Birgit Humpert
May3, 2013
Topic:
What are effective interventions to improve healthy food choices in workplace cafeterias?
Draft Research Questions
1. How can food labeling influence food choices in workplace cafeterias?
2. How does menu labeling with logos compare to menu labeling with calories in effecting food choices in workplace cafeterias?
Studies in workplace or university cafeterias:
Chu, Y., Frongillo, E., Jones, S., & Kaye, G. (2009). Improving patrons' meal selections through the use of point-of-selection nutrition labels. American Journal of Public Health, 99(11), 2001-2005.
This study examines the effect of nutrition labeling in a university cafeteria. Sales of 12 hot entrees were captured in a baseline period, a treatment period, and a post-treatment period. In the treatment period boards with simplified nutrition labels (total energy, serving size(g), fat(g), protein(g), carbohydrates(g) were displayed, the patrons had to go past the board to get to the counter.
From the last day of the baseline period to the first day of the treatment period the average energy intake fell 12.4 kcal. The average energy intake kept falling during the treatment period by 0.298 kcal/day. The impact was seen most profoundly with high-calorie entrées. Average energy intake started to rise immediately after the treatment period, when the boards were taken down by 1.512 kcal per day.
Michels, K., Bloom, B., Riccardi, P., Rosner, B., & Willett, W. (2008). A study of the importance of education and cost incentives on individual food choices at the Harvard School of Public Health cafeteria. Journal of the American College of Nutrition, 27(1), 6-11.
The authors investigated the influence of price and education on food choices in the cafeteria of the Harvard School of Public Health in Boston, MA. According to dietary guidelines foods were classified as healthy or less-healthy. During the intervention period prices for healthy foods were decreased by 20%. The intervention was marketed and supported by the education material that was provided as flyers and table tents. During the first two days a nurse provided blood pressure readings. Prices returned to normal after the intervention period of five weeks.
Sales data from baseline, intervention, and follow-up period showed that consumption of healthy items increased in the intervention period and remained high or even increased further in the follow-up period: sales of stir-fried dishes increased 27 % and stayed 25 % over baseline after prices returned to normal. Salad bar sales increased 15 % and increased further to 53 % after the intervention. Sales of less healthy items decreased and stayed below baseline in the follow-up period. Only cake and dessert sales increased in the intervention period (56 %) and stayed increased (59 %).
The authors argue that participants became more aware of healthy food choices in the intervention period and maintained those healthy habits after prices were returned to normal. The increased sale of desserts might indicate that consumers treated themselves to dessert after choosing a healthier item, so that overall calorie consumption remained the same.
Steenhuis, I., Assema, P., van Breukelen G., Glanz, K., Kok, G., de Vries H. (2004). The impact of educational and environmental interventions in Dutch worksite cafeterias. Health Promot Int 19 (3): 335-343.
The authors examined different interventions in seventeen worksites in the Netherlands. One group of worksites received an education program to increase the consumption of fruits and vegetables. In another group the availability of low-fat products, fruits and vegetables in the cafeteria was increased, new foods were labeled, and promoted with posters and table tents. In a third group there was only labeling of low-fat items, fruits, and vegetables.
Participants were asked to fill out a questionnaire that asked for a recall of their last meal eaten in the cafeteria and about fruit, vegetables, and fat intake. In addition sales in the cafeteria were measured.
The study found no significant difference in fat, fruit, and vegetable intake between the different groups. All groups saw a decrease in vegetable consumption which might have been seasonal. For some products, for example desserts, a trend toward lower consumption could be seen, but it was not significant, possibly due to the small sample size. The author also speculates that the short length of the intervention (1 month) and the population (mostly men) could explain the ineffectiveness of the intervention.
Thorndike, A. N., Sonnenberg, L., Riis, J., Barraclough, S., & Levy, D. E. (2012). A 2-phase labeling and choice architecture intervention to improve healthy food and beverage choices. American Journal of Public Health, 102(3), 527-533.
This study examines the effect of a 2-phase intervention on the sales of food and beverages in the main cafeteria of Massachusetts General Hospital. In the first phase all foods and beverages were labeled green, yellow, or red on the basis of the USDA My Pyramid guidelines from 2005. This intervention was advertised; information material and a dietitian were available for questions. In the second phase sandwiches, chips and beverages were rearranged so that green items were on eye-level and water was more readily available.
In phase 1 sales of red items decreased 9.2 % and sales of green items increased by 4.5 %. After the second intervention sales of green items decreased 0.8 % but sales of red items decreased further by 4.9 %. The biggest impact was seen in beverage choices. Sales of red beverages decreased 16.5 % after the first intervention and 11.4 % after the second intervention; green beverages increased by 9.6 % and then further by 4 %.
Vyth, E.L., Steenhuis, I.H., Heymans, M.W., Roodenburg, A.J., Brug, J., Seidell, J.C. (2011). Influence of a nutrition logo on cafeteria menu items on lunchtime food choices at Dutch work sites. J Am Diet Assoc. 0011: 111:131-136.
The authors examine the effect of labeling foods in workplace cafeterias in the Netherlands with the Choices nutrition logo. 24 cafeterias were included in the study and randomly assigned to either the intervention or the control group. The cafeterias had to offer the same items (2 soups and 2 sandwiches) in addition to their regular foods. One of the sandwiches and one of the soups fulfilled the criteria for a Choices item, and in the intervention cafeterias those items were labeled accordingly. In addition to soups and sandwiches fresh fruit was also labeled with the logo. Data was collected in a baseline period, the intervention period, and the post-intervention period. In addition employees were questioned in a survey about their intentions behind food choices.
The authors found no significant change in the sales of sandwiches and soups labeled with the Choices logo. The only significant increase was seen in the sale of fresh fruit. The questionnaire showed that 56 % of employees had seen the labeling and 82 % were familiar with the logo that is also used on supermarket food in the Netherlands. It was found that participants who reported “intention to eat healthy” in the baseline period were more likely to choose the labeled items, but overall those participants were in the minority which might explain the outcome.
Studies from full-service restaurants:
Almanza, B., & Hsieh, H. (1995). Consumer preferences among nutrition labeling formats in a restaurant. Journal Of The American Dietetic Association, 95(1), 83-85.
The authors compared three labeling systems in a restaurant associated with the Purdue University. Theyalso conducted a survey to determine attractiveness, ease of use, and other attributes. The first labeling system used apples to indicate that a meal met three out of five criteria based on the nutrition guidelines. The second labeling used colored dots to indicate which criteria were met. The third labeling system used a pamphlet that gave information about the nutrition content in a numeric format.
Eighty-six surveys were completed. It was shown that costumers preferred the apple labeling or the leaflet depending on demographic attributes. Overall the apple received the highest ratings for ease of use and least time required. Education level also played a role. The more complex leaflet got higher ratings from costumers with a higher education level.
Ellison, B. (2012). Three essays on the effects of calorie labeling in full service restaurants. (Doctoral dissertation) Retrieved from
The author conducted two field experiments to research the effect of food labeling.
1. The value and cost of restaurant calorie labels: results from a field experiment.
In a restaurant on the campus of Oklahoma State University three different sections of the restaurant were assigned different menus for 19 weeks. The first menu didn’t provide any calorie information, the second menu offered the calorie content of the menu items, and the third had calorie information and “traffic light” symbols (red for high calories, yellow for medium calories, and green for low calories). After 12 weeks there was also a price manipulation: four high-calorie items were increased in price, three low calorie items were decreased in price.
Both labeling intervention resulted in an overall calorie reduction, but the “traffic light” version lead to a more significant decrease. Both labeling options had a greater effect than price manipulation on calorie consumption.
2. Do calorie labels reduce caloric intake? Evidence from two full service restaurants.
In the first restaurant menus were labeled with calorie content of the items and sales were compared before and after the intervention. In the second restaurant different sections of the restaurant received different menus. One section had normal menus, the second had menus with calorie labeling, and the third had calories and a “traffic light” symbol next to the menu items. Both restaurants had similar clientele, but the first emphasized healthy food, the second focused more on the dining experience.
In the first restaurant menu labeling with calorie did not lead to a difference in ordered calories, even an insignificant increase in ordered calories was seen. In the second restaurant labeling with calories lead to a reduction in ordered calories, but the change was not significant. When “traffic light” symbols are added a significant reduction from 741 to 672 calories can be seen.
Heathcote, F. F., & Baic, S. S. (2011). The effectiveness and acceptability of a traffic light labeled menu with energy information to signpost customers towards healthier alternatives in a table service restaurant. Journal Of Human Nutrition & Dietetics, 24(4), 390-391.
The authors evaluated a menu for an independent full-service restaurant that offered two dishes in either a normal or a healthier version. The dishes were labeled with “traffic light” symbols and sales were analyzed over a two-week period. Customers were also surveyed.
46 % of customers chose the healthier version of the dishes. 90 % of customers stated that they would order or did order a healthier version. The survey also revealed that costumers who ate out more often are more likely to order a healthier version. But patrons who ordered a healthier dish were also more likely to order dessert.
Pulos, E., & Leng, K. (2010). Evaluation of a voluntary menu-labeling program in full-service restaurants. American Journal Of Public Health, 100(6), 1035-1039.
The authors examined the effect of menu labeling on the ordering pattern in six full-service restaurants. The menu showed calories, fat, sodium, and carbohydrates for all the foods on the menus. There were usually numbers next to the food and a key explaining those numbers someplace else on the menu. Sales data was analyzed for thirty days before and after the intervention. The intervention was combined with a survey.
There was a mean decrease of 15 kcal, 1.5 g fat, 46 mg of sodium per entrée, and an insignificant change in carbohydrates after the intervention. The survey revealed that 71 % noticed the labeling, 54 % understood the labeling system to make healthier choices, but only 20 % were influenced by the labeling.
Roberto, C., Larsen, P., Agnew, H., Baik, J., & Brownell, K. (2010). Evaluating the impact of menu labeling on food choices and intake. American Journal Of Public Health, 100(2), 312-318.
The study examines the effect of menu labeling with calories and on food choices and actual calorie intake in a study setting. Participants were recruited to participate in a market research study; the actual purpose of the study was concealed. They were randomly assigned to three menu groups when they ordered a meal: menu with calorie labeling, a menu with calorie labeling and recommended energy intake, and a menu without calorie labeling. Data was collected for: Calories ordered, calories eaten at the meal, calories eaten during the rest of the day, and the difference between the actual calorie content of the meal and what the participants though they had consumed.
Participants who had menus with calorie information ordered significantly less calories, but it made no difference if they had a menu with recommended intake or without. Those who had menus with calorie information consumed less at the meal than those without the information. If the menu also had recommendation it was even less. Those who were provided with the recommended calorie intake consumed significantly less during the rest of the day, while there was no significant difference for the rest of the day between those without labeling and labeling only with calories.
Studies from Fast Food restaurants:
Elbel, B., Kersh, R., Brescoll, V. L., & Dixon, L. (2009). Calorie labeling and food choices: A first look at the effects on low-income people in New York City. Health Affairs, 28w1110-21.
The authors conducted a study on the effect of the calorie labeling in New York City, after it became mandatory for fast-food restaurants to post these information on the menu boards. 1156 Diners of 14 fast-food restaurants that belonged to the largest chains were asked for their receipts and they were asked survey questions. This was 2 weeks before and 4 weeks after labeling was introduced. Restaurant costumers in Newark served as the control group.
In the survey it could be shown that the number of people who saw the information after it was displayed in New York increased significantly. Almost 28 % reported that labeling influenced their choices and 88 % of those that were influenced stated that they did buy fewer calories. But labeling did not have an influence on actual calories purchased.
Harnack, L., French, S., Oakes, J., Story, M., Jeffery, R., & Rydell, S. (2008). Effects of calorie labeling and value size pricing on fast food meal choices: results from an experimental trial. The International Journal Of Behavioral Nutrition And Physical Activity, 563.
This study examined the impact of four different menu versions on ordering patterns in fast food restaurants. In an experimental setting (conducted in hotel conference rooms and church basements) participants, who ate regularly at fast-food restaurants, were asked to order and consume a fast-food meal. The menus either contained calories labeling with or without value pricing, or no calorie labeling with or without value pricing. Value pricing describes prices for different sizes of menu items, for example French fries. Participants also were surveyed.
Overall the authors found no significant difference among the four different groups. Calories ordered, calories consumed, and menu items were similar. Some significant differences were found when the data was combined with analysis from the survey. Those participants that reported that nutrition is important when they choose fast-food order fewer calories when they are provided with a menu that displays calories. Those that state that price is not important ordered significantly more calories from a menu with calories and prices than from a menu without calories but with prices (948 compared to 598 kcals).
In the survey taste was found to be the most important factor, nutrition was the least important.
Vadivello, M.K., Dixon, L.B., & Elbel, B. (2011) Consumer purchasing patterns in response to calorie labeling legislation in New York City. Int J Behav Nutr Phys Act. 2011;8:51.
The author examined the data from Elbel et al. (2009) to see if the types of foods purchased had changed after mandatory labeling was introduced. They looked at the type of beverages, salad purchases, salad dressing purchases, French fries, cheese on burgers, and number of purchased desserts .They also looked at the reported frequency of fast food consumption.
The results of the study were very mixed. There were no significant positive influences of the labeling in NYC, and even some negative. The number of caloric beverages increased, the number of purchased salads decreased, the use of cheese decreased, and the number of desserts ordered also decreased. They provide a better picture of the shifts that occurred and that overall did not lead to a reduction in calories purchased. There was a significant decrease in fast food consumption frequency among those in NYC that noticed the labeling and reported to use it.