Potential for healthy eating in a Brazilian public university food environment

ABSTRACT Objective To identify the food environment healthy eating potential in the authorized food services on the campus of a Brazilian university. Methods This is an observational study carried out between March and April 2014 on a campus of a public university in Rio de Janeiro, Brazil. The elements of the food environment were evaluated using an audit instrument, containing 86 questions: availability (amount of food services and types of food offered), convenience (days and hours of operation), incentives and barriers to healthy eating (nutrition information, payment strategies and food advertisements). Preparations based on whole grains and fresh or minimally processed products were considered healthy foods; and, preparations with high caloric density, sugar-sweetened beverages, sweets, desserts, cookies, and savory snacks were considered unhealthy food. The questionnaire allowed us to calculate a score for each service, which could range from 0 to 36 points, and indicates the potential for promoting healthy eating in the establishment, with higher scores indicating a greater presence of elements that contribute to healthy eating. The different types of food service facilities were compared based on the scores (p<0.05 for statistical significance). Results Among the establishments evaluated, 24% were snack bars/cafeterias, 26%, restaurants, and 50% offered mixed services. Healthy food items were scarcely available in the establishments (fruits: 24%; vegetables: 20%; brown rice: 15%); while added-sugar beverages (98%), sweets and treats (76%) were widely offered. There was a higher frequency of advertising encouraging consumption of unhealthy items than that aimed at healthy foods (44% vs 30%). In general, the score was 13,2 points (SD=3.3) and the mean score for snack bars/cafeterias (9.3 points) was lower (p<0,05) than that of restaurants and mixed establishments (14,4 points). Conclusion On the campus assessed, the food environment had limited potential for healthy eating, since elements that did not favor healthy food choices were more frequent.


I N T R O D U C T I O N
The food environment is the area of interaction between individuals and the food system, in the acquisition or consumption of food [1].It is made up of physical [accessibility, availability, quality, and promotion], economic [price], sociocultural [norms and behavior] and political [government policies] elements [2].When these spaces are located in institutions such as schools, colleges and hospitals, they are called organizational food environments [3,4].Institutions can act by modifying the organizational food environment through internal policies to regulate food services (authorized or own), in order to encourage the availability of healthy foods and reduce the availability of those considered to be a risk to health.They can also encourage the development of continued actions to promote healthy eating [5,6].
Healthy eating, according to the Dietary Guidelines for the Brazilian Population, is understood as a diet based on the consumption of fresh or minimally processed foods, which contain fats and sugars in small amounts and limited proportions of ultra-processed foods, which are products formulated by the industry, generally, with high levels of sugars, fats, and sodium and high energy density [7].
Studies indicate that the diet quality is related to the food environment, whose characteristics can create barriers or incentives to the adoption of healthy food choices [3,4,8,9].In food environments, the barriers to healthy eating are, mainly, the poor supply of healthy items and the encouragement of the consumption of large portions or the consumption of items with high energy density, while among the incentives for healthy food consumption, the following should be considered: availability of healthy items at affordable prices, nutritional information about foods and preparations, visual highlights of healthy options, availability of reduced portions at low cost and possibility of sharing portions with no change in meal price [9][10][11][12].
The university food environment can be considered a privileged space to promote healthy eating [13,14]; it can significantly influence the eating habits of students, professors, technicians and other professionals, who, for a long period, regularly eat several meals a day at the university.
College students are considered a vulnerable group in the adoption of risky behaviors, including food choices [15][16][17].Young university students have autonomy for food selection decisions, which, however, are constrained by their limited budget and exposure to food environments with reduced potential for the promotion of healthy eating [18].Unfavorable changes in lifestyle are observed in university students since entering college, which can have important health consequences, such as excessive weight gain [14,19,20], which is recognized as an important public health problem in Brazil [21].Studies in Brazil have shown that food consumption of college students is characterized by high intake of foods such as snacks, sweets, and soft drinks and low consumption of fruits, vegetables, fish and grains [16,22,23] and that the dietary patterns of university professors and technicians are characterized by low consumption of fruits, legumes and vegetables, dairy products, water and high meat consumption [24][25][26].
Despite the previous observations, few Brazilian studies have evaluated the food environments in universities [9,11,27,28].In a study carried out in the city of Rio de Janeiro, Franco et al. [27] estimated changes over five years in the availability, price, and advertising of food in authorized food services in a public university and observed unfavorable changes in food availability.Tavares et al. [28] proposed the healthiness index for commercial establishments to be used in the evaluation of the authorized food services in universities.This index is based on the ratio between the proportion of unprocessed or minimally processed foods and ultra-processed foods available in each establishment.For its calculation, an inventory of beverages, food, preparations and convenience items offered for sale in these establishments was carried out.The authors assessed the proposed indicator in two public universities in Rio de Janeiro, having verified its good discriminatory power and its ease of application and interpretation, and the ability to provide a concise diagnosis of commercial establishments on university campuses.
Since the food environment is multidimensional, it is important that its assessment considers these different dimensions [2,3,29].Therefore, the purpose of this study was to identify the potential of the university food environment for healthy eating in the authorized food services on the campus of a public university in Rio de Janeiro.

M E T H O D S
This is an observational study carried out at a public university located in the city of Rio de Janeiro, Brazil, during the period from March to April 2014.Approximately 100 thousand people, including students, professors, technicians, outsourced workers and visitors circulated daily on the campus where the survey took place [30].In 2014, there were 59 authorized food services that operated under a contract and/or bidding modality and sold meals (preparations that are usually consumed at lunch or dinner) and/or snacks (preparations that are usually consumed at breakfast or consumed occasionally between the main meals).Four establishments did not allow the survey to be carried out and one was closed between the period of mapping and data collection, so 54 establishments were assessed.
Data were collected by six trained evaluators, always between 11 am and 2 pm (lunch time).
The raters filled out the questionnaire based on direct observation and questions asked to the person in charge of the service.For a standardized performance of the raters, a careful training was developed based on a field manual, which details are presented in Rodrigues et al. [31].
The following elements of the food environment of each establishment were evaluated: availability (offer of food services; types of food sold), convenience (opening hours consistent with those of the university), incentives and barriers to healthy choices (nutritional information and advertisements encouraging food consumption).
For this evaluation, the Questionário para Avaliação do Ambiente Alimentar em Estabelecimentos que Comercializam Refeições e Lanches Prontos para Consumo (QAA, Questionnaire for Assessment of the Food Environments in Establishments that Market Ready-to-Eat Meals and Snacks) was used [31].The questionnaire includes three sections: service description, assessment of the availability of food/preparations (organized into subsections) and assessment of incentives and barriers to healthy eating, totaling 86 questions.For the last two sections, the questions deal with the presence of specific items and the answer options are yes, no and not applicable, for an item present, absent or when the item is not consistent with the type of establishment being evaluated.
The instrument QAA was considered reliable to assess commercial food services in relation to their potential for healthy eating.The test-retest reliability tested in a previous study was considered high (kappa >0.81), good (kappa >0.61 and <0.80) or moderate (kappa between 0.41 and 0.60) for 89% of the questions.In addition, it demonstrated satisfactory internal consistency with 0.72 Cronbach alpha [31].
A score, taking into account 36 items of the questionnaire, was estimated to assess the potential for healthy eating in the university food environment.The score ranges from 0 to 36 points, with no cut-off point for classifying establishments, with higher scores indicating a greater presence of elements that favor healthy eating and the interpretation of the score must be performed by comparing the estimated values for the different establishments.More details on the methodological bases and on the process of elaboration of the score are found in Rodrigues et al. [31] paper.
The menu or any list that described the beverages and the food items presented on a board, poster, panel, or paper was used to identify the products and preparations sold in the establishments.Whole grain-based preparations (e.g., bread or brown rice and granola) and fresh or minimally processed food (e.g., fruits, vegetables, and beans) were considered healthy foods.In contrast, mixed preparations with high caloric density (e.g.feijoada, piamontese rice), drinks with added sugar, sweets, desserts, treats, cookies and savory snacks were considered unhealthy [7].
The project was approved by the Human Research Ethics Committee of the Hospital Universitário Clementino Fraga Filho, from the Federal University of Rio de Janeiro, Project nº.062/11.
The services were classified according to the products sold, as follows: snack bars/ cafeterias (fast foods, salty and sweet snacks, treats, non-alcoholic beverages, coffee, tea), restaurants (items consumed in meals such as lunch and dinner) and mixed (items relating to snacks and meals).
The absolute and relative frequencies were estimated for the categorical variables and, for the QAA score [continuous variable], the means, standard deviations and minimum and maximum values were estimated according to the type of establishment.The score distribution was asymmetrical (Shapiro-Wilk test p-value=0.043)and the Kruskall-Wallis test was used to assess the differences in the distribution of scores for snack bars/cafeterias, restaurants and mixed establishments.The chi-square test was applied to evaluate the categorical variables analyzed, according to the type of food service.Statistical significance of 5% was considered.For the analyses, the SPSS version 19 (SPSS Inc., Chicago, United States) software was used.

R E S U LT S
Table 1 presents the characterization of the 54 services evaluated, with half of these establishments classified as mixed services and approximately 1/3 operating the night shift.Among the food outlets that operated at night, 58% were snack bars/cafeterias and 42% had mixed services (data not shown).Note: a Snack bar/cafeterias (fast foods, snacks, treats, non-alcoholic beverages, coffee, tea); restaurants (items consumed in meals such as lunch and dinner); and mixed (items relating to snacks and meals).
Table 2 shows the frequencies of healthy and unhealthy items available by type of establishment.In general, items considered healthy were little offered, such as vegetables without the addition of sauce (20%), vegetarian options (9%), brown rice (15%), milk (50%), fruits (24%) and whole grain bread option (48%).On the other hand, drinks with added sugar, sweets or treats, and savory snacks or cheese breads were widely marketed.It should be noted that alcoholic beverages were present in 20% of the establishments, and energy drinks in 9% of them.
Table 3 shows data on incentives and barriers related to the promotion and information release on food and preparations.Among the food outlets assessed, 70% displayed advertisements or signs that promoted both healthy and unhealthy menu items (data not shown).However, there was more encouragement to consume unhealthy than healthy foods (44% vs. 30%) (p<0.01).Incentives to healthy eating such as "drink more juices" messages, photos of fruit juices, fruit juices and salads were more present in mixed (33%) establishments and snack bars/ cafeteria (54%) than in restaurants, where no advertising of healthy items was observed.Incentives to the consumption of foods considered to be a risk to health, such as soft drinks, hamburgers and savory snacks pictures and images were found more frequently in mixed establishments (48%) and in snack bars/cafeterias (69%) than in restaurants (14%) (Table 3).
Likewise, offers that provided price reductions for the combined purchase of fast foods and beverages with added sugar (for example, purchasing a salty snack and a soft drink simultaneously would cost less than the two items purchased separately), were identified more frequently in mixed establishments (37%) and snack bars/cafeterias (77%).In contrast, no establishment charged additional amounts if one or more customers wanted to share a single meal.Restaurants allowed diners to serve their own food more often (43%) than snack bars and mixed-type establishments.
No food outlet exhibited nutritional information on the sold preparations (Table 3).
The mean scores for all establishments and for the establishment categories type are shown in Table 4.The overall average score was 13.2 points (SD=3.3points).The average score of snack bars/ cafeterias (9.3) was lower (p<0.01)than those of mixed services and restaurants which average scores were similar (14.4) (p>0.05;Kruskal-Wallis test).However, the variation in the restaurants scores (minimum and maximum: 12 to 17; SD: 1.5) was smaller than that of the mixed services (minimum and maximum: 6 to 19; SD: 3.1) (Table 4).

D I S C U S S I O N
The campus food environments assessed had several barriers to healthy choices [7], such as lack of nutritional information on preparations, and low availability and few visual incentives of healthy items.In general, the highest score of the QAA score for the establishments assessed was half of the maximum score in the questionnaire, reiterating that the university food environments investigated had reduced potential to encourage healthy eating.In general, the QAA scores attributed to the assessed authorized food services were low, and the lowest average was estimated for snack bars/cafeterias, establishments that provided few opportunities for healthy choices.
When compared to snack bars/cafeteria, restaurants and mixed services had greater availability of vegetables, fruits and whole grains.Roy et al. [13] evaluated seven university campuses in Australia and they also found that establishments with higher average scores offered healthier products such as salads or salad preparations.The global mean score presented in the Australian study reached half of all possible scores.Studies that assessed food environments have recognized that establishments such as snack bars, cafeterias, convenience stores and food vending machines are not very supportive of healthy eating, as they offer many ultra-processed products and few healthy options [9,28].Note: * Kruskal-Wallis test to compare the distribution, according to the type of establishment; Mixed and Restaurants vs. Snack bars/cafeterias: p<0.01. a The QAA score can range from 0 to 36 points.The higher the greater the incentives to healthy eating in the evaluated food environment; b Snack bar/cafeterias (fast snacks, snacks, treats, non-alcoholic beverages, coffee, tea), restaurants (items consumed in meals such as lunch and dinner) and mixed (items relating to snacks and meals).
On the campus where the survey was carried out there are day and evening classes including on Saturdays.However, only one third of the food establishments remained open at night and on Saturdays, most of which were snack bars/cafeterias.Therefore, the campus assessed in our study presented an even lower potential of offering healthy food at night and on Saturdays, impacting even more on the healthy choices of customers who would go there at such time.Access to food services at these times was different from other universities in the same area, which varied from 60% to 93% [9,27].
Food availability, affordability and advertising are key elements of the food environment and influence the diet; thus, these elements may represent incentives or barriers to promoting healthier food choices.As an example, the limited supply and high prices of healthy foods or the expanded supply, advertising and economical advantages of risk foods are barriers to healthy eating [32].Such elements have been described in the university food environment [9,11,14,17,27,[33][34][35][36].Franco et al. [27] identified a weak potential for healthy eating on a campus in Rio de Janeiro, where fresh foods and dishes prepared on site were sparingly offered; there was a lack of nutritional information, prices were higher for healthier options and there was a greater presence of unhealthy food advertisements.
Tam et al. [14] interviewed university campus students in Australia and asked about suggestions for improvements in food services and identified similar elements to those mentioned above.The students suggested an increase in the availability of fresh and unprocessed foods, a greater variety of offered foods and beverages, more affordable prices for healthy products, and access to nutritional information on preparations.
The exposure of nutritional information on commercialized foods and preparations is a strategy that can contribute to the improvement of eating habits of organizational food environments users [14,36].However, both in this and similar studies, this information was not available to consumers [9,27,36].
A study that evaluated the faculty eating patterns on the same campus where this study was carried out pointed out that the patterns identified were related to the food environment of the campus, especially with the predominant type of service, snack bars/cafeterias and mixed establishments.Among the investigated professors, the dietary patterns called "snack" and "fast food" constituted most of the variance in food consumption.Both patterns presented typical foods from snack bars and cafeterias, the first one was composed of items such as bread, butter/ margarine, fruit juice, dairy products and caffeinated beverages; and in the second pattern, cold cuts, savory snacks, soft drinks, sweets and meats were included [26].The low nutritional quality of these patterns corroborates the observations that the campus assessed offered few opportunities for healthy food consumption.
It is important to consider that this study has limitations.The sample consisted of the services available on a given campus, so the data may not be generalized to other university campuses.
In addition, other factors of interest about the organizational environment such as food policies, spaces and time made available to the community that attends the campus to have their meals, water supply, informal market, virtual context, among others [4,37] were not evaluated in this study.
The instrument used in this study evaluates different domains of the food environment, although an assessment of the construct validity has not been made, a limitation of the questionnaire that can be the target of future assessments in order to evidence the discriminatory capacity of the instrument.However, it is important to point out that the results obtained were consistent with similar studies and with the characteristics of the establishments assessed.A strong point of this study is the use of a reliable questionnaire, which showed good internal consistency and adequate test-retest reliability to assess the potential for healthy eating in commercial authorized services [31].This instrument was considered appropriate for evaluating food services in universities, as it does not require establishments to have a structured menu or a list of ingredients used in the preparation, and its main advantage is the fact that it is appropriate for different types of food services (restaurants, snack bars, cafeterias, among others), as is the case of the campus that was assessed.In addition, it is noteworthy that the score was estimated in a way adapted to the type of establishment assessed, since the food items that were not expected in a certain type of establishment were not included in the score (for which the option "not applicable" was ticked).Such items did not imply an advantage or disadvantage for the establishment, not being scored either positively or negatively (for example, the question about the availability of brown rice, a healthy item, was not applied to snack bars/cafeterias and this item was not computed in these establishments).
Our evaluation highlights the paradox of the universities that apply incipiently the precepts that are addressed in teaching and research, especially those in the health area.Health promotion is a recurring subject in courses such as Medicine, Nursing and Nutrition; however, the food environment in universities does not reflect the recommendations indicated by the health area in encouraging healthy eating, as shown by the results presented in this study [38].
Analyses that address the food environment in universities are relevant, given the recent increase in the Brazilian higher education network [39,40].Universities have the potential to contribute to improving the health of individuals who attend their spaces through discussions and development of policies that encourage healthier habits.Hence, the university food environment is considered an opportune space to encourage the purchase and consumption of healthy foods, as interventions mediated by institutional policies seem to be easier to implement [13].These policies could encourage the availability and accessibility of healthy foods on campus, control their prices and create more social spaces equipped with appliances, aiming to increase food choices [41].However, similar to what was found in this study, the authorized food services located within the universities have been characterized by the wide commercialization of beverages with added sugar, sweets, desserts or other treats, quick snacks and scarcity in the supply of fresh or minimally processed products [14,27].
The evaluation of spaces geared to eating is an essential step to direct potential targets for intervention with a view to promoting health in universities.Research has contributed significantly to the literature on university food environments, a topic that is still little explored in low-and middleincome countries [42].In addition, the instrument and the score used allowed a more comprehensive assessment since the instrument considered three different dimensions of the food environment for the classification of healthy eating potentials.
The study points out the need for changes to improve the incentive for healthy consumption in the food services of the university campus assessed, especially in commercial authorized restaurants.The university restaurant policy proves to be very effective in improving students' eating habits, as it makes healthy food available and accessible [11].However, institutional policies that rule the authorized food services are necessary and should include, limitation or prohibition of unhealthy products marketed within these organizations, requirement for training of the food handling teams, incentives for the adoption of more affordable prices, among others.
CB RODRIGUES et al. | UNIVERSITY FOOD ENVIRONMENT

Table 2 -
Availability of healthy and unhealthy items in authorized food services, according to the type of establishment.Rio de Janeiro (RJ), Brazil, 2014.
Note: a Snack bar/cafeterias (fast foods, snacks, treats, non-alcoholic beverages, coffee, tea); restaurants (items consumed in meals such as lunch and dinner); and mixed (items relating to snacks and meals); b Chi-square test.NA: Not Appplicable.

Table 3 -
Incentives and barriers for healthy eating promotion in authorized food services, according to the type of establishment, Rio de Janeiro (RJ), Brazil, 2014.Snack bar/cafeterias (fast foods, snacks, treats, non-alcoholic beverages, coffee, tea), restaurants (items consumed in meals such as lunch and dinner) and mixed (items relating to snacks and meals); b Chi-square test.

Table 4 -
Score a means (standard deviations) of the Questionnaire for Assessment of the Food Environment of the authorized food services on a university campus (n=54).Rio de Janeiro (RJ), Brazil, 2014.