RESUMO
OBJECTIVES: International growth charts have been used in the past decades to identify atypical growth and diagnose the nutritional status of individuals. The aim of this study was to construct and compare growth patterns of normo-nourished children between 6-59 months from Afghanistan, Haiti, and the Democratic Republic of the Congo, to assess if it would be worth developing growth charts at a national level. METHODS: We used an international sample of 46 466 subjects (53.7% female; 46.3% male) from the aforementioned regions. To create the growth charts, we used different statistical methodologies: the Lambda-Mu-Sigma (LMS), LMSP, and LMST models, and regression models based on fractional polynomials. The LMSP models were the ones that fitted our data best and were therefore the ones used to make comparisons between countries using percentiles (3rd, 50th, and 97th). RESULTS: We found that Haitian children were both, taller and heavier than their Afghan and Congolese equals of the same ages. Moreover, differences were bigger in the highest percentiles (i.e., 97th percentile). These differences might be the result of the influence that genetics and diverse social and environmental contexts have on growth rates. CONCLUSIONS: Using the same international reference standards for all populations could result in the overestimation or underestimation of the proportion of malnourished children. In light of our results, we recommend the future development of national and regional growth charts to provide health workers with more precise tools to evaluate the nutritional status in the child population.