Diana Barbosa Cunha

Design of a school randomized trial for nudging students towards healthy diet and physical activity to prevent obesity: PAAPAS Nudge study protocol

Objective: To evaluate the effectiveness of nudge activities at school on the students’ body mass index (BMI). Design: School-based factorial randomized community trial. Setting: Eighteen public schools in the municipality of Duque de Caxias, metropolitan area of Rio de Janeiro, Brazil. Participants and intervention: The 18 schools will be randomized into 4 group arms: group 1—control (without any activity); group 2—will receive educational activities in the classroom; group 3—will receive changes in the school environment (nudge strategies); group 4—will receive educational activities and changes in the school environment. Activities will occur during the 2018 school-year. Main outcome measure(s): The primary (BMI) and secondary (body fat percentage) outcomes will be assessed at baseline and after the study using a portable electronic scale with a segmental body composition monitor. The height will be measured by a portable stadiometer. Analysis: Statistical analyses for each outcome will be conducted through linear mixed models that took into account the missing data and cluster effect of the schools.

Changes in breakfast frequency and composition during adolescence: The Adolescent Nutritional Assessment Longitudinal Study, a cohort from Brazil.

Objectives To estimate changes over time of breakfast frequency and foods/food groups consumed at breakfast. Methods Cohort of 809 students aged 10–16 years old from Rio de Janeiro, Brazil, enrolled in 2010, and followed for three years. Breakfast skippers were those not eating breakfast. Those with breakfast frequency of 4 or less times per week were considered irregular breakfast eaters. Changes over time of breakfast frequency and breakfast foods/food groups were analyzed by generalized estimating equations. Results At baseline, overweight/obese girls had higher prevalence of irregular breakfast than those non-overweight/obese (40% vs. 26%; p = 0.005); among boys, there was no significant difference in the prevalence of irregular breakfast according to weight status. After three years, among girls there was an increase in the risk of irregular breakfast consumption (RR = 1.29; 95% CI: 1.08; 1.54) and breakfast skipping (RR = 1.63; 95% CI: 1.12; 2.38). Also, overweight/obese boys (RR = 1.40, 95% CI: 1.03, 1.89) and non-overweight/obese girls (RR = 1.54, 95% CI: 1.17, 2.04) had higher risk of irregular breakfast in three years, compared to baseline. After three years, boys changed the consumption of foods/food groups at breakfast and presented higher risk of decreased intake of fruits (RR = 1.60, 95% CI: 1.20, 2.13), sandwiches and snacks (RR = 1.58, 95% CI: 1.12, 2.22), chips (RR = 1.43, 95% CI: 1.01, 2.13), and ham (RR = 1.52, 95% CI: 1.12, 2.07), and lower risk of cheese intake (RR = 0.73, 95% CI: 0.53, 1.00). Girls had higher risk of decreased intake of fruits (RR = 2.08, 95% CI: 1.47, 2.95), milk (RR = 1.49, 95% CI: 1.07, 2.08), chocolate powder (RR = 1.54, 95% CI: 1.11, 2.14) and ham (RR = 1.65, 95% CI: 1.16, 2.36). Conclusions Changes in breakfast patterns are different according to sex. Also, different changes in breakfast frequency according to BMI category were found. Consumption patterns of some foods/food groups have a tendency to become changed from initial to middle adolescence.

Quality of food choices of Brazilian adolescents according to individual earnings

OBJECTIVE To evaluate the quality of food choices according to adolescent individual earnings in Brazil. DESIGN Adolescents were classified according to their individual earnings as having or not having spending power for their own expenses. Food records from two non-consecutive days of the Brazilian National Dietary Survey (NDS 2008-2009) were used to estimate food intake. Quality of food choices was based on two approaches: (i) the NOVA classification, which classifies processed and ultra-processed foods and drinks as unhealthy food groups; and (ii) traditional classification, with beans, milk, fruits and vegetables as healthy food groups, and soft drinks, sweets, snacks and crackers classified as unhealthy. We compared mean food intake (g/kJ or ml/kJ) according to per capita household income (tertiles) and adolescent individual earnings, with adolescent earnings adjusted for household income, using multiple linear regression. SETTING Brazilian households (n 13 569). SUBJECTS Adolescents aged 14-18 years (n 3673). RESULTS Males without individual earnings had higher per capita household income than those with individual earnings. Household income was associated with all three food groups of the NOVA classification and seven of the eight groups of the traditional classification. However, only beans and snacks were consumed in significantly greater quantities by adolescents with individual earnings compared with those without earnings. CONCLUSIONS Adolescent individual earnings were not the main driver of food choices; however, per capita household income was associated with food choices. The consumption of both healthy and unhealthy items increased with increasing household income.

Authors’ response to Li et al. Obes Facts 2015;8:DOI: 10.159/000381434.

Li et al. [1] have concerns regarding our statement that adjusting for the baseline values of parameters that are highly influenced by baseline values can bias the results in randomized studies. They argued that this is a standard procedure and that our study should be retracted mainly because it contradicts the epidemiology literature. Our study cannot be retracted exactly because it indicates this possibility of bias. We are not arguing that randomization procedure with unbalanced baseline is due to chance, therefore there is no selection bias. However, in this situation, usually in school-based interventions to prevent obesity, adjustment may have biased the results. We observed in our systematic review that 35% of the reviewed studies had statistically significant unbalanced measures of BMI at baseline. We also highlight that cluster designs favor unbalanced baseline measures. The statement highly criticized in the letter by Li et al. [1] was demonstrated in our analysis of a cluster randomized trial, as indicated in the discussion of our paper [2] . This analysis showed no association between intervention and BMI change when analysis was uncontrolled for baseline BMI and changed to a statistically significant association when baseline BMI was included in the model. Li et al. [1] demand that the article should be retracted based on two points that we will address further, but did not comment on this specific result [2] showing a biased conclusion after baseline adjustment. Also, they ignore our main suggestion that ‘procedures that disentangle the effects of group, time and changes in time, such as mixed effects models, should be used as standard methods in school-based studies’. This kind of analysis clearly indicated that mean baseline BMI was the main difference between groups, without differences in the longitudinal change in our study [2] . Therefore, we have documented the possibility of bias by baseline adjustment for BMI in this specific study. In the review paper of school-based interventions [3] , we evaluated the magnitude of unbalanced BMI between groups at baseline and how authors have analyzed their data. Li et al. [1] concluded that our assumption is simply untrue, but we have shown i) the effect of baseline adjustment and ii) the high percentage of unbalanced BMI data (35%) at baseline in the reviewed studies. Received: January 6, 2015 Accepted: February 18, 2015 Published online: April 9, 2015