Background

Current evidence suggests that screen media exposure leads to obesity in the pediatric population through increased eating while viewing; exposure to high-calorie, low-nutrient food, and beverage marketing that influences children’s preferences. However, still scarce on the literature about the reliability and predictive validity of the subjective (such as questionnaire) measure methods in children and adolescents from low-in-middle income countries. To estimate the reliability and predictive validity of food and beverage marketing/advertising questionnaire in children (3-10 years old) and adolescents (11-18 years old) from seven South American cities: Buenos Aires, Lima, Medellin, Montevideo, Santiago, São Paulo, and Teresina.

Methods

We included a subsample from SAYCARE Study (children: n=330; adolescents: n=215]). The questionnaire consists of seven questions about food and beverage marketing/advertising and decision influence. The reliability assessed by temporal stability (2-week interval) and internal consistency. The validity was assessed by predicting the chance of an excess of weight (classified by IOTF body mass index cutoff points) for a given reported preference.

Results

In children, reliability agreement ranged from 63.7% to 86.3% by kappa coefficients, and the Cronbach’s α (internal consistency estimate) ranged from 0.14 to 0.75. In adolescents, reliability agreement ranged from 78.9% to 85.7%, and the Cronbach’s α ranged from 0.14 to 0.76. On the other hand, the predictive probabilities for an excess of weight ranged from 22.3% to 61.1% in children and ranged from 24.9% to 64.1% in adolescents.

Conclusions

The questionnaire shows acceptable to strong reliability and moderate predictive validity in a pediatric population from low– to middle-income countries. These findings suggest that obesity probability could be estimated by the subjective measurement of screen/marketing media exposure.