The FDA Should Not Repeat Past Mistakes on Sugar

The FDA Should Not Repeat Past Mistakes on Sugar

By: Justin Schwegel

The FDA’s past dietary guidelines contributed to an obesity epidemic by encouraging food manufacturers to reformulate food by removing fat and adding sugar to make it more palatable. The new food labeling guidelines on added sugar are triggering a similar reformulation of processed products that could do little to benefit consumer health.

The obesity epidemic and past dietary guidelines

The United States faces an unprecedented obesity epidemic. A 2017 report from the Centers for Disease Control found that nearly 40% of adults and 20% of adolescents are obese.  The CDC also noted that obesity is correlated with, inter alia, an increased likelihood of all causes of death, an increased risk of type II diabetes, mental illness, coronary heart disease, hypertension, and certain types of cancers. Sugar consumption is a major cause of the growing epidemic.

In addition to the individual health costs of obesity, in 2010, Hammond and Levine found four major categories of societal costs associated with obesity, “direct medical costs, productivity costs, transportation costs, and human capital costs.” Direct medical costs alone were estimated to be as much as $147 billion per year in 2008 dollars (or over $175 billion in 2018 dollars).

Beyond these direct and indirect costs, there are also hidden costs such as the detriment to national security. Nearly 6 million men and 17 million women “exceeded the Army's enlistment standards for weight and body fat…” and a gain of “1% in weight and body fat would further reduce eligibility for military service by over 850 000 men and 1.3 million women.” Since these findings were published in 2012, the situation has continued to deteriorate.

America was not always obese. In the 1960’s and 70’s only 13 percent of adults were obese, but this increased to about 34% by 2012, nearly triple the earlier rate. The current upswing in obesity began in the late 1970s and early 1980s, at a time corresponding to the introduction of new dietary guidelines advising reduced consumption of saturated fats and cholesterol. The guidelines were based on research conducted by Ancel Keys commonly known as the Seven Countries Study. At the same time other health scientists, such as John Yudkin in the United Kingdom, were pointing the blame at sugar. Keys’s study stressed the correlation between consumption of saturated fat and negative health outcomes for men in seven countries. Fat tastes good. To make the reduced fat food more palatable industry plied it with sugar. American industry responded to the FDA guidelines with new low-fat, high-sugar product lines. American consumers responded with expanded waistlines. 

Recent analysis indicates that the seven countries in Keys's study may have been handpicked by Keys to support his hypothesis on the danger of fat. He excluded other European countries with available data, including France and West Germany, both of which had low rates of heart disease despite diets high in saturated fats and cholesterol. Even so, researchers who later reviewed the original data from the Seven Countries Study found that even in the data evaluated by Keys, eating sweets was more closely correlated with heart disease than eating saturated fat. At the same time as Keys was discrediting Yudkin’s findings on sugar the sugar industry paid a team of Harvard scientists to discredit any findings blaming cardiovascular disease on excess sugar consumption. Some researchers today argue that the “[d]ietary advice not merely needs review; it should not have been introduced.” 

Recent nutritional research (e.g. here, here, and here) indicates that replacing fats with carbohydrates, especially sugars, is not correlated with improved health outcomes. The FDA’s newest dietary guidelines no longer have a limit on cholesterol, but they do note the common correlation between cholesterol and saturated fats.

Added sugar, new disclosure obligations--useful information or misleading implication?

Acknowledging the link between sugar and detrimental health outcomes, the FDA’s 2015-2020 Dietary Guidelines for Americans recommend that Americans limit calories from "added sugars" and recommend that added sugars make up less than or equal to 10% of total daily calories. While it almost certainly is great advice to limit sugar intake, “added sugar” is not a meaningful concept when applied to most processed food products Americans commonly consume. The FDA has admitted as much when it stated that it was “not aware of an analytical method that is capable of distinguishing between added and intrinsically occurring sugars in a food product.” 

A standard four-pound bag of pure sugar at any grocery store has no added sugar in it under a standard interpretation of what the word “added” means. For the most part the contents of the bag are the remnants of sugar beets (or cane) with all of the fiber, water, vitamins, minerals and impurities removed. Yet “the FDA distinguishes added sugars from the fructose and glucose present in fruits and some vegetables, despite their identical chemical structures.”

To help consumers determine which foods have the most added sugar, the FDA has adopted a new rule that will require changes to the information panel on foods. The new label will contain a new category under “total sugars” called “added sugars” which will also bear a percentage of how much of your daily allotment of added sugars the product contains. 

Consumers will now have the information to distinguish between products where the sugar is naturally occurring and products where the sugar has been “added.” Their bodies will remain physically incapable of making the same distinction, as will analytical chemists.

So what's the point?

Whole fruits are a healthy part of a balanced diet. Whole fruits contain sugar. In oranges, for example, the majority of this sugar is fructose, commonly known as fruit sugar. Recent studies undertaken at the Mayo Clinic (and others) indicate that fructose may be more closely linked to obesity than glucose or sucrose as a dietary sugar, though these findings have been contested by the industry-funded International Food Information Council. The sugar present in most fruits is primarily fructose, a monosaccharide, while table sugar is comprised of sucrose, a disaccharide composed of a glucose and fructose molecule bonded together. 

Why, then, does the FDA distinguish between “natural” sugar and “added sugar” when one gram of sucrose is likely less harmful than one gram of fructose? The most obvious reason is that fruit contains vitamins, minerals, and most importantly, fiber. Fiber slows the absorption of sugar into the bloodstream, prevents spikes in blood sugar, and provides a feeling of satiety that prevents overeating. 

However FDA’s binary classification of sugars as “added” or “naturally occurring” does not adequately regulate naturally occurring fruit sugars that have been heavily processed, even when they form a part of the ingredients of a processed product. Oranges, for example, can have 100% of their fiber removed to make pulp-free orange juice, and there is no “added sugar.” This process (juicing) is also the first phase in sugar refinement.

This has dietitians concerned. Most dieticians recognize that whole fruits generally do not carry a risk of sugar overconsumption due to the presence of fiber to slow absorption and promote satiety. However, dietitians worry that many people incorrectly think that fruit juice is healthy, when removing fiber during the juicing process makes it easier to both over consume sugar and absorb the sugar too rapidly. The new FDA guidelines on added sugar will only reinforce this misconception and further promote unhealthy consumption choices.

The arbitrariness of the rule has forced the FDA into binds already. The result could mean complicated math for food manufacturers that reformulate products in response to the new labeling requirement. The guidelines allow a food manufacturer to use as a food ingredient dehydrated fruit juice, i.e. fruit which has had 100% of its fiber and most of its water removed. This process is similar to refining cane or beet sugar, but there are important differences in how this type of sugar is counted. According to the FDA, for a processed product made this way the amount of “added sugar” would be the “sugars…in excess of what would be expected from the same volume of 100 percent fruit or vegetable juice of the same type.” As a result, only some of the sugar in a processed product containing sugar from highly processed oranges would need to be declared as “added.” However, the FDA has issued contradictory guidance for the expressed juice of sugar cane, and presumably 100% of the sugar from this evaporated juice would be by definition, added sugar. If food manufacturers use evaporated cane juice as a sweetener it counts as 100% added sugar; if food manufacturers use evaporated apple juice as a sweetener it doesn't count as 100% added sugar. There is no good explanation why.

The FDA initially waffled, but ultimately decided that tomato paste (a processed form of tomato with much of the fiber removed and which contains more sugar than a similar volume of whole fresh tomatoes) will not have any added sugar. However, some fruit and vegetable powders (it's not clear which) will have added sugar. The FDA also seems to have backtracked on honey and maple syrup, both of which must be listed as “added sugar” when added to a product, but perversely will likely not need to have their pure form products labeled as containing “added sugar” when final guidance is drafted.

Even more confusing could be whether a product that has been processed specifically to have sugar removed could have negative added sugar. A cup of Fa!rlife ultrafiltered skim milk, for example, has six grams of sugar compared to 12 in traditional skim milk. Given that under the FDA’s new rule, skim milk would have zero grams of added sugar, it would stand to reason that Fa!rlife could legitimately claim to have negative six grams of added sugar. Fa!rlife also makes a chocolate milk product that has 12 grams of sugar, where the 12 grams naturally occurring lactose has been removed to make it friendly for lactose intolerant consumers and another sugar (presumably sucrose) has been added. Would such a product have zero grams of added sugar? 12? What if Fa!rlife re-added some lactose instead of sucrose? The inherent arbitrariness of the rule has already sparked significant backlash.

Industry has begun responding to the arbitrary rule with arbitrary product reformulation

The FDA’s current guidelines do not meaningfully address the total sugar contained in processed products. It categorizes sugars as naturally occurring and therefore not problematic, or added and therefore subject to restricted intake. However, the actual health impact of a food is determined by its entire composition, including the amount of sugar, fiber, vitamins, minerals, fat, protein, sodium, etc, and not the source from which those contents originate. However, with the new FDA guidelines, food manufacturers, and health food manufacturers especially, are switching their formulations to avoid the “added sugars” that are subject to restricted intake. A recent and concerning trend in sugar replacement is to use dates as the first ingredient in “health” bars.

Some reformulated health or nutrition bars contain extraordinary amounts of sugar as a proportion of their total weight. The Snackless Bar is up to 43% sugar; the RxBar is up to 29% sugar; the That’s It bar is up to 69% sugar; and the Larabar is up to 36% sugar. These products contain no declared “added sugar.” Most of these products contain dates as the first ingredient. Given that dates may contain in excess of 70% sugar, primarily in the form of fructose and sucrose (which is 50% fructose) and little dietary fiber, their use as a first ingredient in a “health food” is cause for concern. While the U.S. sugar industry’s commercial motivations were clear, it foresaw this rush to reformulation and its dangerous potential to mislead consumers in 2014.

This is just the tip of the spear in product formulation. Given the perverse incentives the FDA has created with its new rule, it is just a matter of time before more reformulation is undertaken that will likely retain the nutritional (and sweetness) characteristics of familiar processed products while minimizing the amount of sugar considered "added." Another foreseeable concern is that improvements in both conventional breeding and gene-editing technology could result in varieties of fruits and vegetables with "naturally" higher sugar content and less fiber. This could further allow food manufacturers to create a palatable product while avoiding the commercially unpopular “added sugar” labeling requirement.

Implicit labeling claims that mislead consumers

Given the FDA’s not so distant history of propagating dietary guidelines focused on reducing fat consumption, fat content claims are likely a source of confusion to consumers. Yakult, a widely available product bears a “non-fat” label. The most common commercial uses it as a selling point. The problem with fat content claims is that consumers are often misled into thinking the food is healthy. This is not necessarily the case, especially with high sugar, low fat foods.  Low fat claims are already regulated under 21 CFR 101.62(b)(i-iii), but this framework proves problematic because of the implication to consumers. One Yakult drink, for example, has 11 grams of sugar in a 2.7 oz serving.

Because of the link between sugar and obesity, consumers could often choose foods that are more likely to cause fat weight gain by selecting a nonfat or low fat option. FDA could require a disclaimer for fat free or low fat claims on products high in sugar. The disclaimer could read "This product contains sugar and sugar consumption is strongly correlated to increased risk of diabetes and obesity," or something to that effect. Similar disclaimers are required for dairy producers who label their products as free of recombinant bovine somatropin (rBST). The FDA has suggested that any rBST-free statement be accompanied by the disclaimer “No significant difference has been shown between milk derived from rBST-treated and non-rBST-treated cows.” This is based on language in the Food Drug and Cosmetic Act which indicates that “both the presence and the absence of information are relevant to whether labeling is misleading.” Given the FDA's own role in misleading the public on the relative dangers of fats vis-a-vis sugars this type of disclaimer may be warranted.

Another concern regarding true but misleading labeling claims relates to glycemic index, which is a measure of the relative rise of blood glucose level two hours after eating a food. Packaging for many foods now contains a GI (glycemic index) number. A food’s GI number can be useful labeling information for diabetics. However, this information seems unlikely to be useful for non-diabetics attempting to ensure a healthy diet and avoid weight gain. Pure fructose is a low glycemic index food (19 out of 100), and as noted above, fructose consumption can cause obesity despite its low GI number. Labels with a GI number may cause consumer misconceptions about what constitutes a healthy food if unaccompanied by a disclaimer.

Is there a better approach?

It is difficult to plot the correct path forward. The World Health Organization uses the construction “free sugars” to differentiate between the sugars that are present in whole fruits, vegetables, and milks from sugars that are present in table sugar, honey, syrups, and fruit juices. This construction is more useful than the one adopted by the FDA because it correctly identifies that both beverages that have had sugar added to them and fruit beverages that have had many of the things that are not sugar removed from them, pose a similar threat. But it is not clear how this approach would treat processed products.

The best solution would be to educate the public on what constitutes nutritious food based on an appropriate levels of fiber, essential nutrients, protein, fats, sodium, sugars and other carbohydrates. This is likely impossible in the short term. A second best solution would be to treat all sugars the same, regardless of source, in any processed food or beverage and to express this amount of sugar as a percentage of daily recommended intake of all sugar. This would avoid encouraging manufacturer sleight of hand that ignores the nutritional value of food in favor of limiting what sugars are arbitrarily considered “added.” So long as the FDA fails to take this kind of approach, food manufacturers will continue to sell candy as a health food.