Home » Feature Stories » Informed Ingestion

What’s NOT on the Menu?

By Matt McGowan

Americans consume 54 billion meals a year at fast-food and table-service restaurants. This means that, on average, they eat a meal outside the home about every other day. Research has shown that these meals have contributed to the rise in obesity.

Scot Burton and Betsy Creyer are willing to bet that if Americans knew the number of calories and amount of fat and sodium in many meals consumed outside the home, they might choose a more nutritious menu item. They aren’t psychologists, but Burton and Creyer think many people would consider modifying their behavior if restaurants provided nutrition information on menus and menu boards. The information could help consumers make wiser decisions about the type of food they eat, which in turn could have an overall positive impact on public health.

For the past several years, Burton and Creyer, marketing professors in the Sam M. Walton College of Business, have studied consumer knowledge and attitudes about food purchased at McDonald’s, Subway, Chili’s, Applebee’s and scores of other establishments in which Americans eat more than half their meals and spend more than $400 billion annually. Specifically, Burton and Creyer examine the potential health benefits associated with providing nutrition information and the effects that information could have on consumer perceptions about health and disease.

In two recent studies, they investigated the accuracy with which consumers estimate the nutritional content of restaurant meals and how consumer attitudes about those meals change when nutrition information is provided. The researchers made two important discoveries.

A majority of people consistently underestimate the number of calories and amounts of unhealthy nutrients in the food they eat at restaurants. This is especially true for diners who prefer hamburgers and fries, baked potatoes loaded with cheese, sour cream and bacon bits and other meals with many calories and a lot of fat and sodium.

“Our findings illustrate how poorly consumers understand the nutritional content of many of the meals they eat outside the home,” Burton says. “Reasonable consumers know that a large number of these items aren’t healthy, but they still do not realize how unhealthy the meals can be and the possible effects of frequent, long-term consumption.”

Perhaps more importantly, Burton and Creyer discovered that when consumers are given specific and objective nutrition information about “unfavorable” meals -- those high in calories, fat and sodium -- they regard those meals unfavorably. In other words, if consumers know fettuccini Alfredo has 1,500 calories and 97 grams of fat -- more than 30 grams above the recommended daily amount -- their attitude about that dish changes for the worse. They perceive the meal to be less healthful and more likely to contribute to obesity and heart disease. The nutrition information also motivates them to change their purchase behavior. 

“When we tell them how many calories or how much fat the food has, they get it,” Creyer says. “They see this and say, ‘Oh my gosh, I didn’t know. I didn’t know it had so much fat.’”

Who cares? So what if your Big Mac or Whopper has almost twice as much fat as you think? Does it really help to know that your turkey sandwich with mustard instead of mayonnaise has slightly fewer calories than you think it has? Several facts underscore the relevance of Burton and Creyer’s research:

More Americans are eating out more often. They eat 54 billion meals a year at fast-food or table-service restaurants. That’s more than a billion meals every week. On any given day, Americans spend more than $1 billion on restaurant food. According to the National Restaurant Association, American consumers will spend about $476 billion in 2005 on meals purchased and consumed outside the home. Approximately half of all the money Americans spend annually on food is spent in restaurants.

Also, Americans are fat. Recent reports indicate that almost two-thirds of all adults in the United States are obese or overweight. According to the Centers for Disease Control and Prevention, obesity rates for adults doubled between 1980 and 2000. During this same period, overweight rates doubled for children and tripled for adolescents.

Americans are sick and dying because they are fat. While public health officials and policymakers have softened assertions that obesity alone is an epidemic, there is no doubt that obesity has a positive connection to type 2 diabetes, cardiovascular disease and other chronic illnesses. Initial figures of death rates due to
obesity were overestimated, but David Satcher, former U.S. Surgeon General, has warned that obesity will soon be the leading cause of preventable death in the United States.

Obesity hurts the U.S. economy. In 2003, an estimated $75 billion in health care costs were attributed to obesity. The CDC reports that from 1979 to 1981, U.S. hospitals spent $35 million treating obese and overweight children and adolescents. From 1997 to 1999, hospital costs for the same population groups soared to $127 million.

There are direct connections between the consumption of restaurant food and the rise in obesity. One recent study found a significant relationship between frequent fast-food dining and increases in body weight and insulin resistance, the two primary risk factors for type 2 diabetes.

Some large restaurant chains understand this connection and have exploited it as powerful marketing strategy. Subway, the fast-food chain with 23,741 restaurants worldwide, encourages customers to choose one of seven sandwiches with only six grams of fat. Chick fil-A, encourages consumers to “eat mor chikin” if they’re counting carbs, fat or calories.

Despite the health claims, restaurants, unlike producers of packaged food for sale in grocery stores, are not required to disclose nutrition information. However, this could change. Rep. Rosa DeLauro of Connecticut and Sen. Tom Harkin of Iowa have introduced the Menu Education and Labeling Act (MEAL), a potential sibling of the Nutrition Labeling and Education Act (NLEA), the 15-year-old federal law that led to the ubiquitous Nutrition Facts label on packaged food sold in markets, convenience stores and vending machines. If passed, MEAL would require table-service restaurant chains with 10 or more outlets to disclose calories, saturated fat and sodium levels on menus. Fast-food chains would be required to provide calorie information only adjacent to items on menu boards, the large, often-lighted signs near counters where customers order food. Some states, including Arkansas, have drafted their own bills based on MEAL’s language. 

For their first study, Burton and Creyer asked participants to estimate the number of calories and amount of fat, saturated fat and sodium in nine popular entrees served in dinner-house restaurants. Before estimating the number of calories and nutrient levels, participants were given brief descriptions and serving-size information for each meal. 

In designing the study, researchers selected meals that they categorized as “favorable,” “unfavorable” and “very unfavorable.” Favorable items -- chicken breast, pot roast and turkey sandwich -- contained no more than 640 calories and 26 grams of fat. Unfavorable items -- fettuccini Alfredo, hamburger and fries, chicken fajitas, chef salad and patty melt and fries -- ranged from 930 to 1,660 calories and from 63 to 97 grams of fat. One very unfavorable item, cheese fries, had 3,010 calories and 217 grams of fat.

Differences between estimated levels of calories and nutrients and actual levels varied greatly depending on the meal category. Most importantly, for the six unfavorable meals, participants consistently and significantly underestimated the number of calories and amount of fat, saturated fat and sodium. In other words, calories, fat and sodium in less favorable meals such as hamburger and fries and chicken fajitas are much higher than consumers expect.

For example, participants on average thought a large hamburger and fries had 777 calories, 37 grams of fat, 17 grams of saturated fat and 523 milligrams of sodium. The meal actually had 1,240 calories, 67 grams of fat, 29 grams of saturated fat and 1,270 milligrams of sodium. In other words, people may have a vague idea that a hamburger and fries isn’t good for them, but they have no idea that that meal alone contains more grams of fat and saturated fat than the recommended daily allowance for adults.  

For the three favorable items, consumers either slightly underestimated or even overestimated the number of calories and amount of undesirable nutrients. For example, on average participants estimated a turkey sandwich with mustard instead of mayonnaise to have only 12 fewer calories than it actually has. Also, they thought the sandwich had nine more grams of fat and five more grams of saturated fat than it actually has.

“We hypothesized that a greater percentage of consumers would underestimate calorie and nutrient levels for the meals with more unfavorable nutritional content,” Burton says. “The data confirmed this prediction. The accuracy of estimates for the more favorable items shows us that people who prefer more healthful meals probably have a more realistic idea about the level of undesirable nutrients, such as fat and saturated fat, in those meals.”

In fact, the researchers found that the more unfavorable an item in terms of undesirable nutrients, the larger the gap between consumer expectations and actual nutrition levels. For example, consumers underestimated the only “very unfavorable” item -- cheese fries, which had 3,010 actual calories and 217 actual grams of fat -- by 2,141 calories and 177 grams of fat. They underestimated “unfavorable” items by an average of 642 calories and 44 grams of fat. For “favorable items,” participants underestimated by an average of only 43 calories and overestimated by an average of eight grams of fat. Another way of saying this is that 97 percent of the participants underestimated the grams of fat for a very unfavorable item, 90 percent of the participants underestimated the grams of fat for an unfavorable item and only 37 percent of the participants underestimated the grams of fat for a favorable item.

Knowing that consumers are generally unaware of the high levels of undesirable nutrients in restaurant meals, Burton and Creyer then asked: What are the potential health benefits associated with providing nutrition information on menus? They predicted that access to nutrition information would influence consumer purchase intentions and perceptions about the likelihood of weight gain and heart disease.

Participants in this portion of the study received a survey and mock menu with four items: deluxe hamburger with fries, chef salad, turkey sandwich and chicken breast with baked potato. They also received information on the daily value recommendations for fat (65g), saturated fat (20g) and sodium (2,400mg), based on a 2,000 calorie diet.

The researchers designed this part of the study to simulate the consumer experience if MEAL or similar legislation requiring the provision of nutrition information were passed. Some participants received menus with calorie and nutrient information next to each meal choice. Others received menus with calorie information only, simulating menu boards, and a third group received menus without any calorie or nutrient information.

Burton and Creyer asked participants to choose a menu item and, with the nutrition information provided, explain the choice. Also, participants were asked to share perceptions about the likelihood of weight gain and heart disease as it related to choosing one product instead of another on a regular basis.

The researchers discovered that providing calorie and nutrient information increased choices for the turkey sandwich, the item with the fewest calories and least amount of fat and saturated fat. Likewise, the provision of nutrition information caused participants to decide against meal items with calorie and nutrient levels that were worse than expected. Burton and Creyer emphasized, for example, that choices for the hamburger and chef salad items decreased from 37 percent to 24 percent when the number of calories and amount of fat, saturated fat and sodium were provided.

Furthermore, for those participants with menus that did not include nutrition information, the researchers could not find a marked difference in perceptions about weight gain and heart disease between the turkey sandwich, chef salad and chicken with potato. However, when the information was present, there was a large difference in disease-risk perceptions.

What about fast food? Do consumers consider the nutrition content of the meals they eat at Burger King, Subway, Sonic, Taco Bell and scores of other establishments Americans patronize on their lunch breaks or on the way home from work?

Previous research by Burton and Creyer demonstrated that consumers generally can distinguish the relative healthiness of one meal compared to other meals. In other words, most reasonable consumers know that a grilled chicken sandwich is better for you than a Big Mac. But, do they know how healthy or unhealthy these items are? How accurate are their expectations of nutrition content, and would point-of-purchase nutrition information change consumer attitudes about specific meal items?

To answer these questions, the researchers conducted a follow-up study of consumer experiences at fast-food restaurants. A new group of subjects -- 155 undergraduate students -- were asked to maintain a diary of their purchases at fast-food restaurants over a one-week period. Researchers then asked the participants to estimate the number of calories and amount of fat, saturated fat and sodium for each meal purchased.

In the third stage of data collection, participants consulted the restaurants’ Web sites to obtain actual levels of calories and nutrients. Finally, researchers asked the students to repeat their ratings for the meals. Specifically, they were asked to report their perceptions of meal healthiness, likelihood of gaining weight and heart-disease risk and the likelihood that they would repurchase those meals.

The fast-food study confirmed findings of the dinner-house study and answered several other questions. Most importantly, results of the subsequent study illustrated how poorly consumers understand the nutritional content of many fast-food meals. The researchers found that consumers significantly underestimate calories and levels of undesirable nutrients in meals that contain a lot of those nutrients. 

This was especially true for young men. The researchers discovered that male participants eat at fast-food restaurants more than four times a week and, as predicted, offered estimates of nutrition content that were more inaccurate than those estimated by the female participants. In other words, young men eat fast food more often than young women and have a poorer understanding of how unhealthful the fast-food meals they eat are. Male participants thought the average meal contained 324 fewer calories than it actually had. They also underestimated the amount of fat in the average meal by 14.7 grams.

“For the week, this suggests that male participants consumed a full day’s worth of fat and some 1,400 calories they don’t know about,” Burton says. “This would project to more than 72,000 calories annually. This number of unknown calories per week is clearly capable of resulting in significant long-term weight gain if the consumption was not modified over time or counterbalanced by a significant amount of exercise.”

For all participants, knowledge of specific nutrition information about unfavorable fast-food meals led to unfavorable evaluations of those meals. In other words, with access to nutrition information for meals with high levels of undesirable nutrients, participants regarded those meals as unhealthy and thought the meals would increase the likelihood of weight gain and heart disease. Access to the information would also decrease their intentions of purchasing the meal again.

However, the researchers emphasized that participant exposure to nutrition information for lower calorie meals led to no significant changes or more favorable evaluations of those meals. Burton says this finding could result in market advantages for restaurant chains serving lower calorie meals.

Burton and Creyer’s research shows that attitudes about unhealthy meals change when consumers have access to specific information about calories and undesirable nutrients contained in those meals: Perceptions are more unfavorable or negative. Consumers think less of the meals, but does this mean they would choose a different, more healthful meal? Would they say, “I’ll have a grilled chicken sandwich without mayonnaise” instead of “I’ll take a Whopper, a large order of fries and chocolate milkshake”?

The researchers think many people, particularly those concerned about their weight or health, would modify their behavior and pick the chicken sandwich. They argue that disclosure of nutrition information at the point of purchase may reduce consumer choices for higher calorie, less nutritious items and may even lead to restaurants offering more nutritious meals.

“There’s a segment of the population that wants this information to be available,” Burton says. “And there’s a segment of the population who couldn’t care less. But, for those who do care, our findings suggest significant long-term health benefits may be obtained by providing nutrition information on menus and menu boards.”