The purpose of this study is to assess the effects of a six months nutrition intervention, delivered and taught by classroom teachers with in-service nutrition training, on the prevention of overweight and obesity among children in grades 1 to 4.

In this randomized controlled trial, 464 children from seven elementary schools were allocated to a six months nutrition educational intervention by their own teachers or to standard care. Intervened teachers had 12 sessions of three hours each with the researchers throughout six months, according to the following topics: nutrition and healthy eating (sessions 1 to 4); the importance of drinking water (session 5); strategies to encourage fruits and vegetables consumption and to decrease high energy-density foods intake (sessions 6 to 8); strategies to increase physical activity (sessions 9 and 10); and healthy cooking activities (sessions 11 and 12). After each session, teachers were encouraged to develop activities in class that focused on the learned topics. Sociodemographic, anthropometric, dietary, and physical activity assessments were performed at baseline and the end of intervention.

We expect that fewer intervened children become overweight and the consumption of fruit and vegetables will be higher. In addition, we expect a less consumption of low-nutrition high dense foods.

We also expect to provide further support to decrease overweight epidemic, involving classroom teachers in a training intervention and making them dedicated interventionists.

The prevalence of obesity continues to increase (Ogden et al., 2006) and is growing concerns in Portugal and in the world. As successful treatments have been elusive, the prevention should be considered a high priority. So, present efforts are being dedicated to develop and implement overweight and obesity prevention interventions (Gortmaker, et al, 1999; James, Tomas & Kerr, 2004), in the hope that multilevel approach, if occur early in childhood, may improve eating habits and physical activity and therefore, contribute to prevent or reverse this epidemic state.

In each school, previously trained persons performed anthropometric evaluation, using standardized procedures (WHO Expert Commitee 1995). Anthropometric measurements were performed in children with light indoor clothing and barefooted. Weight was measured in an electronic scale, with an error of ±100g (Seca, Model 703, Germany), and height was measured using a stadiometer, with the head in the Frankfort plane. BMI was computed as mass, Kg/height, m2. The prevalence of underweight, normal weight, overweight and obesity was calculated according to the International Obesity Task Force (IOTF) criteria, making a correspondence between the traditional adult cut-off and specific values for children according to gender and age (Cole et al. 2000). A z-score (the number of standard deviations (SD) from the reference population) was calculated for each child using the LMS method and the calculation was determined using the LMS growth add-in for excel (Pan and Cole 2009).

Dietary intake was gathered by 24h dietary recalls obtained by nutritionists and/or trained interviewers. We assessed a single recall before and another after the intervention. The respondents had no prior notification of when the recalls would occur. Training of interviewers included practice using photos and food models to quantify portion sizes, experience in probing information from children without suggesting responses as well as the type of food consumed in detail of fat content, brand name, constituents of mixed dishes and so on. The 24h dietary recall is the most commonly used dietary assessment method because it is easy to administer, can be performed in large-scale studies (Kranz and Sie-Riz 2002, Gomez-Martínez et al. 2009), and can be used to assess adequacy of energy and macronutrient intakes. During the 24h dietary recall, each child was asked to recall all food and beverages consumed during the past 24h. Daily routines were used as prompts (waking up, going to bed, time between classes, and before or after school) to enhance recall. Portion sizes of foods and beverages consumed were also estimated using food models, photos and other props (cups, glasses, food wrappers or containers) as an aid in determining serving sizes. Energy and nutritional intake were estimated using an adapted Portuguese version of the nutritional analysis software Food Processor Plus (ESHA Research Inc., Salem, OR, USA). The 24-hour dietary recall interviews captured the time and name of each eating occasion, the foods and beverages reported at each eating occasion, and the source from which item was obtained.

In order to assess the level of physical activity of children, parents were asked five questions with four answer choices (4-point scale) ranging from 0 (very seldom) to 4 (very often): a) Outside the school does your child take part in organized sport? b) Outside school does your child take part in non-organized sport? c) Outside school, how many times a week does your child take part in sport or physical activity for at least 20 minutes? d) Outside school hours, how many hours a week does your child usually take part in physical activity so much that he gets out of breath or sweat? e) Does your child take part in competitive sport? (Mota and Esculcas 2002). The total sum of the points was computed reaching a maximum of 20 points. To express the activity levels, a Physical Activity Index was obtained, which divided the sample into four activity classes: sedentary group (0-5); low activity group (6-10); moderately active group (11-15); and vigorously active group (16-20), on the basis of their reported physical activity (Mota and Esculcas 2002).

Social, demographic and family characteristics were assessed by questionnaire. The survey sent to parents contained questions about gender and age of children, education of the parents (recorded in five categories: 0, 1-4, 5-9, 10-12, and more than 12 years of formal education). This information was further grouped for analysis into three categories: up to 9 years, 10-12 years, and more than 12 years of education.

The research team proposed, in January 2008, the accreditation of the sessions that were developed with the teachers to the Minister of Education, Scientific-Pedagogic Council for In-service Training (Conselho Científico Pedagógico da Formação Contínua, Ministério da Educação). This proposal was approved in September 2008 in the form of "training workshop" with 72 hours duration, distributed by active learning strategies (36 hours) and work contact with the children (36 hours). Teachers of the intervention group had 12 sessions of three hours each with the researchers throughout six months, which included the following topics: nutrition and healthy eating for the children and the families (sessions 1 to 4); the importance of the water (session 5); strategies to encourage fruits and vegetables consumption and to decrease high energy density foods intake (sessions 6 to 8); to increase physical activity and to reduce screen time exposure (sessions 9 and 10); and healthy cooking (sessions 11 and 12). After each session, teachers resend the learned contents to develop creative and engaging classroom activities about the addressed topic. All the questions that arose during the implementation of classroom activities were addressed and resolved shortly with the researchers. Teachers were allowed to develop and refine the activities and learning strategies that were proposed by researchers. At the end of this period, the teachers delivered a critical report of activities focused on the work contact with children.

The data analysis was performed using SPSS ®, Version 18.0 (SPSS Inc; Chicago, IL).

Inclusion Criteria:

• Children from the schools involved in the study with informed consent provided by parents.

Exclusion Criteria:

• Age more than 12