TABLE 2.
Overview of characteristics and key findings of food interventions considering human health in postsecondary campus settings1
| Authors | Location | Setting | Sample size | Intervention | Duration | Study design | Outcomes and measures | Summary of findings | Funding |
|---|---|---|---|---|---|---|---|---|---|
| Aboul-Enein and Bernstein (73) | Houston, TX, USA | Online | Students; enrolled in undergraduate online nutrition course (n = 65) | Online nutrition course about the Mediterranean diet (incorporated assigned readings, article reviews, recipe analysis, weekly discussion forums, asynchronous presentations, writing assignments, exams) | 1 semester (∼4 mo) | Pretest, posttest experiment (no control) | Survey: perceived quality and value of Mediterranean diet (Mediterranean Diet Quality Index—16-item questionnaire index; score range 0–12, where ≤3 is low, 4–7 is medium, and ≥8 is high) | Compared with scores before the intervention, participants viewed the Mediterranean dietary pattern more positively and demonstrated an increase in understanding of the pattern | Not stated |
| Bernardo et al. (74) | Brazil | University campus | Students (n = 82; 41 in intervention and 41 in control) | Cooking class intervention (5 weekly 3-h hands-on cooking sessions) and food selection and purchasing workshop. Sessions included cooking demonstrations, discussions about nutrition, hands-on food preparation, and group meals. Session was conducted in groups of 10–12 students, divided into 2–3 students/bench. | 6 wk | Randomized control trial with follow-up at 6 mo | Sample characteristics: demographics, cooking characteristics (e.g., self-reported knowledge, cooking experience, and location). Outcomes related to cooking skills and healthy eating practices using a questionnaire tested for validity: 1) accessibility and availability of fruits and vegetables at home; 2) cooking attitudes; 3) cooking behaviors at home; 4) cooking behaviors away from home; 5) produce consumption self-efficacy; 6) self-efficacy for using basic cooking techniques; 7) self-efficacy for using fruits, vegetables, and seasonings (while cooking); and 8) knowledge of cooking terms and techniques. | Survey responses demonstrated an increase in cooking confidence and self-reported knowledge of cooking terms after the intervention. Participants increased the availability of fruits and vegetables at home after the intervention. Effects were sustained at the 6-mo follow-up | Federal Agency for Support and Evaluation of Graduate Education in Brazil; Human Resources Development Program; National Council for Scientific and Technological Development |
| Bevet et al. (66) | Burlington, VT, USA | Cafeteria (buffet-style) | Students (n = 681 in pretest; n = 128 late-at-night) | Choice architecture 1: vegetable-heavy entrées added to beginning of self-serve line.Choice architecture 2: healthy choices snack bar added to salad bar (e.g., hummus, popcorn, trail mix, yogurt, precut fruit). | ∼3 wk | Pretest, posttest quasi-experiment (no control) | Pre-survey (to identify preferences for food options among diners at different eating occasions): preference for foods, sociodemographics. At-late-night survey: health score, food choices, satisfaction. Tallied number of students who took food from each intervention station. | Between 54% and 69% of students incorporated a vegetable-heavy entrée into their dining session throughout the week. Significant increase in purchases from the healthy choices snack bar. | University of Vermont Dean of Students Office |
| Biden et al. (70) | London, Ontario, Canada | Residence cafeterias | Sales from cafeterias. Students were surveyed to provide context to findings (n = 1476). | Point-of-purchase labels using a checkmark indicator based on multinutrient “FRESH” criteria. Those not meeting criteria were classified as non–FRESH approved. FRESH Reward Card program provided students with 1 free fruit/milk after 9 purchases of whole fruit/milk. | ∼2 y for intervention period | Pretest, posttest ecological study | Sales outcomes: number of items sold, number of fruit and milk items sold, net sales, number of free fruit or milk items redeemed, annual cost of free fruit items redeemed calculated by taking the mean cost of the fruits with the highest sales quantities. | Healthy items were sold more often and made less expensive by the food operating service. Healthy items also appeared more frequently on menus than less healthy items | Authors had no funding to report |
| Cárdenas et al. (71) | Lima, Peru | Cafeteria | Sales from cafeterias (∼150 students/d). Students for semistructured interviews (n = 12) | Repositioned fruit next to point of purchase with a sign that said, “Consuming five fruits and vegetables per day prevents many illnesses - World Health Organization” with price per item of fruit. Poster was also placed at entrance of cafeteria.Follow-up intervention where price of fruit was reduced by 33%. | 13 wk; 3 wk per intervention period | Pretest, posttest experiment with 2 intervention phases (no control) | Outcomes (collected by salesperson): number of pieces of fruit sold each day; number of full meals sold; fruit ratio of total fruit purchased to total meals sold in the same day. Visible information about each consumer was also collected (e.g., sex, student vs. nonstudent adults). Qualitative interviews: perceptions of marketing strategies of fruit, demographic questions, reasons for purchasing or not purchasing fruit, noticed reduction in price. | Fruit purchasing significantly doubled from phase 1 (repositioning of fruits) to phase 3 (added 33% reduction). There were no differences in fruit sold between the other phases. | National Heart, Lung, and Blood Institute; NIH; US Department of Health and Human Services; Fogarty International Center |
| Christoph and Ellison (60) | Midwestern USA | Dining hall | Students (n = 1069) | Nutrition label items (with dish title, serving size, number of calories, and grams of fat, carbohydrates, and protein) | ∼16 wk; data collected at weeks 8 and 12 | Cross-sectional | Survey: gender, age, college classification, height, weight, self-perceived eating habits, exercise frequency, enrollment in college-level nutrition courses, and nutrition label use. Dietary intake coding (using pre- and postmeal photographs): coded selection, type, servings, and consumption according to the US Department of Agriculture's MyPlate food categories. | When compared with non–label users, a greater proportion of label users selected fruits, vegetables, and beans. Fewer label users selected fried foods, foods with added sugars, potatoes, and refined grains than nonusers. | US Department of Agriculture; National Institute of Food and Agriculture |
| Christoph et al. (59) | Midwestern USA | Dining hall | Students (n = 2729) | Nutrition label items (with dish title, serving size, number of calories, and grams of fat, carbohydrates, and protein) placed directly in front of food | ∼16 wk; data collected at weeks 8 and 12 | Cross-sectional (3 waves) | Survey: nutrition label provision, awareness, and use; behavior characteristics (sleep, exercise, food habits); and specific topics (e.g., what types of information the diner would prefer to see, reasons for label nonuse, frequency of dining services mobile application). Sociodemographic variables (age, sex, year in university) and anthropometric variables (height and weight, weight intention) were collected for correlates. | Nutrition label placement did not influence nutrition label awareness. | US Department of Agriculture; National Institute of Food and Agriculture |
| Cioffi et al. (61) | Ithaca, NY, USA | Dining hall | Sales from dining halls | Nutrition labeling on prepackaged food purchases | ∼2 y; 1 y for pre- and posttest periods, respectively | Pretest, posttest experiment (no control) | Mean calories purchased per week, sales patterns, change in high- vs. low-calorie and high- vs. low-fat foods. | Food labels led to a 7% reduction of mean total calories purchased per week. Total fat for purchased foods also demonstrated a 7% reduction among labeled foods. | Division of Nutritional Sciences, Cornell University |
| Deliens et al. (69) | Brussels, Belgium | On-campus restaurant | Students for French fries experiment (n = 2930 during control week; 2344 during week 1 and 2325 during week 2). Students for fruit experiment (n = 3235 during control week; 3802 during week 1 and 3728 during week 2). Students for interviews (n = 230 for French fry experiment; n = 227 for fruit experiment) | Price increases on French fries by 10% and 20%.Price reductions on fruit purchases by 10% and 20%.Posters and information boards to communicate the price adjustments | 5 wk; 1 wk per intervention period | Pretest, posttest quasi-experimental study and qualitative interviews | Data collected at register: student status, chosen menus, whether French fries or fruit was chosen, sex of the student. Interviews: demographics (gender, age, residency, study discipline, height, weight), food choice and whether price manipulation influenced their choice, asked if price adjustment would change their consumption in the long run, asked if they believed it was a good initiative to support healthy eating. | Increasing the price of French fries by 10% and 20% led to a 10.9% and 21.8% reduction in purchases, respectively. Decreasing the price of fruit by 10% and 20% led to a 25.1% and 42.2% increase in fruit purchases, respectively. | Authors had no funding to report |
| Dingman et al. (75) | North Carolina, USA | Vending machines in residence halls | Vending machines (sales data for n = 18 machines; 9 intervention and 9 control sites). Students (n = 670). | Poster boards adjacent to each vending machine, listing Nutrition Facts Panel for each product in the vending machine. Five products that met nutrition criteria were highlighted (<200 kcal, ≤2 g saturated fat, 0 g trans fat, ≤7 g sugar, 300 mg Na) and labeled as “Better Choice” snacks with an interpretive label. Criteria and promotional messages were described on posters and emails. | 8 wk; 4 wk for pre- and posttest periods, respectively | Pretest, posttest, 2 × 2 experimental design | Sales data: the mean calories sold per snack, and the proportion of snacks that contained fewer calories and less saturated fat, sugar, and sodium than the usual snacks (i.e., Better Choice snacks). Used sex and year of schooling as covariates in models. | Compared with the 4-wk pretest period, there were no significant differences in calories sold. | Not stated |
| Hammond et al. (62) | Ontario, Canada | Residence cafeterias | Students (n = 159) | Numeric calorie labels | 2 wk; 1 wk for pre- and posttest periods, respectively | Pretest, posttest naturalistic cohort | Demographic information: sex, race, self-reported health status, BMI, weight perceptions and aspirations. Exit surveys: noticing information, what types of information, when they saw it, use of information, calories ordered and consumed. | Compared with the pretest period, calorie labels led to significant increases in noticing and use of nutrition information to guide food purchases. The calorie content of foods purchased and estimated amount of calories consumed decreased after the intervention. | Canadian Cancer Society Research Institute; Canadian Institutes of Health Research |
| Hua et al. (76) | New Haven, CT, USA | Vending machines | Vending machines (n = 56) | Improved availability of healthier snacks (using “FitPick” criteria of ≤250 kcal, ≤20 g sugar, ≤230 mg Na, ≤10 g fat, ≤3 g saturated fat, and no trans fat). Water and other beverages (≤25 kcal/237 mL) were classified as healthy. Reduced price for healthy items by 25%. Water reduced price to $1.3. Promotional signage to indicate price change and labels to indicate healthy items. Signs promoted water consumption. | 5 mo | 2 × 2 × 2 balanced factorial experiment | Sales and revenue data: total number of units sold and total revenue, stratified by type of vending machine (i.e., snacks or beverages). | The interaction between improving the availability of healthier options and promotional signage increased revenue of healthier snacks. Price reductions alone did not affect consumer choice. The interaction of all 3 interventions did not increase purchasing of healthier snacks. Compared with the pretest period, there was an overall increase in healthier snack purchasing across all intervention machines. | None to report |
| Mistura et al. (77) | Victoria, British Columbia, Canada | Residence cafeterias | Students for surveys (n = 340) | Improved availability and appearance of vegetables. Poster at eye level to indicate vegetable option and with a character/message. | ∼11 wk; 3 wk per intervention phase | Pretest, posttest quasi-experiment (no control) | Tallied number of students observed purchasing a vegetable option compared with the total count of students that purchased from the hot table. Counts were recorded by sex. | Compared with the 2-wk pretest period, there was no change in the mean number of vegetables purchased. | British Columbia Ministry of Health |
| Mora-García et al. (78) | Bogota, Colombia | Cafeteria | Consumers (n = 228 for control; n = 257 for intervention) | Randomly informed people about the existence of the front-of-package Nutri-Score labeling system. | 4 wk | Pretest, posttest randomized controlled trial | Receipts: products bought, price of product. Survey: anthropometric and demographic characteristics, using the system, physical activity measurements. | The intervention led to an increase in total expenditure of $0.18. Spending on healthier items was 21% or $0.26 higher than purchases made during the pretest period, with no change for less healthy items. Compared with those in the control, customers were 10% more likely to buy healthier items. | Apoyo de Proyectos Interdisciplinarios de Investigación |
| O'Brien and Palfai (79) | Northeastern USA | Online | Students (n = 148; n = 49 for control group, n = 50 in web-only group; n = 49 in web and mobile group) | Online nutrition course (on-screen, open-response prompts, writing task, personalized feedback on healthy eating) only. Online nutrition course intervention and daily messages for behavior planning. | 4 wk | Pretest, posttest randomized controlled trial | Dietary habits survey: vegetable and fruit consumption (three 7-d recall items). Healthy food choices: how often students selected university-sponsored healthy items (Sargent Choice selection). | Compared with the control group and the group that received the online nutrition course, those who received the online course and daily messaging demonstrated a significant increase in the likelihood of improving vegetable consumption and were 3 times more likely to meet dietary guideline standards. However, the intervention did not significantly change fruit consumption. | Not stated |
| Policastro et al. (63) | NJ, USA | Dining hall (take-out line) | Students (n = 9765) | Modified sandwich order form where healthier ingredients were listed first within each category, printed in bold and larger font, and designated with a star symbol. | 8 d of data collection over 8-wk study period | Pretest, posttest experiment (no control) | Participants’ choice of ingredients and measures for calories, fat, sodium, and fiber for each order. | When compared with orders using the unmodified sandwich order form, the modified form led to an increase in selection of healthier ingredients and decreased selection of less healthy ingredients. | None to report |
| Rodgers et al. (80) | Northeastern USA | Online | Students (n = 43) | Online intervention which included taking photographs of every meal and text messages to encourage healthy eating (3 times/d at mealtimes). | 10 wk; 3 wk for intervention period | Pretest, posttest experiment (no control) | Demographic survey: age, year in school, housing status, weight and height using scale to calculate BMI. Food Frequency Questionnaire: fruit and vegetable intake. Beverage Intake Questionnaire: caloric intake from sugar-sweetened beverages. Drive for Thinness (subscale from Eating Disorder Inventory): 7 items, such as preoccupation with thinness. | There was a significant increase in fruit and vegetable consumption among those in a higher BMI category. There were no significant changes in vegetable consumption and a decrease in fruit consumption was observed among those in a lower BMI category. | None to report |
| Rose et al. (58) | Columbus, OH, USA | Vending machines (campus dorms) | Students (n = 128) | Vending machines with milk installed in high-traffic areas and near other vending machines. | 2 mo | Pretest, posttest experiment (no control) | Healthy Habits Survey: vending purchasing habits, motivators and barriers to consuming calcium-rich foods. Calcium consumption questionnaire: assessed calcium and milk intake from foods and beverages. | No changes in milk and calcium intake. | American Dairy Association-Mideast |
| Roy et al. (81) | Australia | Food outlet | Students (n = 713) | Calorie labeling placed adjacent to items on menus and contextual statement. Social marketing campaign with comprehensive website/interactive calculators, advertising slides for digital screens, and banners. Dietitians were also available during lunch hours and stood next to banners. | 10 wk; 5 wk/intervention | Pretest, posttest experiment (no control) | Sales data: change in itemized food sales data (also compared with the same weeks during the same period the year prior). Intercept interviews: customer attitudes, awareness of knowledge, and use. | There were no differences in sales between the calorie labeling with social marketing campaign and calorie labeling alone. Of participants, 30% were aware of the calorie labels on the menus and 75% were accepting of the labeling intervention after they were made aware of the changes. Participants selected meals with lower mean calories after the social marketing intervention, despite most participants claiming that the campaign would not influence their food and beverage purchases. | None to report; corresponding author received Australian Government, Department of Education support |
| Schroeter et al. (82) | USA | University campus | Students (n = 57; n = 18 for control group; n = 8 for incentive and education group; n = 31 for education group) | Four-week educational session on 5 food groups outlined in US Department of Agriculture's MyPlate, including 1-h lessons with 3–4 quiz questions, interactive style, and with tailored personalized suggestions at the end of the lessons. Education and incentive if Healthy Eating Index score was improved by 5%. | 6 wk; 4 wk for intervention | Pretest, posttest randomized controlled trial | Survey: demographics, media behavior (self-rated), nutrition knowledge (self-rated), health behavior, and dietary recalls using the Automated Self-Administered 24-Hour Dietary Assessment Tool. | Dietary quality scores increased by 15% among those who received either intervention, whereas the control group increased by 8%. | Cornell Center for Behavioral Economics in Child Nutrition; California Polytechnic State University College of Agriculture, Food, and Environmental Sciences |
| Scourboutakos et al. (83) | Toronto, Ontario, Canada | Residence cafeteria (buffet style) | Students (n = 368–510) | Education campaigns for beverages and fruits and vegetables (e.g., healthy eating plate infographic). Physical activity calorie equivalent labeling for sugar-sweetened beverages. | 5 mo; 2 mo for pretest and 3 mo for intervention period | Pretest, posttest quasi-experiment (no control) | Beverage choices and fruit and vegetable choices. | The proportion of students who purchased fruit significantly increased from 30% to 36%. The proportion of students who purchased sugar-sweetened beverages significantly decreased from 49% to 41%. | Cancer Care Ontario; Canadian Institutes of Health Research; University of Toronto |
| Seward et al. (72) | Cambridge, MA, USA | Residence cafeterias | Students (n = 1329) | Traffic light labeling and healthy-plate stickers. Healthy foods and beverages moved to high-traffic and accessible areas. | 13 wk; 6 wk for pretest and 7 wk for intervention period | Pretest, posttest quasi-experiment (no control) | Cafeteria servings: change in proportions of red, yellow, and green items (overall and by subgroups of food categories). Online surveys: how and whether students used the nutrition information to guide choices and whether they wanted to have nutritional labels/what information should be on those labels. | Of postintervention students who noticed the traffic-light labels, 59% thought they were helpful and 73% said that they should continue to be used after the study. No significant changes in proportions of red-, amber-, and green-labeled items sold. | Harvard College Museum of Comparative Zoology; National Heart, Lung, and Blood Institute |
| Shi et al. (64) | Australia | University campus | Vending machines (n = 60 across 7 campuses; n = 27 snack machines, n = 33 beverage machines, n = 11 with both) | Health Star Rating (voluntary front-of-package labeling system; ranging from half a star to 5 stars where more stars indicate a healthier choice). | Not applicable | Cross-sectional | Audits: name, weight, volume of items; promotions. Nutritional quality: energy, positive nutrients, risk nutrients. | Compared with preimplementation data ∼3 y prior, there was an increase in the proportion of snacks and beverages rated 3.5 stars or higher offered in the vending machines. | None to report |
| Sogari et al. (65) | USA | Dining hall | Consumers (n = 3734) | Messages about the nutritional benefits for whole-grain pastas. | 18 d over 9-wk study period | Pretest, posttest quasi-experiment | Number of diners who selected whole-grain pasta vs. other types of pasta. | Compared with the nonmessage condition, messages about vitamin benefits demonstrated a 7.4% increase in the probability of selecting whole-grain pastas. | European Union's Horizon 2020 Research and Innovation Programme |
| Tallant (84) | USA | University campus | Students (n = 33) | Nutrition education course focused on self-efficacy, nutrition behavior skill-building, and nutrients and lifestyle behaviors. Lecture on food labels was also provided. | 16 wk | Pretest, posttest quasi-experiment | Survey: food label use; healthy food choices; modified survey instrument from the United States Department of Agriculture Diet and Health Knowledge survey measuring perceptions on diet adequacy, nutrition knowledge, nutrition self-efficacy, and perceptions for a peer nutrition education component. Dietary self-assessment.Scores for food label reading. | Compared with the pretest period, the intervention led to a significant increase in food label–reading behavior and food choice behavior. Of students, 27% practiced food label reading more frequently at the posttest period and 29% indicated choosing healthier options more frequently. | Not stated |
| Valdez et al. (85) | Tuscan, AZ, USA | University campus | Students (n = 10) | Experiential learning course: value of whole foods plant-based diet (and nutrition) for chronic conditions, provided meals. | 10 d | Pretest, posttest quasi-experiment using mixed methods | Qualitative questionnaire: perspectives on whole foods plant-based diets. Blood samples for cholesterol and glucose. | The intervention improved participant understanding of how plant-based diets can affect chronic disease. Compared with the pretest period, there was an observed mean change in total cholesterol of −26 mg/dL, HDL cholesterol of −6.1 mg/dL, and LDL cholesterol of −21.6 mg/dL. | Authors stated that they did not receive financial support; however, the article stated that grant funding was secured to cover the costs of meals |
| Vermote et al. (67) | Brussels, Belgium | On-campus restaurant | Students (n = 2056 at pretest; n = 2175 at posttest). Students and employees for interviews and dietary recall (n = 296). | Reduced portion sizes for French fries (∼20%) | 1 wk | Pretest, posttest quasi-experiment | Consumption and plate waste. Interviews: satiety and caloric intake; food recall for meal, side, dessert, and beverage; sociodemographics. Noticing of portion sizes and estimates for how much and if portion size was sufficient. | Reduced portion size was effective in reducing consumption and plate waste, with no differences in satiety and caloric intake. Mixed consumer perceptions regarding portion size changes. | Not stated |
| Vermote et al. (86) | Brussels, Belgium | On-campus restaurant | Students (n = 556) | Food labeling (interpretive) at point of purchase and information campaign (posters). | 35 wk; interventions were phased in ranging from 1 to 23 wk in duration | Pretest, posttest quasi-experiment using mixed methods | Sales data: amount sold for desserts. Interviews: demographics, visibility of intervention, influence and use, feasibility for long-term use. | Compared with the pretest period, there was a significant increase in fruit purchasing. The effect was sustained over a 35-wk study period. | Not stated |
| Viana et al. (87) | Los Angeles, CA, USA | Vending machines | Vending machines (n = 3). Consumers (n = 100). | Information campaign using Healthy Campus Initiative label (included web address for more nutritional information). Interpretive labeling for healthy products. Healthy products were grouped at eye level. | 2 mo | Pretest, posttest quasi-experiment | Sales reports: revenue, profit, number of total products sold, and healthier products sold. Survey: demographics, typical frequency of vending machine purchasing. | Compared with the pretest period, intervention machines demonstrated an increase in profits of healthier items. Revenue was not compromised after the intervention. | Not stated |
| Walmsley et al. (68) | Coventry, England | On-campus grocery store | Not applicable | Moving fruits and vegetables from the back of the store to the front and moving beverages. | 5 y; 2 y per intervention period | Pretest, posttest experiment (no control) | Sales data: daily total quantity sold for each product, description, price, profit, barcode, unit size, and food category. | The rearrangement of the shop led to an increase in the percentage of total fruit and vegetable sales. There was a decline in sales of fruits and vegetables over a 5-y period. | None to report; corresponding authors supported by University of Warwick; National Institute of Health Research |
1
n = 29 interventions across 30 records.