What are the First Amendment obstacles to mandatory front-of-pack labeling?

Suppose the government wanted to require front-of-pack nutrition labels for packaged food and beverages, making it easier for consumers to see at a glance some key nutritional qualities.

In the United States, could the manufacturers claim that such a rule violates their First Amendment rights?

In the journal Food Policy this past summer, Jennifer Pomeranz, Dariush Mozaffarian, Renata Micha, and I study the precedents. Much depends on whether a particular labeling policy could satisfy a legal principle called the Zauderer test. This test stipulates, for example, that the mandatory labeling information must be factual and uncontroversial. 

Looking at the wide array of front-of-pack labeling schemes around the world, we find that some varieties are more likely than others to pass this test. Certain proposals for simple mandatory symbols indicating "healthy" or "unhealthy" broadly, without communicating much nutrition information, could be ruled unconstitutional.

The article is titled, "Mandating front-of-package food labels in the U.S. – What are the First Amendment obstacles?"

Pomeranz et al. (2019). Table 1 (small excerpt).

How old is the term "coconut milk"?

In the New Republic this week, Emily Atkin reviews the renewed Trump administration interest in restricting the word "milk" on labels for products such as "soy milk" and "coconut milk."

“As the [FDA] Commissioner noted, the dictionary definition of the word ‘milk’ does include coming from nuts, and this is not a new concept,” the Plant Based Food Association said in an emailed statement. Indeed, Gottlieb on Tuesday acknowledged that “if you open up a dictionary, it talks about milk coming from a lactating animal or a nut.” This is one of several reasons why non-dairy milk companies reject the idea that they’re misleading consumers.

The argument made me wonder how old is the use of "milk" for products other than cow's milk? Here are a couple entries from Merriam-Webster (which seemed to require sign-in after the first few lookups):

• Almond milk (14th century).
• Coconut milk (1698).

I also looked up 100 Bible verses with the word "milk" in English translation (Hebrew and Greek may be another matter). For the dairy industry, the good news is that most verses did refer to excretions from a lactating mammal. Isaiah provided the most metaphorical use I could find: "You shall suck the milk of nations; you shall nurse at the breast of kings." And the dairy industry may hope that Isaiah was just being aspirational in some of his comments: "He who has no money, come, buy and eat! Come, buy wine and milk without money and without price."

Others have recently pointed out the many names of food products that could get caught up in an overly literal FDA rulebook, if it were applied consistently.

• Hamburgers (contain no ham ... and aren't from Hamburg either).
• Hot dogs (contain no dog).

The comment period for the FDA proposal will soon open, and I suspect there will be plenty of submissions on this topic.

Gershoff Symposium, April 27

The 2018 Gershoff Symposium, on April 27, will feature 4 presentations:

• "Food Labeling Chaos" (Keynote). Kathleen Merrigan, Executive Director of Sustainability and Professor of Public Policy at George Washington University. Merrigan was Deputy Secretary of Agriculture in the Obama administration. She led a fascinating workshop on sustainability in dietary guidance in 2014, leading to a 2015 commentary in Science (whose co-authors included my Tufts colleague Tim Griffin and myself), which was part of the discussion and debate around sustainability issues in the 2015 Dietary Guidelines for Americans. 
• "Deal or No Deal, What Do You See in a Label?" John Bernard, Professor, Department of Applied Economics & Statistics, University of Delaware. John is an old friend from graduate school at Cornell. His recent research addresses fascinating topics in food labeling, including (a) one article in Food Policy anticipating what will type of mess will happen if cloned meat and dairy products enter the marketplace without labeling rules, and (b) another article in Food Quality & Preference with amusing results about how much more highly consumers rate the taste of equivalent food products when they are labeled organic compared to when they are not labeled.
• "State of the Evidence: Organic vs. Non-Organic." Qi Sun, Associate Professor of Medicine, Brigham and Women's Hospital and Harvard Medical School.
• "Creating a Sustainable Food System: What Matters and What Counts?" Amanda Beal, President and Chief Executive Officer, Maine Farmland Trust. Beal is a Tufts Friedman School alum and a great source of insight about local farming and fisheries in Maine.

A registration link and more information about this annual symposium, in honor of former dean Stanley Gershoff, are here.

Peeling back the wrapper: Why would Mars support the new added sugars label?

By Melissa Hudec and Parke Wilde

The Food and Drug Administration (FDA) is proposing a new Nutrition Facts Panel that identifies added sugars (rather than just mixing added sugars in with all other sugars).

The proposed label supports a recommendation from the 2015 Dietary Guidelines Advisory Committee, which—if adopted by the federal government—would advise that added sugars be limited to no more than 10% of daily calories.

Of course, many food and beverage companies hated these policy changes. Yet, there were some exceptions. For example, Mars, Inc. — yes, the maker of M&Ms and other candies—supported the new added sugars label. As the Wall Street Journal reported:

“It might appear to be counterintuitive, but if you dig down a bit more, we know candy itself is not a diet,” said Dave Crean, global head of research and development at Mars. “It shouldn’t be consumed too often, and having transparency of how much it should be consumed is actually quite helpful to consumers.”

We are not sure why Mars would support the new labels, but we have three theories, which may in combination offer some explanation.

First, Mars is a privately owned company, not beholden to shareholders. Perhaps this gives managers greater freedom to pursue the company’s long-term interest and reputation rather than just short-term sales.

Second, chocolate products may have less to lose than other high-sugar products have. We created this mockup approximate comparison of a Mars candy and a major soda brand (assuming a standardized serving size for the candy and making some assumptions about how much sugar is added sugar). A chocolate candy has some calories from fat, and even some of the carbohydrates (in the dairy ingredients) may not count as “added sugars.” By comparison, a soda gets all of its calories from added sugars. In total, the Mars candy provides only approximately 13g of added sugars (26% of the daily value), while a soda provides 39g of added sugars (fully 78% of the daily value!).

Third, Mars is more than just chocolate. In fact, it may surprise the reader that confectionery products comprise only about 1/3 of its portfolio (.pdf). The rest of sales come from primarily from pet food, as well as from the Wrigley subsidiary, and a smaller amount from savory food products, drinks, and the up-and-coming Symbioscience (“cocoa flavanols” and a topic for future discussion) products. 

Source: Mars, Inc.

Surprisingly, the food industry is not unanimous about the new added sugar label. On the contrary, for a variety of reasons, some major players say they can work just fine under the new system.

This post was adapted from Melissa Hudec's term paper for Parke Wilde's class on U.S. Food Policy.

Update (Jan 5, 2016): A former student in the same food policy class -- who now works in corporate affairs at Nestlé -- emails to point out quite rightly that her company also supported the FDA proposal: "I loved your and Melissa Hudec’s post on added sugars labeling. Now that I work at Nestlé and spend a lot of time on our nutrition-related positions, I feel compelled to comment! Nestlé has also publicly supported FDA’s proposal for mandatory declaration of added sugars. In fact, this article came out a few days after your post."

KIND Snacks petitions FDA to change the definition of "healthy"

In early 2015, FDA wrote a warning letter to KIND Snacks, calling out the company for using the term "healthy" on nut-filled snacks that -- among other things -- had a level of total fat level exceeding the official standards for "healthy" foods.

At the time, media sources included quotes from leading national figures in nutrition policy, who pointed out that recent editions of dietary guidelines do not come down hard on total fats (instead focusing on the types of fats). For example, NPR quoted Walter Willett.

Willett says that the FDA's letter to Kind is based on outdated guidelines, at least when it comes to nuts. The government updates its Dietary Guidelines for Americans every five years, and the latest report from the advisory committee for those guidelines does indeed point to research supporting the inclusion of nuts in a healthful diet.

But the FDA seems to be lagging, in part because the agency doesn't revise its guidelines as frequently. "I think there's wide consensus that nuts are a healthy food," Willett says.

Last week KIND Snacks petitioned FDA, asking for a change in the definition of "healthy." Here is part of the firm's publicity in support of that petition, making the rhetorical point that current rules focus too much on fat content and not enough on processing or sugar content. 

What do you think?

A couple complications:

1. This post was spurred by an email from a food policy student, who also has worked for KIND Snacks. I think I would have chosen this topic anyway, but introspection cannot quite confirm that. 
2. Some KIND Snacks are sweetened, and, in any case they are packaged manufactured foods, not quite so simple as the nuts in the image above.

Food policy is never simple.

Vanity beverage name: "Just Sugar Water"

Let's have a little fun with a major beverage manufacturer's website that allows you to order soda bottles with vanity names printed on them.

Update (11:42am): Twitter friends point out that ... of course ... others have already been experimenting with the limits of the words the website will accept. And here is CSPI's contribution.

Is it really so bad that WTO ruled against U.S. in country-of-origin labeling dispute?

The World Trade Organization (WTO) today ruled in favor of Canada and Mexico, saying that U.S. country-of-origin labeling (COOL) rules violate our commitments in previous trade agreements.

The COOL rules in dispute required new labels on fresh beef, pork, and lamb, but not on processed foods such as hot dogs. The labels would say what country the product comes from. In some cases, the labels might have to be complex ("this cow was born in Canada, fed in the United States ...").

Many people who care about food policy from a public interest perspective will say the WTO ruling is terrible. For example, Food & Water Watch Executive Director Wenonah Hauter says today:

The WTO’s continued assault against commonsense food labels is just another example of how corporate-controlled trade policy undermines the basic protections that U.S. consumers deserve.

But let me ask, is the WTO ruling really so bad? I have two reasons for asking this question.

First, perhaps the WTO ruling has some merit.

Canada and Mexico claimed that the U.S. law was designed as a trade barrier, not merely as a consumer labeling provision. They pointed out that the tracking and record-keeping burdens fell more heavily on Canadian and Mexican producers than they did on seemingly similar U.S. producers. They also cast doubt on the consumer information merit of all this tracking and record-keeping, because so much of the meat was destined for processed food that never would carry a country-of-origin label anyway. If the U.S. asks Canada and Mexico to incur tracking and record-keeping costs, and then fails to share the resulting information with processed meat consumers anyway, it does look like maybe the whole point was just to create a burden for our trading partners.

It would be one thing if the United States simply never negotiated a trade agreement in the first place. But, it is another thing altogether if the United States does negotiate a trade agreement, saying we will reduce trade barriers in return for Canada and Mexico doing the same, but then we fail to do what we promised.

Second, from the perspective of Food & Water Watch and other trade-skeptical consumer groups, perhaps the consequences are not so terrible.

WTO critics claim such rulings violate U.S. sovereignty. For example, Lori Wallach, director of Public Citizen's Global Trade Watch says today:

Many Americans will be shocked that the WTO can order our government to deny U.S. consumers the basic information about where their food comes from and that if the information policy is not gutted, we could face millions in sanctions every year.

This is not true. The WTO cannot order our government to deny U.S. consumers such information. The WTO cannot rewrite U.S. laws. The U.S. can simply refuse to comply.

When the WTO rules against the United States, the only remedy the WTO has is the power to permit Canada and Mexico to put up trade barriers of their own, without these trade barriers being ruled non-compliant with the same trade agreements. In other words, the only power the WTO has is the power to state for the record, "fair is fair."

For example, if the U.S. chooses not to honor the WTO ruling, Canada has proposed a list of U.S. exports that may get new import tariffs at the Canadian border: corn, meat, apples, pasta, orange juice, and so forth.

The United States may honor its trade agreements, or we may fail to honor them. If we don't honor them, surely it is right to be good sports when Canada and Mexico establish new import tariffs.

On what grounds should Food & Water Watch or Public Citizen's Global Trade Watch complain about the WTO ruling? If it is fine for the United States to exercise our sovereignty, it is fine for others also. Far from griping about the WTO ruling, trade-skeptical public interest organizations could just tip their hat to the WTO, contemplate the consequences for reduced food trade, and declare, "all is well."

There is a good reason why they don't do so. Many Americans, especially in agricultural regions of the country, are glad for the business that trade brings in the form of increased export markets for our farmers. When we remember that Americans are food producers as well as consumers, the trade agreements begin to seem more sensible and the WTO begins to seem more reasonable.

Further information

Chapter 4 of Food Policy in the United States: An Introduction addresses trade issues. I do my best to make the case for a public-interest pro-trade perspective in a talk at Cornell University last fall (video here).

The food policy of Hampton Creek's "Just Mayo"

"Just Mayo" from Hampton Creek is a vegetable oil spread offering a new vegan egg-free alternative to traditional mayonnaise. For people who follow food policy, it generates a pile of interesting things to think about, including: (1) food industry innovation, (2) nutrition content, (3) food labeling for people who care about animal welfare and the environment, and (4) standards of identity. 

I heard about Just Mayo from a recent Friedman School alum, who works in marketing for the company and who now has hired some current students on her marketing team. Josh Balk, from the Humane Society of the United States, who spoke at the Friedman School last year, is one of the founders.

(1) In part because of interest from Bill Gates in alternatives to eggs, Just Mayo has been heavily covered in the business and technology press. According to Forbes in December, Gates backs Hampton Creek as one of a handful of companies that may transform the food system. The New York Times last year described the investment of venture capitalists in the start-up. Bloomberg BusinessWeek in October focused on Just Mayo's recent success in getting shelf space in Whole Foods.

(2) Part of the consumer motivation for choosing an egg-free product is likely to be nutrition goals. Like traditional mayonnaise, Just Mayo is a high-fat high-calorie vegetable spread. Its role in a healthy diet likely depends heavily on how the consumer chooses quantities and also on what foods the product complements (in reasonable quantities, mayo in cole slaw complements tasty vegetable consumption and serves as a gateway salad, while mayo on fried potatoes Belgian-style is a high-energy occasional treat).  I am not a dietitian, but I would think of Just Mayo's nutrition profile as neither better nor worse than mayonnaise.

INGREDIENTS: Non-GMO Expeller-Pressed Canola Oil, Filtered Water, Lemon Juice, White Vinegar, 2% or less of the following: Organic Sugar, Salt, Apple Cider Vinegar, Pea Protein, Spices, Garlic, Modified Food Starch, Beta-Carotene.

(3) Many other consumers may seek an alternative to mayonnaise for animal welfare or environmental reasons. Both Just Mayo and traditional grocery store mayonnaise are processed food products, with similar transportation and packaging issues. For animal-friendly consumers at a loss to navigate confusing cage-free and free-range egg labels, Just Mayo certainly makes things easier by simply avoiding the use of eggs as an input. As a rule, plant food sources exhaust less of the planet's land and energy resources per unit of food energy produced. For these consumers, Just Mayo would be an improvement over traditional mayonnaise.

(4) As a product name, "Mayo" was close enough to "mayonnaise" to prompt me to look up the standard of identity for mayonnaise [Edit Feb 13: this sentence toned down from an earlier version saying Just Mayo "flirts with violating" the standard of identity]. A standard of identity is the federal government's official definition of a food. Sometimes, having a standard of identity protects consumers, by forbidding the sale of foods that are adulterated with fillers. At other times, a standard of identity may become outdated, preventing healthful or environmentally sound innovations. For example, there have been major controversies over whether the word "milk" can be used in the label for "soy milk."

The presence of eggs in mayonnaise is almost an archetypal example of a standard of identity. The Congressional Research Service glossary of agriculture policy terms (.pdf) specifically mentions mayonnaise under the heading for "standards of identity." According to the standards of identify in the Code of Federal Regulations, which are accessible on the Food and Drug Administration (FDA) website, "mayonnaise" includes eggs. 

The label for Just Mayo uses the slang "mayo" rather than "mayonnaise" in the product name. A favorable review on a vegan website calls the product "vegan mayonnaise," while a Whole Foods press release goes instead with "vegan mayonnaise-style product." In a similar fashion, Kraft Mayo or low-fat Hellmann's may use the term "mayo" or "mayonnaise dressing" without observing complete agreement with the standard of identity for mayonnaise [Edit Feb 13: sentence added]. The American Egg Board, a semi-public federal government checkoff entity, has begun a campaign to address the labeling of egg substitutes. It has a related white paper appealing for a "clean label" to oppose some of the egg labeling alternatives that may implicitly or explicitly be critical of traditional industrial egg production.

There may be some legal fireworks before the labeling of vegan alternatives to mayonnaise is settled.

OMG! Sodium!

I showed this cereal box to my 10-year-old daughter at breakfast this morning, but she couldn't see the pun.  At first, I thought maybe she didn't know the text message shorthand, OMG.  On the contrary, she could only read the message, "Oh my God, sodium!"

Albany Law School professor Timothy Lytton has a new book, Kosher: Private Regulation in the Age of Industrial Food.  A key point is that this topic is more broadly relevant than one might think, because kosher food is just one of many examples of food regulation systems that can be adopted by the private sector.

Lytton was interviewed on the What is Your Food Worth? blog.

As a general matter, private food safety audits and industry-sponsored nutrition labeling schemes have been a great disappointment. Behind most major food-poisoning outbreaks is some private auditing firm that gave the food producer a phony five-star rating. And when nutritional rating schemes give high marks to sugary cereals and full-fat ice cream, you have to wonder.

As a kosher-observant Orthodox Jew, I realized that kosher certification offers a 2000 year old example of private food certification. My initial suspicion was that kosher certification was full of price gouging and unnecessary, super-stringent standards. As I began to get into my research, however, I found that, although fraud and corruption were rampant a century ago in kosher meat production, today’s kosher system is highly reliable. My book tells the story of how, within the span of a century, kosher certification became the one of the most reliable systems of private certification in the food industry, indeed, perhaps in any industry.

POM Wonderful claims are false and misleading

An administrative judge for the Federal Trade Commission (FTC) concluded (.pdf) last week that POM Wonderful marketing claims about health benefits were "false and misleading."

For example, POM Wonderful advertisements imply that the juice protects against prostate cancer.  Could this be true?  POM Wonderful cited a study with some evidence that "PSA doubling time" -- a measure of prostate cancer's progress -- is slowed by drinking POM Wonderful.  Yet, truthfulness requires more than selective quotation from a favorable study.  In the FTC hearing, the balance of scientific evidence failed to support POM Wonderful's implied prostate cancer claim. 

POM Wonderful argued that some of its claims were merely puffery, not intended actually to convince grown-up consumers that the juice protects against cancer.  Yet, truthfulness does not permit the kindergarten defense: "Okay, I implied it, but I didn't really say it, so it's not a lie."

The administrative judge is correct to tar the claims as false and misleading.

What is the policy implication?  Some reasonable people would say the FTC should crack down on misleading health claims.  Other reasonable people would say "buyer beware," while maintaining that regulation will do little good.  In either case, let us all acknowledge that the claims are false and misleading.  There can be no defense of the claims themselves.

Or, so I thought.

POM Wonderful's response to the ruling this week has a breathtaking audacity.  I see today on the NYT website, POM Wonderful advertisements boasting of the FTC judge's ruling.  For the prostate issue above, here is the key quote in the POM Wonderful ad today:

“Competent and reliable scientific evidence supports the conclusion that the consumption of pomegranate juice and pomegranate extract supports prostate health, including by prolonging PSA doubling time in men with rising PSA after primary treatment for prostate cancer.”

– Judge Chappell, Chief Administrative Law Judge, FTC

In the Matter of POM Wonderful LLC, Initial Decision (5/17/2012), page 282

How is this possible?  Did the judge really endorse the very cancer-protective claim that POM Wonderful had implied?  Here is the full passage from page 282 of the judgment, with the sentences not quoted by POM Wonderful underlined.

Competent and reliable scientific evidence supports the conclusion that the consumption of pomegranate juice and pomegranate extract supports prostate health, including by prolonging PSA doubling time in men with rising PSA after primary treatment for prostate cancer.  However, the greater weight of the persuasive expert testimony shows that the evidence relied upon by Respondents is not adequate to substantiate claims that the POM Products treat, prevent, or reduce the risk of prostate cancer or that they are clinically proven to do so.  Indeed, the authors of the Pantuck Study and the Carducci Study each testified that their study did not conclude that POM Juice treats, prevents, or reduces the risk of prostate cancer.

Let anybody who was tempted to criticize the FTC or defend POM Wonderful read these two passages and evaluate for themselves the company's standard of honesty.

In my view, POM Wonderful is truly a bold titan of the dubious claims industry.

Update (1:45 pm): I just noticed that Marion Nestle also covered the NYT ads.  Soon perhaps POM Wonderful will quote Marion's sentence: "Fruit juices are healthy and especially delicious when fresh."  Of course, Marion goes on to say she doubts the cancer claims too.

GAO: FDA lacks the power it needs to address structure / function claims

The nonpartisan Government Accountability Office this week recommended that the Food and Drug Administration take steps to clear up consumer confusion about health claims that food companies make on their products.

Currently, companies are allowed to make several types of claims:

• health claims -- aka REAL health claims -- which must have evidence that meets a standard called "significant scientific agreement"
• qualified health claims -- aka WISHFUL health claims -- which can meet a weaker standard of evidence, so long as the company admits the weaker evidence on the label
• structure / function claims -- aka STEALTHY health claims -- which purport not to be health claims (for example, a high-fiber product that "keeps your bowels moving" without explicitly claiming to "treat constipation"; or a high-calcium product that "builds strong bones" without explicitly claiming to "prevent osteoporosis")

Food manufacturers only use a few major officially recognized health claims, because, truthfully, there is not a lot of good evidence for many of the claims that companies might want to make.

FDA was forced several years ago to allow qualified health claims, because companies had sued in court over their First Amendment rights to make claims that had some scientific evidence, even if there was not enough to count as Significant Scientific Agreement.  However, the necessary cautionary statements about weak evidence are a turnoff for the consumer, so manufacturers have not made major use of these qualified health claims.

As a consequence, manufacturers have placed most of their bets on structure / function claims, which do not have to be backed by much evidence, because these claims pretend not to be health claims in the first place.  The only time FDA can take action against a structure / function claim is if it is blatantly false or misleading.

The most striking thing about the new GAO report is that it recommends FDA strengthen its ability to find out if such claims are false or misleading.  GAO points out that the Federal Trade Commission, which regulates advertising, has the authority to ask companies to present evidence for their claims.  FTC officials told GAO that they would be unable to enforce the rules without this authority to ask for evidence.  However, FDA lacks the authority to make companies share their evidence.  GAO recommended that FDA identify and request from Congress the authority it would need to effectively judge whether structure / function claims are false or misleading.

I am curious what Alex Tabarrok would say about this GAO recommendation.  Alex is a widely read economist blogger at Marginal Revolution, who tends to be ferociously (and thoughtfully and provocatively) critical of FDA on other regulatory issues.

IOM report on front-of-pack food labels

In recent years, many consumers have been confused by the wild and ill-coordinated array of front-of-package (FOP) food labeling efforts: Smart Choices, Smart Spot, NuVal scores, Guiding Stars, Heart Check, and so forth.

The Institute of Medicine (IOM) in October released its Phase I report on front-of-pack labeling options.  This report is available for free download on the website of the National Academies Press (requires brief registration of email address).

The report is politely worded, but it essentially seeks to rein in some of the excesses.  In place of complex multi-nutrient schemes, the report recommends emphasizing just a few key nutrition components for which the research base is most solid and the connection to major chronic diseases is strongest: calories, saturated fat, trans fat, and sodium.  I was a little surprised that added sugars failed to make the list, but the report authors had some practical concerns about how sugars are measured, and they felt that listing calories addressed much of the concern about sugars.

The report seemed unfavorably disposed toward algorithm-based systems (discussed previously on U.S. Food  Policy), especially if the algorithm is proprietary or complex and not all the details are shared.

Perhaps the most damning section of the report is an illustrative comparison of how the various systems rated the same set of products.  For example, the IOM report compared six grain products: regular oatmeal, instant oatmeal, unsweetened toasted oat cereal, sweetened toasted oat cereal, crisped rice cereal, and an apple cinnamon cereal bar.  These foods reflect the options that a grocery shopper really might face on a supermarket shelf, choosing the family's breakfast supplies for the next week.  All six products would win the Smart Choices and Heart Check standards, which seem fairly permissive.  NuVal would give a higher score to regular oatmeal and a lower score to instant oatmeal, while the Nutrient Rich Foods Index and Guiding Stars would do just the opposite.  Sensible Solutions would favor only the two oatmeal products plus unsweetened toasted oat cereal (not the sweet cereal and cinnamon cereal bar).  Taken as a whole, the comparison makes the current status quo in the grocery store aisle look like a big confusing mess.

A future report from the same committee will investigate what consumers actually understand.

A fine line: nutrient content claims and health claims

Kellogg can no longer use this marketing strategy (see Time Magazine in June). 

In federal government lingo, the image above is a "health claim" or "function claim" with insufficient evidence.

But the company will still use this strategy (see Marion Nestle and Food Navigator this week). 

In federal government lingo, this is a "nutrient content claim." For official purposes, everybody agrees to pretend that the word "antioxidants!" has no more health implications or evidence requirements than does the word "crunchy!" The only legal issue is whether the manufacturer actually has added in the claimed antioxidants.

If you are convinced that this distinction between health claims and content claims succeeds in protecting consumers from misleading marketing, you will be reassured that somebody out there is watching that line like a tennis referee.  Everybody else may want to continue to be skeptical of health-related packaged food marketing claims across the board.

Principles for front of pack scoring systems

Several front-of-pack or supermarket-shelf nutrition labeling programs use scoring systems to rate food products.

Examples include the Hannaford supermarket chain's Guiding Stars system ...
... and the NuVal system used by some other supermarket chains.
The computation of the scores seems fairly arbitrary to me. It might help to start with some basic mathematical principles. Here are three principles that I think a food scoring system should satisfy.

1) The score should not depend on the serving size. A major limitation of all such scoring systems is that our health really depends on how much of each food we eat, and in what combination. Unfortunately, the scores necessarily apply to each food as it sits on the supermarket shelf, not as we consume it. Because the creators of the score have no knowledge of how much we eat, the score should be treated as a description of the density of good and bad nutrients per unit of weight or food energy. The score should not depend on the serving size convention used or the number of servings in a package. I have seen some scoring systems that appear at first glance to be independent of the serving size, but on closer inspection have quirky limits on total daily nutrients per serving that contribute to the score.

2) The score should rate mixed foods in a consistent way. Take the example of a ham and cheese sandwich. Suppose the sandwich is, by weight or by calories, 40% bread, 30% meat, and 30% cheese. The score for a packaged ham sandwich in the supermarket should be the same as a weighted average of the scores for bread, meat, and cheese purchased separately:

SandwichScore = 40%*BreadScore+30%*HamScore+30%*CheeseScore.

Without this principle, the scoring system will be biased in favor of or against manufactured mixed foods in place of separate ingredients. I think a good scoring system would have no such prejudice.

3) The score should rate each good and bad nutrient independently. Suppose adding 20% more salt to a high-fiber food reduces the score by 20%. How much should adding 20% more salt change the score for a low-fiber food? Many of us would say the score should again change by 20%. Unfortunately, existing scoring systems may have strange interactions across the good and bad nutrients, so that the effect of one nutrient on the score depends on the value of other nutrients in a way that the authors probably did not intend.

Currently, the scoring systems seem to me to have an ad hoc quality that makes it difficult to take the quantitative scores very seriously. Broadly speaking, a scoring system may seem to work correctly in giving healthy foods a good score, and unhealthy foods a bad score, but consumers understood those broad outlines of their food options already even without a scoring system. The whole purpose of a scoring system is to add quantitative rigor to the information provided. They may have a long way to go.

In other recent reading on front-of-pack labeling, see the recent series by Timothy Lytton at the Fooducate blog. In general, Lytton suggests that FDA should not get into the business of developing its own front-of-pack system, but instead should just enforce existing rules against making implicit healthy claims for foods that fail to meet FDA's definition of "healthy." One exception to Lytton's hands-off recommendation is that he feels there may be a need for stronger regulation of the complex scoring systems.

Smart Choices suspends operations

In an effort to better coordinate with the Food and Drug Administration, the Smart Choices front-of-pack labeling program today announced that it would suspend operations for the present time.

The labeling program had been criticized for giving a stamp of approval to marginal products, such as a somewhat reformulated version of Froot Loops. This blog had previously covered both this criticism and the response from the program's supporters, including leading nutrition experts at the Friedman School and elsewhere.

In today's announcement, Mike Hughes, chair of the Smart Choices Program and vice president for science and public policy at the Keystone Center, stood by the actual nutrition criteria used in the program. "Our nutrition criteria are based on sound, consensus science," said Hughes.

We suggested in the previous post, "the program could have considered stricter criteria in some areas, such as sweetened cereals. More importantly, it could have achieved a different emphasis even with the program's current criteria. It could have more strongly highlighted fruits, vegetables, and whole grains, while giving a lower profile to products that have been slightly reformulated and artificially enriched to just barely meet the nutrient criteria."

In either case, the FDA may be in a better position than the manufacturer-led Smart Choices Program to referee this question. In today's announcement, Hughes said, "[W]ith the FDA's announcement this week that they will be addressing both on front-of-package and on-shelf systems, and that uniform criteria may follow, it is more appropriate to postpone active operations and channel our information and learnings to the agency to support their initiative."

Setbacks for Smart Choices

The Smart Choices labeling program has created quite a stir.

Following a critical New York Times article earlier this month, which Ashley Colpaart discussed here, the story has been picked up by other major outlets. Rebecca Ruiz scrutinized the program's funding sources in Forbes magazine [update: sentence corrected 9/25/2009]. Mark Bittman shared his wit in a tour of a supermarket aisle on ABC's Nightline. Tom Laskawy at Grist called the program a dumb move.

My dean at the Friedman School at Tufts, Eileen Kennedy, who is a board member for the Smart Choices program, was quoted in the Times defending the inclusion of Froot Loops, which has become the poster child for questionable products included in the program. She has taken a lot of grief for this, including unfair emails and telephone calls. She argues, in person and in public, that the participating companies deserve credit for the social responsibility they showed in giving up their separate food labeling schemes and agreeing to the stricter "Smart Choices" standards. If Froot Loops meets well-defined standards, then wouldn't it be wrong to exclude the brand simply because one doesn't like its marketing associations?

The Froot Loops example highlights a weakness of the Smart Choices program, which seems to favor reformulated branded manufactured products over traditional simple healthy foods, such as fresh fruits and vegetables. Dean Kennedy responds that all fruits and vegetables without additives qualify for the program, a point that was omitted from the New York Times article.

The Smart Choices program would have been wise to anticipate the criticism that it favors highly processed foods. The program could have considered stricter criteria in some areas, such as sweetened cereals. More importantly, it could have achieved a different emphasis even with the program's current criteria. It could have more strongly highlighted fruits, vegetables, and whole grains, while giving a lower profile to products that have been slightly reformulated and artificially enriched to just barely meet the nutrient criteria. Then, reformulated Froot Loops might still have qualified, but the program would have been on stronger ground choosing a different poster child product -- any of thousands of simple, healthy, delicious, traditional foods.

Instead, the program's one-page fact sheet (.pdf) promotes plenty of manufactured packaged food brands but no traditional healthy foods. The program's board includes major manufacturers, but no producers or retailers of less processed fruits, vegetables, and whole grains.

The program should have anticipated criticism of its fee structure also. Although it has a sliding cost scale for food companies, with larger fees for products that have bigger sales, the low end of the scale is still too expensive for commodity producers (comparatively small-scale producers of a non-branded food product). If the program is not just a marketing ploy for food manufacturers, or a revenue stream for the non-profit program itself, then it should permit the seller of an apple without additives simply and freely to use the Smart Choices logo. If an apple automatically meets the program's criteria, it is difficult to see what type of review the program would undertake that would justify even a modest application cost. Currently, if you search for "apple" on the program's website, you find all about Apple Jacks and very little about apples.

Marion Nestle's blog has covered this issue with cutting insight. Advocacy groups have been having a field day. Change.org is running an email campaign, with thousands of signatories already. Somebody has apparently circulated an email list of people to contact that includes faculty like myself. I read every email with interest, even though there is not much mileage in lobbying me on this topic.

The Food and Drug Administration, which oversees federal policy on food labeling, wrote to the Smart Choices Program in August:

In the past five years, competing FOP [front-of-pack] symbols on food labels have proliferated. Consumer research suggests that these competing symbols, which are based on different nutrient criteria, are likely to confuse consumers. In this context, we recognize the potential value of a more standardized approach for FOP labeling.

However, since products bearing the Smart Choices symbol are just beginning to appear in the market, we will need to monitor and evaluate the products as they appear and their effect on consumers' food choices and perceptions.

FDA and FSIS would be concerned if any FOP labeling systems used criteria that were not stringent enough to protect consumers against misleading claims; were inconsistent with the Dietary Guidelines for Americans; or had the effect of encouraging consumers to choose highly processed foods and refined grains instead of fruits, vegetables, and whole grains.

Of course, some of FDA's concerns would seem to apply equally well to other front-of-pack labeling programs, not just Smart Choices.

The Board of Directors page on the Smart Choices website formerly listed affiliations for directors with senior roles at the American Diabetes Association, Baylor University, and Tufts University, but these affiliations have been removed. In an August 5 press release, the American Society of Nutrition (ASN) seemed proud to "jointly administer" the Smart Choices program along with a non-profit organization called NSF International: "Together, ASN and NSF International are committed to ensuring that the Smart Choices Program is credibly implemented, governed, and monitored." Now, however, some of the references to the ASN role have been deleted from the Smart Choices site -- for example, they have been removed from the one-page fact sheet (.pdf). ASN has tried to clarify its role in the program in a letter to members: "ASN does not own the program and does not endorse the products under Smart Choices." I wonder if leading institutions in the nutrition profession are reconsidering the program.

10 foods approved by the new Smart Choices program

This week the New York Times wrote an article criticizing the new industry sponsored Smart Choices program which aims to help "strapped for time" consumers make fast choices by way of a front-of-the-label logo. From the Smart Choices website:

The Smart Choices Program™ was created by a diverse group of scientists, nutritionists and food industry leaders, to harmonize existing front-of-pack nutrition labeling icons, symbols and systems. The intent is to provide a single, simple message for the consumer - regardless of which brands they buy or stores they shop in. Our vision is that the Smart Choices Program will be the most widely-used front-of-pack nutrition labeling program in the U.S. across retail channels and brands.

The Smart Choices Program provides a front-of-pack symbol and calorie indicator that helps consumers make smarter choices for products in 19 categories, including: cereals, meats, fruits, vegetables, dairy and snacks.

The categories also include: snack foods and sweets, desserts, water (plain and carbonated), and fats, oils and spreads.

Here are 10 foods approved by the labeling scheme:

10. Breyers Smooth & Dreamy Fat Free Ice Cream (Chocolate Fudge Brownie)- Unilever

9. Frosted Flakes Cereal (Original)- Kellogg

8. Cocoa Puffs Cereal- General Mills

7. Keebler Cookie Crunch (Original)- Kellogg

6. Country Crock (Churn Style)- Unilever

5. BAGEL-FULS Bagel with Cherry Filling & Cream Cheese (Cherry & Cream Cheese) -Kraft Foods

4. Healthy Choice French Bread Pizza (Simple Selections Pepperoni French Bread Pizza)- Conagra

3. Kid Cuisine- (All Star Chicken Nuggets, Campfire Hotdog, Carnival Corn Dog, Constructor Cheeseburger, Magical Cheese Stuffed Crust Cheese Pizza, BBQ Shake - Ups)- Conagra

2. Lunchables- Fun Pack (Chicken Dunks, Turkey and Cheddar Sub, Cheese Pizza)- Kraft

1. Betty Crocker Fruit Roll-Ups Crazy Pix (Cool Chix® Berry Wave)- General Mills

New advocacy coalition backs national menu labeling

The National Restaurant Association today announced support for national calorie labeling in chain restaurants.

The proposed legislation (.pdf), the Labeling Education and Nutrition Act of 2008 (LEAN Act), "will provide a national nutrition labeling standard for foodservice establishments with 20 or more locations."

The Center for Science in the Public Interest, a public interest group that has long supported restaurant nutrition labeling, joined with the restaurant trade association in supporting the bill. CSPI director Michael Jacobson has a related blog post at the Huffington Post.

Why would the restaurant industry, which has in the past strongly opposed such policies, now lend its support? There are several reasons. First, the bill is a compromise bill, providing the restaurant chains with some of their key policy priorities, including preserving a good deal of flexibility in deciding how to present the information and protection from what the restaurants describe as "frivolous" lawsuits. Second, the industry is facing the hard facts that menu labeling policies are succeeding at the state and local level around the country. As with other important nutrition labeling policies in the past, such as the current nutrition facts panel on packaged food, an important sector of the food industry chose to support a new government policy in return for more consistent and less burdensome regulation across jurisdictions.

Massachusetts approves calorie labeling measure

Massachusetts joined New York and California passing legislation to provide consumers the calorie information for the food they are purchasing at food establishments.

From the HHS press release:

Food establishments with 20 or more locations in Massachusetts will be required to provide that information at the point of purchase ― either on the menu board or on the restaurant’s menu. The new rules, which will take effect in November 2010, will cover approximately 50 chain restaurant companies, representing more than 5,000 locations in Massachusetts. An 18-month implementation timeline will allow local health departments and the industry the opportunity to familiarize themselves and prepare for the new regulation.

While more research needs to be done, preliminary research shows that providing consumers with calorie information does effect choice. A literature review by Harnack and French, published in the International Journal of Behavioral Nutrition and Physical Activity in 2008, showed that five out of six studies provided some evidence consistent with the hypothesis that calorie information may influence food choices in a cafeteria or restaurant setting. They claim, however, the results "suggest the effect may be weak or inconsistent."

Again, from the HHS press release:

Health regulations like the one passed today are popular with consumers. A study conducted in February 2009, gauging reaction to New York City’s calorie labeling law, showed that of those who visited restaurants with posted information, 89% considered it a positive change — and 82% report that nutritional information on menus had made an impact of their ordering.

The measure adopted today is part of Mass In Motion, a wide-ranging statewide initiative to promote a range of wellness activities for Massachusetts residents, businesses and communities. Last month, the Public Health Council also passed regulations allowing for Body Mass Index measurements for all school children in Massachusetts.

Additionally, health officials will soon announce grants for communities to establish wellness initiatives at the local level. These efforts, combined with an expanded state-sponsored Workplace Wellness program and an interactive web site (www.mass.gov/MassInMotion), represent the most comprehensive effort to date to deal with the serious problem of overweight and obesity in the Commonwealth.

The new regulation underwent a thorough public review process that included two public hearings and the submission of comments from more than 100 individuals and groups. More information at www.mass.gov/dph.

On a national note, Sen. Tom Harkin, D-Iowa, and Rep. Rosa DeLauro, D-Conn have introduced the Menu Education and Labeling Act (MEAL) geared at chains with 20 or more outlets. The restaurants would be required to post calories on menu boards and food displays. In addition, saturated fat, trans fat, carbohydrates and sodium would be required on printed menus. The house version of the bill (HR 2426) has 34 cosponsors and is in the House Energy and Commerce Committee. The senate bill (S 1048) has three cosponsors and is in the Senate Health, Education, Labor, and Pensions Committee.