Papass clinical trial protocol: a multi-component school-based intervention study to increase acceptance and adherence to school feeding

This is a multi-component school-based, clinical trial called PAPASS, which means “Professionals and Students for Healthy Eating in Sumidouro”. PAPASS is a study aimed at increasing adherence and acceptance of school meals, which will be conducted in the 2019 school year in 3 public schools in the city of Sumidouro, Rio de Janeiro, Brazil. All students from 4th to 9th grade, morning and/or afternoon shifts from the 3 selected schools will be invited to participate in the study.

The study design will be parallel, with 3 arms: (1) Control group (without intervention); (2) Intervention group 1 (changes in the school environment) and (3) Intervention group 2 (changes in the menu and school environment) (Fig.1). To allocate interventions, schools will be numbered and randomly allocated to each group using opaque envelopes by people who are not involved in PAPASS.

This study was approved by the Ethics Committee of the Institute of Social Medicine (State University of Rio de Janeiro), under opinion number 2.825.207 and registration number in the CAAE Brazil Platform: 92307118.7.0000.5260 and registered in the Brazilian Registry of Clinical Trials (ReBEC) for randomized controlled trials under code RBR-7mf794 (http://​www.​ensaiosclinicos.​gov.​br). The SPIRIT recommendations were followed to describe the study protocol [27]. The complete checklist can be reviewed in Additional file 1. The signed Informed Consent Form will be obtained from all participants’ guardians and can be reviewed in Additional file 2.

The sample size of the PAPASS study was calculated based on the results of the last edition of National School Health Survey, conducted in 2015, which revealed that only 37.21% of Brazilian students, from 6th to 9th grade, from public schools, morning or afternoon shifts, that reported the availability of school meals, consume them at least three times a week [10, 28]. Considering a 50% increase in adherence to school meals, 80% power and a significance level of 5%, a sample size of 109.4 adolescents per arm was estimated, totaling a sample of 328.2 students. Considering losses and the rate of refusal, the total sample size required was increased by 20%, raising to approximately 132 students in each arm and a total of 396 students in the sample.

This study will be conducted in the municipality of Sumidouro, which has an estimated population of 14,920 inhabitants, is located 174 km from the city of Rio de Janeiro, state capital, and is part of the Rio de Janeiro State Center Fluminense mesoregion [29]. The municipality has a Municipal Human Development Index (MHDI) around 0.611, the worst from the state, far from state capital, which has a MHDI of 0.799, ranking first in the state [30].

For this study, three of the four Sumidouro districts were included and, from the 23 municipal public schools, with classes from 4th to 9th grade, 3 of them were selected to participate in the study. All students regularly enrolled in these schools were considered eligible for this study (Fig.1). Exclusion criteria was students with physical/cognitive disabilities and pregnant adolescents.

Data on the characteristics of the study population (age, gender and skin color/ race), anthropometric measures, socioeconomic data, physical activity practice, sedentary habit, breakfast habit, food intake and adherence to school feeding will be collected in three moments (baseline - March, half of the school year – August, and end of the school year - November), using a structured questionnaire applied by trained researchers. Acceptability tests will be performed bimonthly, in order to include all menus established in the schools participating in the study.

Adherence will be assessed through a specific question in the questionnaire directed to the frequency of school meals consumptionin the week, with the options: never, once a week, twice a week, three times a week, four times a week or all days. Students who consume the food offered 3 or more times a week will be considered as adhering to the school feeding program [9]. Each adolescent will be assessed in 3 moments: at baseline - at the beginning of the school year; the second - at half of the school year; and the third - at the end of the school year. In order to avoid follow-up losses, schools will be visited 4 times at each moment of data collection (Table 1).

To assess the level of dishes acceptance, the acceptability test will be applied for all dishes that make up the school feeding menus, as proposed by the Manual for Application of the Acceptability Test, published by the Ministry of Education [12]. The Department of School Food of Sumidouro establishes that each menu must be worked for a period of 2 months. The first set up menu (valid for February and March) will be the same as for all schools in the municipal school system. The other menus effective for the months of April–May, June–July, August–September and October–November will be differentiated among the schools that are part of the study. Therefore, the acceptability tests will be applied in March, May, July, September and November 2019. The schools allocated to the control group and intervention group 1 will continue to receive the same menus made available to the other schools in the municipality, while the school allocated to intervention group 2 will receive the new menus prepared by the research group.

The acceptability test will be applied to students who access the school cafeteria and consume the food offered. The test will be performed on cards with hedonic scale, a validated instrument which consists in the evaluation of a product - in this case, the menu - from a graded scale, with points representing the student’s acceptance of food, differentiated based on the grades (4th and 5th grade - mixed facial hedonic scale; 6th to 9th grade - verbal hedonic scale) [31, 32]. In these assessments, the student may select the expressions “I loved”, “liked”, “indifferent”, “did not like” or “hated”. When it comes to the mixed facial hedonic scale, besides these expressions, they have pictures of feelings faces. Answers will be counted for each scale expression shown on the card. If the sample has a percentage greater than or equal to 85% in the expressions “I loved it” and “I liked it”, it will be considered that the dish/food produced good acceptance. Acceptability tests will be performed in all 3 schools participating in the study, being on the 4th week of implementation of each new menu in the school of intervention group 2 and in the same period in the schools of control and intervention groups 1 (Table 1).

The rating of the socioeconomic level will be performed by applying the Brazil 2018 Economic Classification Criterion, with the questions included in the questionnaire, that refers to the amount of assets that the family owns (automobile(s), personal computers(s), dishwasher(s), refrigerator(s), freezer(s), washing machine(s), DVD player(s), microwave oven(s), motorcycle(s) and clothes dryer(s)), domestic servant(s), bathroom(s) and if have a Public Utility Services (piped water and paved street) [33].

Age will be calculated from the difference between the date of questionnaire application and the student’s date of birth.

Weight and height will be measured using standard procedures [34]. Portable weight scales (Tanita®BC-558) will be used for weight measurements. Height will be measured using a portable stadiometer (Alturexata®), with a width of 200 cm and a variation of 0.1 cm. Two measurements will be taken and a maximum variation of 0.5 cm between them will be allowed. If the variation exceeds this value, the measurements will be repeated. The average of the two valid measures will be considered for the analysis. The classification of the nutritional status of the students will be performed according to the classification of the World Health Organization [35].

To assess the physical activity level, we used questions included in the 2012 National School Health Survey (PeNSE), which refer to the frequency and time spent in three different activities (commuting to school, physical education classes at school and extra-school physical activities), in relation to the week prior to the interview, to estimate the accumulated physical activity in the period. Adolescentss will be classified as active (≥300 min per week), insufficiently active (150–299 and 1–149 min per week) and inactive (without physical activity) [9].

Food consumption will be assessed by a short food frequency questionnaire (FFQ) (23 items), which consists of a reduced version of a validated FFQ for adolescents from Rio de Janeiro [36], with the following options: less than once a month or never; 1–3 times a month; Once a week; 2–4 times a week; 5–6 times a week; once a day; 2 or more times a day.

Sedentary habits will be assessed in two separated questions based on the number of hours on a typical weekday (Monday to Friday) spent watching television and, using a computer or playing video games, which showed the following options: < 2 h; 2–4 h; 4–6 h, 6–8 h; > 8 h. The sedentary behavior indicators adopted are based on common PeNSE 2012 questions and were based on the number of daily hours allocated to these activities [9]. The habit of having breakfast will be assessed from a question taken from PeNSE 2012, which refers to the frequency in which the student usually consumes the first meal of the day, with the following options: never or almost never; once or twice a week; 3 or 4 times a week; 5 or 6 times a week; every day [9].

In order to promote retention of study participants, telephone calls will be made with the school board requesting that students be aware of the date when data collection will happen, minimizing the number of absentees. Furthermore, meetings will be held with students’ parents and staff from each of the participating schools to raise awareness of the importance of participating and authorizing students participation in data collection. To maximize the data collected, each school will receive 3 visits, on consecutive days, to get information from students who, for some reason, may have been absent on the previous data collection.

The interventions will take place during the 2019 school year, beginning in April, in the two schools allocated to the intervention groups and will combine two strategies: modifications in the school environment and modifications in the menu offered by the school. Interventions in the environment will be based on the choices architecture principles proposed by Thaler & Sustein [23, 37, 38], and the interventions in the menu will be based on the factors that contribute to poor adherence and acceptance of school feeding, identified through a previously conducted focus groups with the lunch ladies (school food service employees) of the all 3 participating schools and the students of the school that will receive the two interventions (intervention group 2).

The focus groups were conducted in the second semester of 2018, lasting 2 days. On the first day, it was held with the students, and on the second day, with the lunch ladies. The focus group with the students was conducted by 1 mediator and 2 observers with number of participants ranging from 8 to 10 people [39‐43]. It was guided by questions that addressed the perception about the frequency of eating school meals; the most accepted foods; the least accepted foods; the environment where the meal is consumed; the way of preparation and distribution service; as well as prior knowledge of the dishes that would be served during the week. The focus group elaborated for the lunch ladies was conducted by 1 mediator and 1 observer with number of participants ranging from 8 to 10 people [39‐43]. It was contemplated with questions about: self-knowledge about the role of lunch ladies in the school; the perception of adherence and acceptance of food by the students; motivation of students for not consuming dishes; form of preparation and distribution of meals; and what actions could be taken to make school feeding more attractive to students. The activity lasted approximately 1 h and 20 min, distributed according to the complexity of each topic to be discussed. After transcribing the participants’ speeches, content analysis was performed by trained researchers in three stages, which enabled the identification of some factors responsible for not consuming school meals [44]. In face of the reports observed during the focus groups with the lunch ladies and students that allowed the identification of dishes with less acceptance among the students, it was observed that the lack of autonomy in the composition of the dish by the students generated dissatisfaction, as well as a waste of the prepared foods due to the standard portioning service performed by the lunch ladies.

Intervention group 1 will receive changes in the school environment. Modifications in the environment, based on the choices architecture, which will consist of making posters that will be fixed to the cafeteria walls, and displays, with images and messages that encourage students to make healthier food choices, that will be placed on the tables; use of tablecloths on cafeteria tables with color images of fresh foods; accommodation of the fruits served in the school meals in a prominent position, improving their visibility in an attractive way; creation of a fruit stand (common in street markets in Brazil); placement of a banner with the menu of daily dishes at the entrance of the cafeteria; application of footprints on the floor that playfully direct adolescents to the cafeteria. Changes in the meals distribution system will be made through the installation of a thermal distribution counter, which will guarantee the autonomy of the students’ choice of dishes, through a self-service system.

. Intervention group 2 will receive the same interventions as intervention group 1, with the addition of the modification in the dishes that make up the menus offered to the students, based on the reports observed during the focus groups.

The new dishes and the menus were developed by gastronomy researchers and by the nutritionist, responsible technician for the PNAE in Sumidouro, with common foodstuffs for all municipal schools in Sumidouro, acquired by bidding process. The new dishes were intended to be more palatable, favoring the addition of fruits and vegetables in the menu, as well as reducing the use of ultra-processed foods, sugar and oil. Local eating habits were respected and all requirements proposed by PNAE resolutions for the creation of school feeding menus were met [13]. It is important to point out that the new dishes suggested by the PAPASS project were previously approved by the School Food Department of Sumidouro.

A Practical Guide was developed for the lunch ladies, composed by notions of hygiene, storage and preparation techniques of the recipes that will be served in the 2019 school year. Also, sixty-one technical cards composed of quantity of ingredients, preparation method, necessary utensils, yield and portion for each dish were prepared and included in the guide. The dishes were previously tested in an experimental kitchen to correct the necessary quantities of inputs and to check the utensils that would be used, as well as the improvement of the techniques needed to prepare the new recipes. The cookery workshops with the lunch ladies were elaborated from their previous report, through qualitative research conducted from a focus group, aiming to meet the demands for new preparation techniques, especially among the less consumed foods by the students. They were conducted by trained gastronomy researchers and took place on two consecutive days in the very kitchen of the school where the lunch ladies are placed. The creation of the technical cards enabled the development of 4 lunch menus, composed by side orders, protein dish, vegetables and/or salad, dessert and drink, destined only to the intervention group 2 school.

The control group will continue to receive the standard treatment previously offered to all schools in Sumidouro’s municipal school system, without any loss to students.