The association between changes in height and obesity in Chilean preschool children: 1996-2004.

OBJECTIVES: To establish the association between changes in height and prevalence of obesity between 1996 and 2004 in Chilean preschool children. RESEARCH METHODS AND PROCEDURES: Children registered in the Junta Nacional de Jardines Infantiles (JUNJI) preschool program were routinely measured three times a year and in consecutive years. Two indices [weight-for-height z-scores (WHZ) and BMI centiles] were used to define obesity. Height-for-age z-scores (HAZ) were used to define stature. Generalized estimating equations were used to describe the relationship between stature and obesity while controlling for repeated measurements in children. RESULTS: The prevalence of obesity in Chilean preschool children has remained relatively constant over the past 9 years when either index is used. The prevalence of stunted (HAZ<or=2) children has decreased, while the proportion of children who are tall (HAZ>2) has increased. Using WHZ to define obesity, stunted children do not seem to be at increased risk of obesity when compared with children of normal height. Tall children were strongly protected from obesity when either criterion was used to define obesity. DISCUSSION: The lack of association between stunting and obesity using WHZ>or=2 likely reflects the fact that Chile is in the post-transitional phase of the nutrition transition, and stunting is no longer a risk factor for obesity; however, the associations observed between stunting and BMI suggest that either WHZ or BMI, or both, are inaccurate criteria to define overweight. CONCLUSIONS: There is a unique relationship between stature and obesity in preschool children that is different from that observed in older children in the same population.