Comment about the article: “Association of metabolic syndrome with low birth weight, intake of high-calorie diets and acanthosis nigricans in children and adolescents with overweight and obesity”

Dear Sir,

We have read with great interest the article entitled “Association of metabolic syndrome with low birth weight, intake of high-calorie diets and acanthosis nigricans in children and adolescents with overweight and obesity”.¹ This study addresses a significant public health problem that is increasingly prevalent in Mexico.² We would like to discuss here some methodological aspects that may be beneficial for future studies.

First of all, overweight and obesity are highly prevalent conditions among Mexican schoolchildren.¹ To investigate them, the authors actually used a cross-sectional study, but they have wrongly reported it as a case–control study.³ In cross-sectional studies, the proportion of subjects in a sample with a given condition is determined, while in a case–control study, the subject's status (case or control) is already known, and separate sampling strategies are used for cases and controls.⁴ In addition, studies conducted at hospitals and on subjects having several conditions and factors are susceptible to Berkson's bias.⁵ It is therefore quite possible that, in this study, the “exposure” variables studied (low birth weight, intake of high-calorie diets, and acanthosis nigricans, particularly the last two) were systematically greater in these cases only by hospital sampling, which should have been mentioned as a limitation.

In addition, the variable acanthosis nigricans is a manifestation of insulin resistance, and cannot therefore be considered as an exposure variable for metabolic syndrome, but rather as a response variable to the former, based on the basic principle of the temporal relationship between variables.⁶ Thus, both variables are already conceptually related, and OR calculation, although mathematically correct, is therefore not appropriate. Finally, measurement of the diet and weight variables may be biased. The 24-h recall is a tool that allows for the characterization of the current or recent intake patterns of a person, but it does not measure the intake of a high-calorie diet over long cumulative time periods nor does it necessarily result in metabolic syndrome.⁷ As regards the evaluation of birth weight, this may be affected by the recall bias of the mother, because of the time elapsed.

Overall, we think that the study helps us to better understand the problem of overweight and obesity in Mexican children and adolescents. We also hope that the above comments may be taken into account in future studies aimed at better characterizing and understanding these public health problems that affect the pediatric population.