This is a retrospective data analysis study of two groups that received different treatments.
The study population consisted of children and adolescents between six and eighteen years old, of both sexes, who were referred for treatment of overweight and followed at the largest central hospital in Portugal between 2012 and 2017. Children/adolescents with overweight were proposed to nutrition counseling of their multidisciplinary treatment team. The dietitian followed the consultation protocol of his multidisciplinary team. The intervention group protocol included the use of the behavioral contract and time between consultations adjusted to the results of children/adolescents. The control intervention protocol included standard follow-up.
Figure 1 systematizes the course of the interventions under study. In the first visit (M0), a behavioral contract methodology was applied in the BG. This methodology consisted of signing a behavioral contract negotiated between the dietitian, tutors and the children/adolescent, in order to correct the main previously identified food errors (e.g.: skip breakfast, don't eat vegetables, drink soft drinks) through a 24-hour recall. A non-material reward was also defined (e.g.: a bike ride, a visit to a theme park, going to a swimming pool) that would be offered by the tutors of the child/adolescent, if they obtained good results (meet at least two-thirds of the objectives set). In the second visit, after 1 month (M1), if the children/adolescent had loss weight, the reward was offered and the nutritional plan was prescribed. Throughout this treatment, the interval between patient visits ranged between monthly and quarterly in the first six months (M2) and between quarterly and half-yearly in the last six months (M3). In the STG, the nutritional plan was prescribed during the first visit (M0) and the time between visits had a higher interval, quarterly in the first six months (M2) and half-yearly in the last six months (M3) relative to the BG.
In both interventions, dietitians also encouraged the practice of physical activity.
For sample selection, we used the child growth curves of the World Health Organization, which adjust the children’s weight, height and Body Mass Index (BMI) according to sex and age [15]. We included only children and adolescents, with BMI z-score > 1 at the time of admission to the consultation and with at least 12 months follow-up. Children and adolescents with previously diagnosed psychiatric pathology (e.g. autism) or other that compromise the adherence to the program or the collection of the parameters under study were excluded.
The initial sample consisted of 79 children/adolescents of the Behavioral Group (BG) and 153 of the Standard Treatment Group (STG). However, for the analysis of results, only children/adolescents who had at least three assessments (baseline, between three and six months and after 12 months) in the 12-months interval were admitted. Thus, according to these criteria, 35 individuals of the BG and 55 of the STG were selected. Figure 2 summarizes the sample selection.
The analyzed data were based on the evolution of the BMI z-score and body composition. All consultations were in-person. In each consultation data was collected for height (measured with stadiometer), weight and body composition, measured by weight scale and bioimpedance equipment with cross-calibration, respectively. Data from the M0 (baseline), M1 (one month after), M2 (six months after) and M3 (12 months after) of follow-up were collected.
Body composition (fat mass, fat mass index, fat free mass and fat free mass index) was interpreted through reference values adjusted for sex and age published by Wells, JC. et al. [16]. Since the fat mass index and the fat free mass index relate fat mass and fat free mass, respectively, with height [17], these two indexes were chosen as preferential to analyze body composition.
All ethical requirements were ensured, and the study was analyzed and approved by the Ethics Committee of the Faculty of Medicine of the University of Lisbon and the North Lisbon University Hospital Center (reference of approval code: 459/18). The informed consent was also approved by the tutors.
Statistical analysis of the data was performed using the Statistical Package for the Social Sciences (SPSS) version 22 (Microsoft Windows®).
Descriptive statistics consisted in the calculation of means and standard deviations (SDs) for continuous variables and in the presentation of relative and absolute frequencies for ordinal and nominal variables.
To calculate the differences between groups, we used the Student T test (for independent samples) and the Mann-Whitney test. We also used the Student T test for paired samples or the Wilcoxon test to compare the mean differences between the beginning and the end of the intervention.
For the study of evolution over time by groups, the ANOVA test of repeated measurements and the Friedman test were used. For the comparison between groups in the qualitative variables, the Chi-Square test, Fisher's exact test or the Chi-Square test by Monte Carlo simulation were used. Statistical significance for p < 0,05 was accepted in all analyzes.