The objective of the study was to demonstrate the efficacy of a nutritional intervention based on the Social Cognitive Theory and the Transtheoric Model in the favorable modification of eating habits in adolescents of secondary education.

The hypothesis was that secondary school adolescents receiving a nutritional intervention based on Social Cognitive Theory and the Transtheoric Model presented more favorable modifications in their eating habits than adolescents who only received nutritional information.

A community trial was conducted with pilot testing for the validation of the study instruments. Adolescents from the second grade of the afternoon shift of the Jalisco Technical High School Nº46 participated in the 2015-2016 cycle.

Self-completed questionnaires evaluated socioeconomic variables, physical activity habits, stage of change (Transtheoric Model), constructs of the Social Cognitive Theory and five eating habits: consumption of fruits and vegetables, natural water, dairy products, sugary drinks and processed foods ; As well as anthropometric variables.

The intervention group had twelve workshops on healthy eating habits and two sessions for parents. The control group received three sessions of nutritional information.

All variables were measured again at the end of the intervention to evaluate results in the short term and three months later to know results in the medium term.

DESCRIPTION OF THE DESIGN: The present study is a community trial, since there was control group and intervention group, random group assignment, intervention at group level, pre-test and at least one post-test (López, 2011).

The study counted with pre-test or initial evaluation - evaluation 1 (E1) and two post-tests:

Evaluation 2 (E2) in the short term, at the end of the intervention (approx 4 months after evaluation 1).

Evaluation 3 (E3) in the medium term, three months after the intervention. Prior to the community trial, the validation study of the instruments used in the research was carried out. For this, the questionnaires were applied to adolescents twice (Evaluation A-EA and Evaluation B-EB), with a temporal difference of one week.

CALCULATION OF SAMPLE SIZE:

The sample size calculation for the intervention group and the control group was based on the data obtained in the Birnbaum et al. to the. (2002) regarding the variable "changes in the average number of servings consumed per day of fruits and vegetables". This was the only study that was found to measure changes in dietary habits (food consumption) in a quantitative way, as planned in the present study.

Was used the "formula to calculate sample size when is expected to find an average difference and is known the standard deviation, given reliability and power (considering different variances)." A control ratio was established: cases of 1: 1, a level of significance or probability of error type 1 (α) of 0.05 (reliability of 95%) and a probability of error type 2 (β) of 0.20 (power of 80 %).

The following results were obtained: Sample size for each group 29

SAMPLING SYSTEM:

The sampling system for group formation is non-probabilistic, since the groups were already formed prior to the experiment (2nd grade school groups). However, the form of participation of the school groups in the research, yes was assigned of simple random way (by means of tombola).

Of the six groups corresponding to the 2nd grade school, two groups were randomly selected to perform the validation of the instruments (stage 1 of the research).

The remaining four groups participated in the other stages of the research. After completion of evaluation 1, they were randomly assigned to either the intervention group (two groups) or the control group (two groups).

CRITERIA AND STRATEGIES OF FIELD WORK:

PREVIOUS ACTIVITIES. Presentation of the project to the directors of the educational institution. Organization of times and schedules with the institution. Signature of informed consent for managers of the educational institution.

Random selection of groups that will participate in stage 1 of the research. STAGE 1. Validation of the instruments in the study population. Presentation with parents of adolescents participating in the so-called "validation group" and explanation of procedures, delivery and signing of informed consent.

Presentation with adolescents participating in the so-called "validation group" and explanation of procedures, delivery and signing of informed consent.

Application of questionnaires (Evaluation A). One week later, the second application of questionnaires (Evaluation B). Taking anthropometric measurements. Delivery of results (nutritional status and eating habits) and educational material (triptych) on healthy eating habits of participating adolescents.

STAGE 2. Recruitment for the community trial and Evaluation 1. Presentation with parents of adolescents participating in the "intervention group" or in the "control group" and explanation of procedures, signed informed consent.

Presentation with the adolescents who participate in the "intervention group" or in the "control group" and explanation of procedures, informed consent sign.

Application of questionnaires in both groups. Measurement of anthropometric measures in both groups. Random selection of groups participating in the "intervention group" and the "control group".

Delivery of results (nutritional status and eating habits) to students and parents of both groups.

STAGE 3. Intervention. Note: simultaneous activities in both groups. STAGE 4. Evaluation 2 (at the end of the intervention). Application of questionnaires in control group and intervention group. Anthropometric measures taken to control group and intervention group. Delivery of results to students and parents of both groups. STAGE 5. Evaluation 3 (three months after evaluation 2). Application of questionnaires in control group and intervention group. Anthropometric measures taken to control group and intervention group. Delivery of results to students and parents of both groups. Delivery of report of global results to executives of the educational institution.

MEASUREMENT TOOLS:

Questionnaires. General data format: It was elaborated to collect the necessary data to describe the population, it includes aspects related to data of identification, sex, age, socioeconomic level, habits of physical activity and measures and anthropometric indices. To evaluate the socioeconomic level, there are six multiple-choice or open-ended questions, which were adapted from validated questionnaires (AMAI, 2017; Bauce, 2010; Vera-Romero, 2013); Only those that were adequate for adolescents to be answered were included. To evaluate physical activity, a section of a validated questionnaire is used (Flores-Vázquez, 2014), with slight adaptations. To ensure the validity, the internal consistency and reproducibility of said sections were evaluated. Only the anthropometric evaluation section is filled by the evaluator and not by the adolescent.

Self-completed food habits questionnaire: Contains 20 open-ended questions about the frequency and amount of consumption of fruits and vegetables, dairy products, natural water, sugary drinks and industrialized foods. Some of the questions were extracted from a validated questionnaire (Flores-Vázquez, 2014), however, due to the adaptations and added questions, the reproducibility of the questionnaire was evaluated.

Cognitive social measures related to eating habits in adolescents: Evaluates self-efficacy, facilitating resources, self-regulation, social support, expectations and self-assessment of expectations in relation to eating habits, using Likert type scales, including a total of 28 questions Four to five for each section). The instrument was translated and adapted from a previously validated one (Dewar, 2012). The validation of the resulting instrument was previously performed, for which reproducibility, internal consistency and factor load were evaluated.

Algorithm to measure stage of change: It consists of a total of 4 linked questions about actions and intentions in healthy eating habits, through which the adolescent change stage is obtained, is based on an algorithm previously translated into Spanish (Villalobos, 2012). Their reproducibility and clarity were evaluated for the adolescent, as well as their correlation with the amount of adequate eating habits (criterion validity).

CAPTURE OF THE DATA:

Anthropometric measurements. They were performed by nutrients previously trained and standardized in the anthropometric measurements. One evaluator performed the measurement of body weight and height and another the one of waist and abdominal perimeter; Another person served as a scorer.

Questionnaires. All questionnaires were designed to be autocompleted. For their application, the evaluators gave the questionnaires to the students and indicated the importance of providing accurate information about the questions included in the formats. Each questionnaire has explicit instructions on how to complete it. However, the evaluators remained with the students during the filling of the instruments and solved the doubts. It took 30 minutes to complete the questionnaires.

DATABASE AND COMPUTER PROGRAMS:

The preparation of the thesis document was carried out in the program Microsoft Office Word 2013. The database was elaborated in the program Microsoft Office Excel 2013 and later exported to the statistical program SPSS version 22, for its analysis. The calculation of the anthropometric indices was done through the WHO Anthro Plus program

STATISTIC ANALYSIS:

Quantitative variables are expressed in quartiles (non-parametric) or some in mean and standard deviation (parametric). The qualitative variables are expressed in frequency and percentage. Non-parametric tests were used for the qualitative variables. It was verified whether the quantitative data are normally distributed by means of the Kolmogorov-Smirnov (n> 50) or Shapiro-Wilk (n≤50) test, which allowed the use of parametric or non-parametric tests (Table 13) .

Intra-group tests. Quantitative variables: Student's parametric T test for related samples, Wilcoxon non-parametric test. - Evaluate changes in food consumption (portions / day) in the short term (evaluation 1 to 2) and medium term (evaluation 1 to 3) in the intervention group and in the control group. -Evaluate modifications in the score of each TCS construct, in the short and medium term, in each group. -Evaluate changes in the anthropometric indicators, in the short and medium term, in each group.

Ordinal qualitative variables: Wilcoxon non-parametric test -Evaluate differences in the change stage, in the short and medium term, in each group. - Evaluate differences in the distribution of nutritional status, in the short and medium term, in each group. -Evaluate differences in the distribution of the food habit "dairy consumption", in the short and medium term, in each group. -Evaluate differences in the distribution of the answers of the questions about cognitive-social measures, in the short and medium term, in each group.

Dichotomous qualitative variables: Non-parametric test Mc Nemar -Evaluate differences in the distribution of each food habit classified as "adequate or inadequate", in the short and medium term, in each group. -Evaluate differences in the presence of abdominal obesity, short stature and overweight / obesity, in the short and medium term, in each group.

Intergroup testing. Quantitative variables: Student's T parametric test for independent samples, Non-parametric test U of Mann Whitney - To compare each eating habit, TCS construct and anthropometric indicator, in each evaluation, between the intervention group and the control group. -Compare the difference in the modification of each eating habit, in the short and medium term, between the intervention group and the control group. -Compare the difference in the modification of each TCS construct, in the short and medium term, between the intervention group and the control group. -Compare the difference in the modification of each anthropometric indicator, in the short and medium term, between the intervention group and the control group.

Qualitative variables: Non-parametric test Chi-square test or Fisher's exact test - Compare the distribution in the socioeconomic level and physical activity habits (baseline), of the intervention group vs the control group. -Compare the distribution in the nutritional status and the stage of change, in each evaluation, of the intervention group vs the control group. -Compare the distribution of each food habit classified as "adequate or inadequate", in each evaluation, of the intervention group vs the control group. -Compare the changes in the stage of change, in the short and medium term, of the intervention group vs the control group.

Inclusion Criteria:

• Informed consent signed by parent or guardian.
• Informed consent signed by the adolescent.

Non-inclusion criteria

• Presence of diagnosed disease, through which dietary habits (diabetes mellitus, kidney disease, etc.) have been modified.
• Pregnancy or lactation period.
• Voluntary practice of vegetarian diets (abstention of food of animal origin).

Exclusion Criteria:

• Two missing assessments.
• Voluntary decision to leave the study.
• Attendance at sessions <50%.
• Initiation of medical or nutritional treatment that influences their eating habits.

Parental participation was low