Enhancing the Effectiveness of Food Labeling in Restaurants

Starting with New York City in 2006 and more recently in California and Massachusetts, an increasing number of municipalities and states have enacted or are considering legislation mandating calorie posting by fast food restaurants. In recent months, the House and Senate introduced health care reform bills with provisions for menu labeling. The House bill—based on the Menu Education and Labeling (MEAL) Act—would require chain restaurants with 20 or more locations to post the number of calories, grams of saturated fat, and milligrams of sodium next to menu items.

Modifying fast food intake is a sensible target for a society confronted with an obesity epidemic that has far-reaching health and economic consequences. The Nutrition Labeling and Education Act of 1990 led to the ubiquitous labeling of packaged foods. However, the total proportion of food consumed in restaurants has increased steadily since then, and there is a strong, albeit only suggestive, correlation between obesity and patronage of fast food restaurants.

It is unlikely, however, that the type of calorie labeling going into effect around the country will have much effect on the prevalence of obesity. Existing evidence on the effectiveness of point-of-purchase calorie labeling is equivocal at best.¹ Calorie labeling can potentially be helpful but whether it is will largely depend on how it is implemented.

Given the momentum of efforts to encourage food labeling—and the opportunity cost of investing in this approach as opposed to other alternatives—it is important to consider the reasons such efforts are unlikely to have the effects that policy makers seem to anticipate.

Although research suggests that consumers consistently fail to correctly estimate the amount of calories in a given food item, evidence is limited on whether individuals generally underestimate or overestimate the calories in food. More research is needed to address this issue.^{1 - 2} If the tendency is to overestimate, accurate information could promote increased consumption, for instance, leading to ordering double as opposed to single cheeseburgers. Dieters are likely to be especially prone to such overestimation if they subconsciously exaggerate the caloric content of food as a strategy to restrain their intake.

Providing calorie information will only be helpful if individuals know how to interpret it. Past research has revealed low levels of understanding of calorie information, as studies have suggested that only 20% of participants can accurately calculate the contribution of a single food item to recommended daily intake.² This is likely to be especially problematic for obese individuals with low socioeconomic status, who disproportionately patronize fast food restaurants and tend to have low levels of numeracy.² A recent study examining the association between calorie labeling and food choices on low-income adults in New York City found that most individuals reported no effect of caloric information on food choices; even those who reported calorie labeling influenced their food choices did not purchase fewer calories.³ Even in cases in which individuals reported an association between labeling and purchase of low-calorie foods, it is not known whether these links are causal, as individuals who intend to order low-calorie meals may notice this information at higher rates.⁴

Optimistic views of the effects of providing calorie information are implicitly based on the assumption that lack of such information is a major driver of overeating. While information provision could be improved, many if not most adverse health behaviors likely result not from lack of information, but from self-control problems.⁵ Most obese persons are aware of that problem, and many if not most want to lose weight but have been unable to do so. Perhaps the biggest contributor to obesity is that dietary modification requires making difficult trade-offs between immediate gratification and delayed benefits. Each single indulgence has little effect on weight and most individuals tend to heavily discount the future so it is easy for them to rationalize that the short-term benefits exceed the costs.⁶

Given these concerns, it is important for regulatory entities to consider ways to increase the effectiveness of food labeling before further legislative initiatives mandate broader application of food labeling.

Much information provided in medical or public health settings is presented in ways that some consumers may not understand. If individuals do not know what 200 calories means, such information is unlikely to be useful.⁷ Presenting calorie information along with calorie targets and representing calories as exercise equivalents are approaches that could highlight the effective cost of eating a high-calorie item, but the relative effectiveness of such approaches has not been adequately tested.⁸ Presenting calories graphically as a proportion of a targeted daily intake could help demonstrate the energy consumption associated with a particular food item. Using a grading system similar to the Hannaford Supermarkets' Guiding Stars system, which rates foods with stars based on nutrient quality per 100 calories, might also be effective. All these approaches have conceptual appeal but need to be rigorously tested, particularly their effectiveness in different target populations.

Information framing is also crucial to efforts to improve food choices in fast food restaurants. Framing information in terms of how many pounds an individual would gain over a year if that person ate that item once a day might help translate the information into terms that better reflect the effects of specific food choices on weight. Different ways of framing calorie information should be tested to find feasible ways to provide accurate information that maximize the beneficial effects on health.

Recent work has demonstrated that changing defaults by listing healthy choices on the front page of a menu was significantly associated with purchase of lower-calorie sandwiches whereas putting caloric information on the menu was not.⁹ Defaults and more effective labeling could be used in a complementary fashion. Many of the current defaults at fast food restaurants are unhealthy. Designing a menu board or menu so that the healthier (lower-calorie) options appeared first, or were printed in larger and bolder font, may help individuals identify and choose these healthier items. Restaurants could replace the default soda and fries in a combination meal with a bottle of water and a side salad or fruit. Lower unit prices of “supersized” sodas and fries encourage purchase of the largest sizes. Changing the differential prices of healthy and unhealthy food choices and of different sizes could be another way of manipulating defaults to guide consumers toward healthier food choices.

There are good reasons to doubt whether caloric labeling, as currently being implemented in different municipalities and proposed nationally, will be effective in increasing healthy food choices. Careful testing of ways to make the information more helpful should be done prior to wide-scale implementation.

The effect of well-designed food labeling and defaults will be magnified by restaurant responses if consumers change their behavior. When Los Angeles required restaurants to post hygiene ratings in windows visible to customers, this not only shifted patterns of patronage but led to an improvement in hygiene and a significant decrease in hospitalizations from food-related illnesses.¹⁰ Menu labeling likely can be made more effective, but it is important to recognize that this is just one of a series of well-designed complementary approaches necessary to address the obesity epidemic.