agricultural markets. This can influence governmental spending and associated public health campaigns to mitigate the detrimental health effects of high UPF consumption, which can add to the bottom line. Even if UPFs are taxed, there are policy implementation costs and other associated administrative and regulatory costs that, once established, are difficult to eliminate.

These negative feedback loops might be expected to check the growth of UPF consumption. However, as I mentioned at the beginning of this op-ed, setting boundaries is critical to understanding the functioning of the system. And as the devil is in the details, we can begin to extricate where to move those boundaries by highlighting a few examples.

A lack of transparency and financial conflicts of interest plague the Department of Agriculture committee that develops and approves the dietary guidelines. As reported by Nina Teicholz in the journal Public Health Nutrition, the 2020-2025 dietary guideline advisory committee had conflicts involving 19 out of 20 members. Four of the six members of the Birth-to-24 Months subcommittee had ties to infant formula/baby food companies.

A recent paper by Scott Slater and colleagues highlighted the prominent role of multistakeholder institutions (MIs). MIs are the result of a movement in global food governance toward “a more decentralized, market-orientated and corporate-engaged system of governance, involving a more diverse range of public and private actors.” This approach has been promoted by the United Nations and through “powerful business associations, most prominently the World Economic Forum (WEF) through its Global Redesign Initiative, reflecting a broader vision of stakeholder capitalism.” The result has been “the proliferation of prominent multi-stakeholder initiatives, partnerships, platforms, and roundtables involving the world’s largest transnational food corporations, their affiliated business interest groups, along with multilateral agencies, international non-governmental organizations, national governments, and research institutions.” Of the 45 global food system MIs, the UPF industry and affiliated interest groups hold 43.8% of the total 601 board seat positions. Just four companies — Unilever, Nestlé, PepsiCo and Coca-Cola — controlled 64 or roughly 11% of those seats.

Big Pharma’s influence

But it is not just purveyors of fast food and junk food and manufacturers of UPFs that are involved. And here is where the goalposts widen: Pharmaceutical companies are also significantly represented, and there is a clear reason. Rising UPF consumption indirectly boosts demand for medications, e.g., those used to treat Type 2 diabetes, hypertension and cholesterol-related conditions. Pharmaceutical revenues have shown a high correlation with UPF consumption trends, reflecting this demand. The global pharmaceutical industry has experienced significant growth over the past few decades as UPF consumption has increased. In 2001, worldwide pharmaceutical revenues were approximately $390 billion. By 2023, this figure had risen to around $1.48 trillion, reflecting the industry’s expansion and intensified demand from a population that is increasingly offered “a pill for every ill” as the predominant medical treatment.

An analysis of the power law fits for the plots of pharmaceutical revenues from antidiabetic medications, UPF consumption and Type 2 diabetes prevalence underscores a strong association in their growth trajectories, suggesting that these trends may be potentially interconnected causally in their broader economic and health impacts (for details see Appendix). These results show that all three variables are closely related, with the strongest correlations observed between diabetes prevalence and both UPF consumption and pharmaceutical revenues. This suggests that as UPF consumption increases, diabetes prevalence rises, resulting in increased pharmaceutical revenues.

The purpose here is not to establish causality beyond any doubt but to provide a reasonable basis based on the available data and analysis to make the assumptions upon which a systems model can be predicated.

A linked system?

Examining UPF consumption, CDD (as represented by Type 2 diabetes in this discussion) and pharmaceutical company revenues, the result suggests a linked system. The output of massive UPF consumption is CDD. CDD is the input flow for pharmaceutical company revenues. Within this linked system, there is essentially no economic reward (especially for patients and individuals) for returning someone to and maintaining good health.

Our current health care industry, with respect to CDD, is based on disease management. When someone breaks a leg, we do not put them on crutches for the rest of their life to manage their infirmity. A surgeon operates and repairs the trauma. This is followed by a program of ancillary health services and rehab if needed. Barring complications, the system returns the patient to a preoperative level of function.

There is no such goal in the current treatment model of disease management, despite widespread evidence that a significant percentage of CDD can be prevented, treated and reversed with lifestyle-focused interventions such as diet. Such measures are often not even a serious consideration in the treatment plan, which all too often focuses on a lifetime crutch of various pharmaceuticals. Again, borrowing from systems analysis, the input flow into the disease management process or stock is people with chronic diseases. People who die quickly may bring in the short term a significant “purchase” — think of the bill for someone receiving maximum therapy in the last 24 hours of life in an intensive care unit setting — but they are not long-term, repeat customers. That requires a steady flow of lifetime consumers, in other words, those with CDD.

Interventions

Increasing UPF consumption creates a complex economic impact. While it generates revenue for the food and pharmaceutical industries, it also leads to rising health care costs, productivity losses, and a significant financial burden on individuals and public health systems, not to mention the human suffering of increased morbidity and mortality. Initiating change to improve health care outcomes begins by defining and acknowledging the system and its stocks, flows, and positive and negative feedback loops.

Figure 6: A simplified overview of the ultraprocessed food (UPF), Type 2 diabetes (a CDD) and pharmaceutical revenue system as presented in this op-ed. Within the system boundaries, producers of UPFs manufacture their products (from flows outside of this system boundary) to fill the UPF stock. The stock of UPFs is depleted through the purchase and consumption of products by consumers. Some of these consumers develop CDD (e.g., Type 2 diabetes, T2DM) and fill the stock of pharmaceutical revenues. Pharmaceutical revenues are partially depleted by flows outside of the system boundary (salaries, operating expenses, advertising, marketing, R&D, etc.). Part of the revenues are channeled into agencies like multistakeholder institutions (MIs). Through actions like those involving global food governance, these agents act as a positive feedback loop for the manufacture of more UPFs. When UPFs are purchased, that generates revenue for their manufacturers, which is another positive feedback loop for their manufacture. Part of UPF manufacturer revenues is also fed back into MIs, creating a third positive feedback loop.

A starting point for a systems analysis of the UPF, Type 2 diabetes (CDD) and pharmaceutical revenue relationship, as presented in this op-ed, can be seen in Figure 6. Within the system boundaries, producers of UPFs manufacture their products (from flows outside of this system boundary) to fill the UPF stock. The stock of UPFs is depleted through the purchase and consumption of products by consumers. Some of these consumers develop CDD (e.g., Type 2 diabetes) and fill the stock of pharmaceutical revenues. Pharmaceutical revenues are partially depleted by flows outside of the system boundary (salaries, operating expenses, advertising, marketing, R&D, etc.). Part of the revenues are channeled into agencies like MIs. Through actions like those involving global food governance, these agents act as a positive feedback loop for the manufacture of more UPFs. When UPFs are purchased, that generates revenue for their manufacturers, which is another positive feedback loop for their manufacture. Part of UPF manufacturer revenues is also fed back into MIs, creating a third positive feedback loop. These positive feedback loops are highlighted in Figure 7.

Figure 7: A simplified representation of the UPF, CDD and pharmaceutical revenues system highlighting positive feedback loops that act to increase system stocks.

When viewed this way, one of the glaring anomalies, particularly with respect to economic drivers, is the lack of a negative feedback loop to impact and put the brake on the manufacture of UPFs. Likewise, there is a lack of effective agency in depleting the stock of consumers of UPFs through drivers associated with positive dietary and health-related choices.

With a system on the brink of unsustainability, barreling toward the tipping point with no brakes, can an intervention at this point make a difference? One of the advantages of approaching the problem from a systems perspective is that, well, they are systems.

And that means we can look across disciplines and apply solutions from other areas as diverse as economics and ecosystems.

Where are our wolves?

The last of the wolves was eradicated from Yellowstone around 1926. That was done intentionally as part of a predator control program based on the idea that this would protect

game species such as deer and elk. The program was wildly successful and worked to such an extent that 14 wolves were reintroduced in 1995 to control the resultant population explosion of deer and elk. The objective was simply for the wolves to hunt the deer and elk and reduce their populations. However, as complex systems, like ecosystems, are wont to do, there were consequences no one foresaw.

It started with wolves hunting deer and elk and reducing their populations. But the presence of wolves also changed the behavior of the deer and elk. They began to avoid lowland areas like riverbanks, where they were easy prey for the wolf packs. In the absence of overgrazing from the deer and elk, diverse species of plants, including aspen and willow trees, thrived. With the production of berries and seeds came more insects, attracting various songbirds. The new environments provided shelter for rabbits and mice that sustained other predators like weasels, hawks and foxes. As the trees grew, the beaver, extinct for many decades from Yellowstone, returned. Their dams provided habitats for otters, muskrats, water birds and reptiles. These changes were accompanied by decreased erosion along the riverbanks. As the course of the rivers stabilized, the channels became deeper and more fixed. The course of the very rivers themselves had been changed. The reintroduction of the wolves into Yellowstone started a cascade of events among the interconnected and interdependent agents of the ecosystem that subsequently altered the very landscape of that ecosystem.

So, where are our wolves?

Figure 8: The effect of several potential negative feedback loops is to apply stabilization to the system.

Healthy profits

Figure 8 provides several potential intervention points for stabilizing the system by introducing negative feedback loops. For example, an opportunity exists to introduce a solution that leverages the interrelationships, allowing us to move beyond the current status quo of disease management. A potential point of action is those consumers of UPFs who develop a chronic disability or disease such as Type 2 diabetes. There is the potential for creating programs and policies that encourage nonpharmacological remedies, such as diet-based solutions that reduce the consumption of UPFs over the long term. Positive economic reinforcers must accompany the implementation of such propositions. They must not only deliver results but also provide a financially viable model for the providers. The result would be a percentage of people with CDD who do not require a lifetime of medication. This negative feedback affects the input flow for pharmaceutical revenues. This is what I refer to as an economic intervention for healthy profits.

Within the ecosystem of Yellowstone Park, the wolf is a predator, taking the lives of deer and elk. But it is also a giver of life, providing an opportunity for other species like birds, foxes, rabbits, reptiles, beavers and fish to thrive. It is neither villain nor hero, as those are human judgments; it is simply a wolf. A functioning, profitable and healthy pharmaceutical industry is critical to maintaining and improving the current level of health care services. Creating a feedback loop of healthy profits that also includes strategic business opportunities for pharmaceutical industry involvement promotes diversification from a “pill for every ill” model.

Another opportunity exists to impact the percentage of consumers of UPFs who subsequently develop chronic disability and disease. This is accomplished by presenting alternatives to UPF consumption through the creation of regional food nodes. Food cultures and consumption are local and regional phenomena. Creating and supporting regional food nodes or resources could offer current consumers of UPFs alternatives to diets mainly composed of junk food, fast food and other ultraprocessed offerings. This could be realized through several mechanisms. Local employers with cafeterias could instruct their food service staff to serve less ultraprocessed meals and create menus with more healthful local and regional ingredients. Education at the local and regional levels could be reinforced through culinary medicine-focused cooking demonstrations, lectures and presentations held during food service hours and offered as after-hours events.

Regional food nodes could also incorporate the existing networks of regional health systems and extend offerings, including educational opportunities for health care professionals and the public. A natural extension of such a systems approach is to include regional grocery store chains, many of which include pharmacies. Involvement as a regional food node supplying ingredients for purchase by consumers could help offset any potential losses from the sale of pharmaceuticals.

Finally, creating firewalls through increased transparency, legislation and consumer advocacy could help limit the untoward influence of entities like international MIs on global food governance. Increased disclosure and reporting could identify potential conflicts of interest involving research, publication and policy mandates.

I am hopeful that, at the very least, this article has served as ample food for thought. A systems perspective allows us to identify, understand and potentially leverage the interdependencies and interconnections that guide the “invisible hand.” Nonetheless, as we move forward, we must do so with a clear understanding of the limitations of such an approach. At the beginning of our discussion, looking to employ a systems methodology to unravel the intricacies of our individual food-health relationship, we observed that even at its very finest, such a methodology is only an approximation, albeit an incredibly powerful and useful one.

We have set our Healing Table and sent the invitations far and wide:

• To pharmaceutical and medical device manufacturers
• To healthcare systems and governmental agencies
• To food manufacturers and grocery stores
• And to every person who cares about their health and their loved ones.

Let the feast — and the conversation — begin!

Michael Fenster, MD, FACC, FSCA&I, FRSM, MIANE, PEMBA, also known as Chef Dr. Mike, is a board-certified interventional cardiologist, a professional chef and a professor of culinary medicine at The University of Montana where he teaches one of the country's leading courses on food and health. He also hosts the PBS program, “House Calls with Chef Dr. Mike" and is the author of four books. He is CEO of BridgMed.health which uses culinary medicine via an interactive app to guide people towards healthier dietary choices. He can be reached by email at: michael.fenster@umt.edu

Appendix

A brief note on curves, correlations and causation

As plotted in Figure 5, the curves were subjected to fitting algorithms, both for exponential and power law fits. Power law fits were much better solutions, with a squared sum residual value of 1.43 for the pharmaceutical revenues from antidiabetic medications curve, a value of 36.3 for the UPF consumption curve, and a value of 1.15 for the Type 2 diabetes prevalence curve.

The Pearson correlation analysis measured the strength and direction of the relationship between two continuous variables. It was used on the power fit curves for simplicity of presentation. The Pearson correlation coefficient between UPF consumption and diabetes prevalence is approximately 0.985, with a highly significant P value. This high positive correlation indicates a strong alignment in the growth patterns of UPF consumption and diabetes prevalence over time, with the relationship being statistically significant. This suggests that the trends in UPF consumption and diabetes prevalence share a similar upward trajectory, following the power law growth dynamic and suggesting a close relationship where increases in UPF consumption are strongly associated with increases in diabetes prevalence. Such a finding is in agreement with the prevailing literature. The Pearson correlation coefficient between Type 2 diabetes prevalence and diabetic pharmaceutical revenues is 0.967, with a highly significant P value. This indicates a strong correlation between these two variables that is statistically significant and unlikely to occur by chance.

A regression-based path analysis also provided insights into the relationships among UPF consumption, diabetes prevalence and pharmaceutical revenues. As you would expect, a strong relationship was demonstrated between the prevalence of Type 2 diabetes and pharmaceutical revenues derived from antidiabetic medication, with pharmaceutical revenues increasing by approximately $1.34 billion for every 1% increase in diabetes prevalence. There was also a significant relationship between UPF consumption and the prevalence of Type 2 diabetes, with diabetes increasing by 0.42% for every 1% increase in UPF consumption. This is slightly more conservative but within the range found within the literature of an increase of approximately 1.5% in Type 2 diabetes prevalence for every 1% increase in UPF consumption.

This model suggests that UPF consumption has a causal role in increasing diabetes prevalence, which is a primary driver of pharmaceutical revenues derived from the sale of antidiabetic medications. It can be visualized as follows:

There are many other disease drivers of pharmaceutical revenues. Still, this illustration demonstrates how, at least in the case of this one chronic disease (T2DM), the illness, UPF consumption and pharmaceutical revenue are all potentially interconnected in a system of stocks, flows and feedbacks.

A note on UPFs in the risk of Type 2 diabetes

There are a number of studies correlating an increased consumption of UPFs with an increased risk of chronic disability and disease. There are fewer studies addressing specific conditions such as Type 2 diabetes, and the degree of risk found varies depending on how the data were collected and analyzed and the composition of the study group. Nonetheless, two particularly well-done studies (Delpino, et al., 2022) (Moradi, et al., 2021) that bracket a reasonable range of risk were used for this op-ed.

A note on stocks within systems descriptions

In the context of systems dynamics, stocks can also be described as accumulations or state variables. They represent the quantities that are built up or depleted over time, serving as the “memory” of the system. Stocks are influenced by the rates of flows, inflows that add to the stock and outflows that reduce the stock.

A note on UPF consumption prior to 2000

As mentioned in the text, a backward extrapolation to estimate the historical data was employed using a combination of historical dietary trends and growth estimates based on documented increases in UPF intake in later years. Accessing precise historical data on UPF consumption from 1970 to 2000 is challenging due to the limited availability of direct data. However, existing studies and reports were used to refine estimates.

The approximation used is as follows:

1. Historical growth patterns: Data (consumer) show that UPF consumption in the United States has steadily increased since the 1950s.

2. USDA food availability data: The U.S. Department of Agriculture provides data on food availability per capita, which serve as a proxy for consumption trends over time. These data include categories like added sugars, fats and processed foods, offering insights into dietary shifts.

3. Studies on processed food trends: Research indicates a significant increase in processed and ultraprocessed food consumption in the United States over the past two centuries, especially sugar, white flour, white rice, vegetable oils and ready-to-eat meals.

4. Known data points for reference:

• Data from the early 2000s suggest that approximately 53% to 55% of total caloric intake in the U.S. is from UPFs.
• From 2001 to 2023, there was an increase from approximately 53.5% to 66% in UPF consumption, indicating an approximate 0.96% annual growth.

5. Backward projection:

Assuming a consistent growth rate from 2000 backward to 1970 may not capture all nuances (and probably does not, as early growth could have been faster with the rise of industrial food production). Still, it gives a general sense of the curve and growth in the earlier years. A reasonable model could assume a slightly higher growth rate in earlier years, perhaps higher than the 0.76% per year used to account for the following:

• 1970s: The rise of convenience foods and fast-food chains likely led to increased UPF consumption. Estimates suggest UPF intake was around 30% to 35% of total caloric intake during this decade.
• 1980s: With the proliferation of processed snacks and ready-to-eat meals, UPF consumption may have risen to approximately 40% to 45%.
• 1990s: The trend continued, with UPF intake potentially reaching 50% to 55% by the decade’s end.

That led to the following estimates:

• 1970: 30.0%
• 1980: 37.6%
• 1990: 45.2%
• 2000: 52.7%

These estimates suggest a gradual increase in UPF consumption over the decades, with the rate accelerating slightly as UPFs became more prevalent in the food system. This model assumes a backward growth rate of approximately 0.76% per year, which aligns with trends in convenience food adoption during these years. This regressive estimation aligns with prior work done by Moubarac et al (2014) examining Canadian dietary trends, which reported an estimated 43.7 % consumption of UPFs in 1969. Substituting this number as the starting point of UPF consumption in 1970 does not change the power law curve fits or the statistically significant relationship among any of the variables.

A note on pharmaceutical revenues from antidiabetic medications

In 2024, the global antidiabetic medication market, as represented by the sales of the top 10 pharmaceutical firms, was approximately $64 billion (Mikulic, 2024). The U.S. represents approximately 44% of the global market, or roughly $28 billion. These values are in agreement with the U.S. diabetes market estimate of 29 billion in 2023 by Research and Markets (2024). In the context of this discussion, pharmaceutical revenues from antidiabetic medications represent the income from the sales of antidiabetic drugs and exclude licensing, royalties or income related to nondrug antidiabetic operations, as this is more relevant to understanding the direct economic impact of Type 2 diabetes and pharmaceutical companies.

References

1. Statistics about diabetes. American Diabetes Association. Nov. 2, 2023. https://diabetes.org/about-diabetes/statistics/about-diabetes
2. Adams J, Hofman K, Moubarac JC, Thow AM. Public health response to ultra-processed food and drinks. BMJ. 2020;369:m2391. doi:10.1136/bmj.m2391
3. Baraldi LG, Martinez Steele E, Canella DS, Monteiro CA. Consumption of ultra-processed foods and associated sociodemographic factors in the USA between 2007 and 2012: evidence from a nationally representative cross-sectional study. BMJ Open. 2018;8(3):e020574. doi:10.1136/bmjopen-2017-020574
4. New evidence links ultra-processed foods with a range of health risks. News release. The BMJ. May 29, 2019. https://www.bmj.com/company/newsroom/new-evidence-links-ultra-processed-foods-with-a-range-of-health-%20risks/
5. Brown S. Most dietary guideline advisors have ties to food and pharma industries, study finds. Verywell Health. March 28, 2022. https://www.verywellhealth.com/dietary-guidelines-committee-conflicts-of-interest-5223556
6. Centers for Disease Control and Prevention. National diabetes statistics report 2020: estimates of diabetes and its burden in the United States. https://diabetesresearch.org/wp-content/uploads/2022/05/national-diabetes-statistics-report-2020.pdf
7. Centers for Disease Control and Prevention. National diabetes statistics report: estimates of diabetes and its burden in the United States. November 29, 2023. https://stacks.cdc.gov/view/cdc/148231
8. Centers for Disease Control and Prevention. National diabetes statistics report: estimates of diabetes and its burden in the United States. May 15, 2024. https://www.cdc.gov/diabetes/php/data-research/index.html
9. Gwira JA, Fryar CD, Gu Q. Prevalence of total, diagnosed, and undiagnosed diabetes in adults: United States, August 2021-August 2023. Centers for Disease Control and Prevention. November 2024. https://www.cdc.gov/nchs/products/databriefs/db516.htm
10. Julia N. Diabetes statistics: facts & latest data in the US (2024 update). CFAH. January 11, 2024. https://cfah.org/diabetes-statistics/
11. Fenster MS. Nutritionism: at the crossroads. Healthieyoo. July 6, 2024. https://www.healthieyoo.com/nutritionism-at-the-crossroads/
12. Ultra-processed foods: a global threat to public health. Global Food Research Program. May 2021. https://globalfoodresearchprogram.org/wp-content/uploads/2021/04/UPF_ultra-processed_food_fact_sheet.pdf
13. Godman H. More evidence that ultra-processed foods harm health. Harvard Health Publishing. June 1, 2024. https://www.health.harvard.edu/nutrition/more-evidence-that-ultra-processed-foods-harm-health
14. Gotkine E. Prevalence of diabetes 15.8% in US adults from 2021 to 2023. Medical Xpress. November 6, 2024. https://medicalxpress.com/news/2024-11-prevalence-diabetes-adults.html
15. Grover E. The overconsumption of ultra-processed foods in the United States. Ballard Brief. March 2024. https://ballardbrief.byu.edu/issue-briefs/the-overconsumption-of-ultra-processed-foods-in-the-united-states
16. Harvard University. National Diabetes Statistics Report 2024. https://repository.gheli.harvard.edu/repository/11854/
17. Hitt Nichols E. First-ever birth-24 month dietary guidelines: deliberations and complications. Nutrition Coalition. March 14, 2024. https://www.nutritioncoalition.us/2020-dietary-guidelines-info/birth-to-24-months-guidelines
18. Juul F, Hemmingsson E. Trends in consumption of ultra-processed foods and obesity in Sweden between 1960 and 2010. Public Health Nutr. 2015;18(17):3096-3107. doi:10.1017/S1368980015000506
19. Juul F, Lin Y, Deierlein AL, Vaidean G, Parekh N. Trends in food consumption by degree of processing and diet quality over 17 years: results from the Framingham Offspring Study. Br J Nutr. 2021:126(12):1861-1871. doi:10.1017/S000711452100060X
20. Juul F, Parekh N, Martinez-Steele E, Monteiro CA, Chang VW. Ultra-processed food consumption among US adults from 2001 to 2018. Am J Clin Nutr. 2022;115(1):211-221. doi:10.1093/ajcn/nqab305
21. Lane MM, Gamage E, Du S, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024;384 :e077310. doi:10.1136/bmj-2023-077310
22. Mande J. Processed foods are making us sick: it’s time for the FDA and USDA to step in. Harvard Public Health. March 1, 2023. https://harvardpublichealth.org/policy-practice/processed-foods-make-us-sick-its-time-for-government-action/
23. Mialon M, Serodio PM, Crosbie E, Teicholz N, Naik A, Carriedo A. Conflicts of interest for members of the US 2020 dietary guidelines advisory committee. Public Health Nutr. 2022;27(1):e69. doi:10.1017/S1368980022000672
24. Diabetes statistics. National Institute of Diabetes and Digestive and Kidney Diseases. Updated January 2024. https://www.niddk.nih.gov/health-information/health-statistics/diabetes-statistics
25. Americans are eating more ultra-processed foods. News release. NYU. October 14, 2021. https://www.nyu.edu/about/news-publications/news/2021/october/ultra-processed-foods.html
26. Pratt E. Ultra-processed foods linked to heart disease, cancer, and 30 other health conditions. Medical News Today. March 1, 2024. https://www.medicalnewstoday.com/articles/ultra-processed-foods-linked-to-heart-disease-cancer-and-30-other-health-conditions
27. Putnam J, Gerrior S. Trends in the U.S. food supply, 1970-97. U.S. Department of Agriculture. https://www.ers.usda.gov/webdocs/publications/42215/5836_aib750g_1_.pdf?v=0
28. Slater S, Lawrence M, Wood B, Serodio P, Van Den Akker A, Baker P. The rise of multi‑stakeholderism, the power of ultra‑processed food corporations, and the implications for global food governance: a network analysis. Agric Human Values. Published online June 11, 2024. doi:10.1007/s10460-024-10593-0
29. Statista Research Department. Diabetes in the United States - statistics & facts. Statista. October 28, 2024. https://www.statista.com/topics/8038/diabetes-in-the-us/
30. Elflein J. The health effects of ultra-processed food - statistics & facts. Statista. September 13, 2024. https://www.statista.com/topics/12722/the-health-effects-of-ultra-processed-food/
31. Teicholz N. 2020 dietary guidelines repeating past mistakes, lacks scientific rigor. Nutrition Coalition. October 31, 2019. https://www.nutritioncoalition.us/news/2020-dietary-guidelines-repeating-past-mistakes
32. Summary findings: food price outlook, 2024 and 2025. U.S. Department of Agriculture Economic Research Service. Updated November 25, 2024. https://www.ers.usda.gov/data-products/food-price-outlook/summary-findings/
33. Wang L, Martínez Steele E, Du M, et al. Trends in consumption of ultraprocessed foods among US youths aged 2-19 years, 1999-2018. JAMA. 2021:326(6):519-530. doi:10.1001/jama.2021.10238
34. Winick LA, Yamaguchi A, Mooney T. The federal loophole that allows food companies to decide what’s safe for you to eat. CBS News. September 7, 2024. https://www.cbsnews.com/news/ultra-processed-foods-fda-health-safety/

Additional references

1. Aboulenein A. U.S. healthcare spending rises to $4.8 trillion in 2023, outpacing GDP. Reuters. June 13, 2024. https://www.reuters.com/business/healthcare-pharmaceuticals/us-healthcare-spending-rises-48-trillion-2023-outpacing-gdp-2024-06-12/
2. Centers for Disease Control and Prevention. Prevalence of total, diagnosed, and undiagnosed diabetes in adults: United States, August 2021-August 2023. https://archive.cdc.gov/www_cdc_gov/diabetes/library/reports/reportcard/national-state-diabetes-trends.html
3. Delpino FM, Figueiredo LM, Bielemann RM, et al. Ultra-processed food and risk of type 2 diabetes: a systematic review and meta-analysis of longitudinal studies. Int J Epidemiol. 2022;51(4):1120-1141. doi:10.1093/ije/dyab247
4. Mikulic M. Top 10 pharmaceutical companies by global anti-diabetic revenue in 2017 and 2024.Statista. December 10, 2024. https://www.statista.com/statistics/309727/-anti-diabetic-revenue-by-top-ten-pharmaceutical-companies-worldwide/
5. Moradi S, Hojjati Kermani MA, Bagheri R, et al. Ultra-processed food consumption and adult diabetes risk: a systematic review and dose-response meta-analysis. Nutrients. 2021;13(12):4410. doi:10.3390/nu13124410
6. Mordor Intelligence. US diabetes care drugs market size & share analysis - growth trends & forecasts (2025-2030). https://www.mordorintelligence.com/industry-reports/united-states-diabetes-drugs-market
7. Moubarac JC, Batal M, Martins AP, et al. Processed and ultra-processed food products: consumption trends in Canada from 1938 to 2011. Can J Diet Pract Res. 2014;75(1):15-21. doi:10.3148/75.1.2014.15
8. O’Hearn M, Lara-Castor L, Cudhea F, et al; Global Dietary Database. Incident type 2 diabetes attributable to suboptimal diet in 184 countries. Nat Med. 2023;29(4):982-995. doi:10.1038/s41591-023-02278-8
9. OECD Data Explorer. Organisation for Economic Co-operation and Development. https://data-explorer.oecd.org/
10. Parasrampuria S, Murphy S. Trends in prescription drug spending, 2016-2021. Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services. September 2022. https://aspe.hhs.gov/sites/default/files/documents/88c547c976e915fc31fe2c6903ac0bc9/sdp-trends-prescription-drug-spending.pdf
11. Global pharmaceutical market size: the top global pharma companies & pharma markets ranked (2024). Pharmchoices. 2024. https://pharmchoices.com/the-top-global-pharma-companies-pharma-markets/ 
12. Research and Markets. United States diabetes market report: 2024-2032. July 2024. https://www.researchandmarkets.com/report/united-states-antidiabetic-market
13. Ripple WJ, Beschta RL. Trophic cascades in Yellowstone: the first 15 years after wolf reintroduction. Biological Conservation. 2012;145(1):205-213. doi:10.1016/j.biocon.2011.11.005
14. Srour B, Fezeu LK, Kesse-Guyot E, et al. Ultraprocessed food consumption and risk of type 2 diabetes among participants of the NutriNet-Santé prospective cohort. JAMA Intern Med. 2020;180(2):283-291. doi:10.1001/jamainternmed.2019.5942
15. Elflein J. Thehealth effects of ultra-processed food - statistics & facts. Statista. September 13, 2024. https://www.statista.com/topics/12722/the-health-effects-of-ultra-processed-food/
16. The Commonwealth Fund. 2024. https://www.commonwealthfund.org/
17. Health, nutrition and population. The World Bank. 2024. https://datatopics.worldbank.org/health/
18. Global Health Expenditure Database. WHO. 2024. https://apps.who.int/nha/database