6

Research on Healthy Foods: Marketing, Not Necessarily Science

REGARDLESS OF WHO ISSUES THEM, GUIDELINES FOR HEALTH promotion and disease prevention universally recommend diets that are largely plant-based, meaning that they include plenty of fruits, vegetables, grains, beans, and nuts. The US dietary guidelines also recommend foods in the “protein” category. Grains, beans, and nuts are good sources of protein, but the guidelines use “protein” to mean low-fat dairy, lean meats, and fish. Recommended eating patterns include all these foods, relatively unprocessed, but with minimal addition of salt and sugars. Such patterns provide nutrients and energy in proportions that meet physiological needs but also minimize the risk of obesity, type 2 diabetes, and other chronic diseases. One more definition: “Patterns” refers to diets as a whole, not to single foods. No one food makes a diet healthful. The healthiest diets include a wide variety of foods in each of the recommended categories in amounts that balance calories.

In their largely unprocessed forms, foods from the earth, trees, or animals are healthful by definition. So why, you might ask, would the producers of foods such as cranberries, pears, avocadoes, or walnuts fund research aimed at proving that these particular foods—rather than fruits, vegetables, or nuts in general—have special health benefits? Marketing, of course. Every food producer wants to expand sales. Health claims sell. The FDA requires research to support health claims and greatly prefers studies that involve human subjects rather than animals.

All of this explains why Royal Hawaiian Macadamia Nut, Inc. petitioned the FDA in 2015 to allow it to say in advertisements that daily consumption of macadamias—along with eating a healthy diet—may reduce the risk of heart disease. The eighty-one-page petition cited several studies done in humans, one of them funded by the Hershey Company, which sells chocolate-covered macadamias.¹ The FDA ruled that it would permit a qualified health claim for macadamia nuts with this precise wording: “Supportive but not conclusive research shows that eating 1.5 ounces per day of macadamia nuts, as part of a diet low in saturated fat and cholesterol and not resulting in increased intake of saturated fat or calories may reduce the risk of coronary heart disease.”² Can a statement this cumbersome help sell macadamia nuts? Definitely, with a little help from the press: “Go nuts, folks! FDA declares macadamia nuts heart healthy.”³

Legitimate scientific questions can be asked about specific foods—their nutrient content or digestibility, for example—but most such issues were addressed ages ago. Foods are not drugs. To ask whether one single food has special health benefits defies common sense. We do not eat just one food. We eat many different foods in combinations that differ from day to day; varying our food intake takes care of nutrient needs. But when marketing imperatives are at work, sellers want research to claim that their products are “superfoods,” a nutritionally meaningless term. “Superfoods” is an advertising concept.

But what is wrong with promoting the benefits of healthful foods? Wouldn’t we be better off eating more of them? Yes, we would, but many industry-funded studies are misleading, which is why the FDA requires so many qualifications in the claims it allows. This kind of research is designed to produce results implying that people who eat this one food will be healthier and can forget about everything else in their diets. Research aimed at marketing raises questions about biases in design and interpretation, may create reputational risks for investigators, and reflects poorly on the integrity of nutrition science. It also raises questions about the role of government agencies in promoting single-food research and about their failure to do a better job of regulating marketers’ claims about health benefits based on that research. To illustrate why such concerns matter, consider some of the marketing issues related to three healthy foods: blueberries, pecans, and pomegranates.

Maine Wild Blueberries: A “Nutrient-Rich SuperFruit”

The trade association Wild Blueberries of North America wants you to understand that frozen, fresh Wild Blueberries (always capitalized) are better for you than unfrozen, fresh supermarket highbush berries: “Jam-packed with a variety of natural phytochemicals such as anthocyanin, Wild Blueberries have twice the antioxidant capacity per serving of regular blueberries. A growing body of research is establishing Wild Blueberries as a potential ally to protect against diseases such as cancer, heart disease, diabetes and Alzheimer’s.”⁴ This is an impressive range of health benefits for a tiny fruit consumed in small amounts even by people who can find them, but selling them as an antioxidant powerhouse has done wonders for the Maine wild-blueberry industry.

For years, I have had a potted highbush blueberry plant on my twelfth-floor Manhattan terrace, satisfyingly productive in years when I can manage to fend off the hordes of voracious finches. Unlike the easy-to-get-at highbush varieties, the wild ones grow close to the ground on sandy soils left behind by receding glaciers and are more difficult to harvest. In Maine, these blueberries are an important agricultural commodity. Since 1945, Maine blueberry growers have supported research—then and now focused on production practices—at the state’s university. As techniques improved, blueberry growers produced more berries. These needed to be sold.

The Maine Wild Blueberry Commission consulted with marketing specialists. In 1992, a consultant advised focusing on taste as a means of differentiating wild from cultivated blueberries. But the consultant then read an article in a USDA magazine extolling the virtues of plant antioxidant pigments in “boosting the immune system, reducing inflammation and allergies, [and] detoxifying contaminants and pollutants.”⁵ The article said that USDA investigators had invented an assay for antioxidants demonstrating that blueberries have the highest levels of any fruit tested (kale was highest among vegetables). The consultant advised the commission to focus on antioxidants. From 1997 to 2000, half the Maine Wild Blueberry Commission’s marketing resources went into repositioning blueberries as a health icon. The strategy worked. Maine’s wild-blueberry industry flourished—at least for a while. Recent overproduction and competition from Canadian fruit have dropped prices below profitability.⁶

I love blueberries, wild and cultivated, but they are a fruit like any other. Their antioxidants may counteract the damaging actions of oxidizing agents (free radicals) in the body, but studies of how well antioxidants protect against disease yield results that are annoyingly inconsistent. When tested, antioxidant supplements have not been shown to reduce disease risk and sometimes have been shown to cause harm.⁷ The USDA no longer publishes data on food antioxidant levels “due to mounting evidence that the values indicating antioxidant capacity have no relevance to the effects of specific bioactive compounds, including polyphenols on human health.”⁸ The US National Center for Complementary and Integrative Health at NIH judges antioxidants as having no special benefits.⁹ People who eat more fruits and vegetables have less risk of chronic disease, but nobody really knows whether this is because of antioxidants, other food components, or other lifestyle choices.

Blueberries, like every other fruit and vegetable, have a unique combination of antioxidants. But so what? It is best not to expect miracles—like this especially wishful-thinking headline: “Blueberries associated with reduced risk of erectile dysfunction.” Two of the authors of this study reported receiving funding from the US Highbush Blueberry Council “for a separate project unrelated to this publication.”¹⁰

Pecans: A New Federal Marketing Order

In April 2016, I received an email from Jeff Worn, vice president of the South Georgia Pecan Company, asking if we could have a conversation about health claims for these nuts. I love pecans, particularly in pralines, but I am not a fan of health claims for foods; health claims are about marketing, not science. I told Worn that if he wanted to promote pecans as a superfood, I would be no help at all. He replied that his request was about education, not marketing. Most Americans, he said, think of pecans as something in pralines, ice cream, or pie, but, he said, pecans are “actually much more than that. They have a ton of nutritional value in the antioxidants they contain.… Plus the fats in Pecans are good fats, but the general public thinks more fat in a product means love handles and that’s just not true with Pecans.”¹¹

Antioxidants again. But I could see his point about the healthy fats in pecans, and we began an email correspondence. He explained that his industry was in the process of obtaining a federal marketing order (FMO) from the USDA. FMOs are the USDA’s way of helping producers of what it calls “specialty crops” (translation: fruits, vegetables, and nuts for human consumption, as opposed to industrial corn and soybeans fed to animals). The FMO for pecans applies to pecan growers in fifteen states; it authorizes the pecan industry to collect data, recommend quality standards, regulate packages and containers, and conduct research and promotional activities. It also establishes an American Pecan Council to work with the USDA to decide how the funds will be used and to oversee these activities.

TABLE 6.1. USDA Marketing Programs for Fruits, Vegetables, and Nuts (“Specialty Crops”), 2018

Research & promotion (checkoff) programs

Hass Avocado Board

Mushroom Council

National Mango Board

National Peanut Board

National Potato Promotion Board

National Processed Raspberry Council

National Watermelon Promotion Board

Popcorn Board

United States Potato Board

US Highbush Blueberry Council

Marketing orders

Almonds

Apricots

Avocados

Cherries, sweet

Cherries, tart

Citrus

Cranberries

Dates

Grapes

Hazelnuts

Kiwifruit

Olives

Onions

Pears

Pecans

Pistachios

Plums/Prunes

Potatoes

Raisins

Spearmint oil

Tomatoes

Walnuts

SOURCES: US Department of Agriculture, Agricultural Marketing Service, “Research and promotion,” www.ams.usda.gov/rules-regulations/research-promotion; US Department of Agriculture, Agricultural Marketing Service, “Marketing orders and agreements,” www.ams.usda.gov/rules-regulations/moa.

I did not know much about FMOs and asked Tufts University professor Parke Wilde, my go-to person for information about checkoff programs, to help me understand how checkoffs and FMOs differ. He explained that both are run by the USDA’s Agricultural Marketing Service and both collect mandatory fees from all producers for marketing, promotion, and research. Their most obvious difference is geographical; checkoffs are national, but FMOs are regional (most of the other differences have to do with their enabling legislation, scope of activities, and vulnerability to legal challenges). Table 6.1 lists the specialty crops covered by checkoffs and FMOs in 2017.

Worn said he was working with researchers at the University of Georgia to do clinical studies on the health benefits of pecans and wanted my thoughts on what his group should or should not say about the nutritional benefits of pecans for marketing purposes. I thought his comment “It’s about education” sounded like marketing. Pecan growers want to sell more pecans. I want people to understand that all whole foods have nutritional value—all relatively unprocessed fruits, vegetables, grains, beans, meats, dairy, and, yes, nuts. But I could understand his emailed response: “I do want people to buy more Pecans. It’s what pays the bills, but it’s more than that to me. I am a facts based person.… I just have a desire to dig deeper and bring something to the consumer that isn’t all about money.”

I go through all this, with his permission, because he is especially thoughtful about the role of research in food marketing. I told him that I do not need more information about the nutritional benefits of pecans. I already know that people who habitually eat nuts tend to be healthier than people who do not. This is true whether the nuts are pecans, macadamias, walnuts, almonds, or any other. But nut-eaters may differ in other ways from non-nut-eaters. What kind of research would it take to demonstrate that eating pecans—as compared to any other nut—produced measurable improvements in health? Such studies would be impossibly difficult and expensive. Do the nutritional differences between one nut and another merit funding a clinical trial to find out? I think not, but I can understand that nut producers might disagree.

Worn sent me the protocol for a study of the effects of a pecan-rich diet on biomarkers of risk for cardiovascular disease and diabetes in overweight people. The investigators presented the preliminary results of this study at the 2016 annual meeting of the American Society for Nutrition. Their poster presentation associated eating pecans with “reductions in fasting insulin, glucose, blood lipids, systolic blood pressure, and inflammation, and longer lag time for LDL [low-density lipoprotein] oxidation… although none of these changes were statistically significant.”¹² This conclusion put a positive spin on neutral (not statistically significant) results, as is typical of industry-funded research. The poster did not disclose the study’s funder, but the protocol did: the National Pecan Shellers Association.

POM Wonderful: A $35 Million Research Investment

Early in 2008, I received a polite letter from Matt Tupper, president of POM Wonderful, a company making pomegranate juice and supplements. The company is owned by the billionaire couple Lynda and Stewart Resnick, owners of the largest farm in the United States. Their $4.5 billion, privately held company also owns Fiji Water as well as 180,000 acres (twenty-one square miles) of almond, pistachio, and citrus orchards in California alone, among them 18,000 acres of pomegranates.¹³ Tupper wrote that he admired my work but was disappointed to see a quotation from me in an article he had just read: “Pomegranates are no better than any other fruit. They’re just brilliantly marketed.… Spend $20 million researching just about any fruit… and you’d discover that it, too, provides unique and miraculous-sounding benefits.”¹⁴ Tupper’s letter explained that “the medical research funded by Pom is not simply a tool to market more pomegranates.… Rather, the fundamental goal of Pom’s research program is to develop a scientifically sound understanding about how and why the pomegranate impacts human health.” He said the company believed that pomegranates are superior to other fruits and vegetables because they are “uniquely endowed with a myriad of potent antioxidant compounds and therefore sit at the top of the nutritional pile.” Ah yes, antioxidants.

Since at least 2001, POM Wonderful had been investing in studies designed to show that daily intake of eight ounces of pomegranate juice or a supplement of pomegranate polyphenol extract would produce higher levels of antioxidants in the body and would therefore reduce risks for cardiovascular disease, type 2 diabetes, prostate cancer, and erectile dysfunction (the holy grail, apparently). It claimed these benefits in its advertisements.¹⁵ In 2011, I wrote a column for the San Francisco Chronicle explaining why nutrition research, particularly research funded by industry, requires careful interpretation. I said that whenever I see studies claiming benefits for a single food, I want to know three things: whether the results are biologically plausible; whether the study controlled for other dietary, behavioral, or lifestyle factors that could have influenced its result; and who sponsored it. I used pomegranates as an example, noting that one of its major producers was sponsoring research to hype its benefits. I agreed that pomegranates might have high antioxidant activity but asked, “Compared with what? Its maker does not say.”¹⁶

FIGURE 6.1. This POM Wonderful advertisement appeared the day after a district court judge ruled that POM’s health claims were not scientifically substantiated. The judge’s statement is slightly out of context; the full statement read, “Pomegranate juice is a natural fruit product with health promoting characteristics. The safety of pomegranate juice is not in doubt.”

In response, a vice president of clinical development at POM wrote me another polite letter stating, “POM Wonderful prides itself in supporting extensive scientific and medical research, well in excess of any other food company. With over 60 publications in peer-reviewed scientific journals, our commitment to understanding the health benefits of the pomegranate speaks for itself.… Comparing the health benefits of our product to other juices is not a key objective of our extensive research program.” Instead, he said, POM intended the research “to unlock the secrets of a particularly healthy fruit.”¹⁷

POM held nothing back in advertising those secrets. Its ads said POM had clinical studies proving that the juice and supplements “reduce the risk of, and treat heart disease, including by decreasing arterial plaque, lowering blood pressure, and improving blood flow to the heart.” The company advertised that POM products reduce the risk of prostate cancer “by prolonging prostate-specific antigen doubling time.” Even better, POM juice “prevents, reduces the risk of, and treats, erectile dysfunction.” Drink POM juice, and you can even “cheat death” (see Figure 6.1).¹⁸

Under the FDA’s arcane rules for health claims, POM was advertising its juice and supplements as drugs, not food. Understanding the distinction requires a bit of history. Until the early 1990s, the FDA did not permit foods or supplements to be marketed with statements that they could prevent, mitigate, or treat a disease; only drugs do these things. If marketers wanted to make drug claims, they had to prove safety and efficacy. Food companies knew that clinical trials would be impossibly expensive and unlikely to prove much. Instead, they lobbied for the right to “inform the public” about health benefits.

When Congress passed the Nutrition Labeling and Education Act of 1990, it instructed the FDA to allow food manufacturers to make health claims that were supported by “substantial scientific agreement” among experts. The FDA’s regulations responding to this act permitted food labels to claim, for example, that fruits and vegetables—as a group—reduce the risk of cancer and heart disease. It also allowed a few other disease claims.¹⁹ Subsequent claims that a food prevents or treats disease needed FDA approval, but POM had not made such requests.

I call the FDA’s rules arcane because POM’s marketers could simply have said that pomegranate juice “supports” healthy heart, prostate, or sexual function. This would have been a structure/function claim authorized by the dietary supplement act, which the FDA also permits for foods. POM’s lawyers surely must have known the legal distinction between “prevents disease” and “supports health,” but the company insisted that its research merited claims that POM products can prevent specific diseases and symptoms.

The FDA disagreed. Early in 2010, the agency warned POM that its advertisements were positioning the juice and supplements as drugs and were therefore illegal under current provisions of the laws that govern the FDA’s activities.²⁰ In September, the Federal Trade Commission (FTC), which regulates advertising and usually follows FDA guidance in such matters, told POM that it had to stop making unproven claims in its advertisements. The FTC did not view POM-sponsored research as adequate evidence for the claims.

POM’s response? It sued. The company argued that the FTC’s actions “detrimentally impacted POM’s freedom of speech now, the value of its research program now, and are violating its First and Fifth Amendment rights.”²¹ Really? I find it hard to believe that America’s Founding Fathers introduced the First and Fifth Amendments to protect the rights of POM Wonderful to make unsubstantiated health claims. The court, apparently, had the same concern. It upheld the FTC’s complaint that POM’s heart disease and cancer claims were false and unsubstantiated. It agreed with the FTC that POM’s research was insufficiently compelling, particularly because many of POM’s studies were uncontrolled or unblinded (study subjects knew what they were drinking or taking). Furthermore, “the erectile dysfunction claims were false and unsubstantiated because the study on which the company relied did not show that POM Juice was any more effective than a placebo.”²²

POM-sponsored research provides additional examples of how easy it is to design studies to give desired results. According to information in the court decision, POM had invested more than $35 million in nearly one hundred studies at forty-four different institutions. At least seventy of its studies were published in peer-reviewed journals. Among other benefits, these demonstrated—using appropriate scientific methods—that pomegranate juice has antioxidant activity and acts as an antioxidant in the body. No surprise there. All fruits and vegetables have antioxidants (they help fend off invaders). POM’s research did not compare the effects of the antioxidants in pomegranates to those in any other fruit. Whether a particular fruit has more antioxidants depends on which particular antioxidants you look for, but what difference does it make? Plant foods as a whole promote health. The particular role of antioxidants in health continues to be difficult to sort out.

POM, however, deserves credit for chutzpah. The day after the judge’s decision, the company placed a full-page ad in the New York Times declaring, “FTC v. POM: You be the judge.” The ad included selected quotations from the judge’s decision. Here is one: “Competent and reliable scientific evidence shows that pomegranate juice provides a benefit to promoting erectile health and erectile function (here).” This statement is indeed here, but the ad omitted the sentence that follows: “There is insufficient competent and reliable scientific evidence to show that pomegranate juice prevents or reduces the risk of erectile dysfunction or has been clinically proven to do so.”²³ The company also put out-of-context quotations from the decision in other advertisements, such as the one shown in Figure 6.1.

POM Wonderful is privately held and does not have to disclose sales figures, but they are estimated at $100 million a year for the juice and supplements. Lynda Resnick once explained, “People needed pomegranate juice in their lives (even if they didn’t know it yet), and I knew they would pay what it was worth.”²⁴ POM’s price premium is impressive. In 2018, a twenty-four-ounce bottle of POM cost $8.99 at my local Manhattan grocery store, but thirty-two ounces of Tropicana orange juice sold for $3.99. Time carried a full-page ad that read, “Refresh your memory: learn about this preliminary research on pomegranate polyphenol antioxidants and memory and cognition.” These, the ad admits, “are early scientific findings on cognitive health and the impact of pomegranate juice on the human brain has not yet been adequately studied. Clinical research is needed to help establish causation.”²⁵ Is such research worth doing? For marketing, apparently so. For science? It depends on the research question.

I end this saga by quoting the conclusion of the one POM-funded study I ran across in my year-long collection: “While pomegranate extract supplementation may reduce blood pressure and increase the antioxidant activity in hemodialysis patients, it does not improve other markers of cardiovascular risk, physical function, or muscle strength.”²⁶ The no-benefit results suggest that this study must have been designed with appropriate controls, which it was. But “may reduce blood pressure” gives the results a positive spin; the investigators observed no reduction when they corrected the results for baseline blood pressure.

Using Science to Market Healthy Foods

Health claims based on research, regardless of how well conducted, help sell foods and food products. When the FDA attempted to block the more far-fetched claims, food companies sued the agency on First Amendment grounds. In 2003, after losing several of these cases one after another, the FDA gave up fighting them and eased its standards.²⁷ Thereafter, it would accept a much broader range of evidence for health claims, would no longer insist on “substantial scientific agreement,” and would only challenge claims—like those of POM Wonderful—that most flagrantly violated the law. This decision encouraged the marketers of whole foods like those shown in Table 6.2 to sponsor research to support health claims for their products.

The studies summarized in the table differ in type; they include reviews, animal experiments, and human trials. I simplified their conclusions to suggest how marketers might use their results in health claims. In fairness, some of these studies qualify their conclusions more carefully. For example, the Hass avocado study says, “Avocados could be an effective dietary strategy for cognitive health in the aging population” (my emphasis). But even without convincing evidence, the FDA accepts health claims using the word “support,” as authorized for dietary supplements. On the basis of this study, avocados can be—and are—marketed for their ability to support cognitive heath in the elderly. Wouldn’t it be terrific if protecting cognitive health were that easy and delicious?

TABLE 6.2. Industry-Funded Studies of Food Plants with Results Useful for Health Claims (Selected Examples), 2015–2018

NOTE: Source notes are located in the “Notes to Tables” here.

One question is why investigators would do this kind of research. John Sievenpiper, an investigator who sometimes partners with food companies, explained to a reporter, “It’s very hard to fund randomized trials properly.… You have to engage the food industry to get those trials done.… [We] see it as our role to try to influence [companies] and produce healthier foods and promote healthier foods.”²⁸

A more critical question is what to make of all this. If I may overgeneralize, the quality of single-food marketing studies does not always hold up to scrutiny. For example, a nutritional biochemist criticized the raisin study noted in the table for its misuse of statistics and for comparing raisins to processed snack foods: “With the design used you can’t really say that raisins were ‘good’ for the participants, just not as bad as the junk snacks.”²⁹ Even when done well, studies so clearly aimed at marketing skew the research agenda. If food companies were not funding marketing studies, investigators might be working on more important biological problems. All these foods are highly nutritious and well worth eating for their taste and texture—as well as for their health benefits. Is one fruit, vegetable, or nut better for you than another? The answer, as I keep saying, depends on everything else you eat or do. People who habitually eat largely plant-based diets are healthier. Variety in food intake and calorie balance are fundamental principles of healthful diets.

Again to be fair, not all studies funded by plant trade associations come out the way they are supposed to. The California Strawberry Commission, for example, sponsored a study to see whether eating forty grams of dried strawberry powder a day—equivalent to a pound of strawberries—would counteract the effects on blood lipids of eating a high-fat diet. It did not.³⁰ I do not want to even think about strawberry powder. But does this result mean we should not be eating strawberries? Of course not. All fruits, vegetables, and nuts have vitamins, minerals, fiber, antioxidants, and other components that collectively promote health. If we are fortunate enough to have choices, we can eat the ones we like.