Premium
Defining Infant Rapid Weight Gain to Best Predict Subsequent Obesity
Author(s) -
Eckhardt Cara L,
Leo Michael C,
Karanja Njeri
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.687.3
Subject(s) - mcnemar's test , medicine , percentile , logistic regression , anthropometry , obesity , weight gain , standard score , weight for age , confidence interval , demography , body mass index , disease control , pediatrics , body weight , statistics , mathematics , environmental health , sociology
Infant rapid weight gain (RWG) is a growth pattern defined as an increase in weight‐for‐age z‐score (WAZ) > 0.67, and has been identified as a potentially useful predictor of childhood obesity. However, the relevant literature is primarily based on the Centers for Disease Control and Prevention (CDC) growth references rather than the currently recommended 2010 World Health Organization (WHO) growth standards, and is inconsistent with respect to the age interval used to determine RWG. We used anthropometric data from n=38,515 singleton, live term births among Kaiser Permanente Northwest HMO members from 1996–2012 to address these concerns. We compared the prevalence of RWG using criteria varying by growth reference and three key age intervals. The prevalence of RWG was significantly different (Exact McNemar's significance probability <0.05) between growth references (CDC v. WHO) for all three age intervals tested: 0–6 months (31.8% v. 21.2%), 6–12 months (3.4% v. 17.1%) and 0–12 months (19.9% v. 27.0%). Significant Cochran's Q tests indicated that prevalence also differed significantly (p<0.05) by age interval within each growth reference. In a series of logistic regressions, RWG was modeled as a predictor of obesity (BMI ≥ 95 th percentile using the CDC growth references recommended for this age group) at 36 months of age for the subset of children with follow‐up data (n=13,424) using each of the different sets of RWG criteria. While RWG, regardless of the criteria used, was a statistically significant independent variable within each model, the ROC curves associated with the models indicated that none of the RWG specifications performed as a strong screening tool for obesity at 36 months; the highest area under the ROC curve value using the CDC references was observed for the 6–12 month age interval (0.64), and was seen for the 0–12 month age interval using the WHO growth standards (also 0.64). Clearly the growth reference and age interval used to define RWG have implications for determining prevalence and for potential screening of subsequent obesity risk, highlighting the importance of using a consistent reliable definition of this measure in practice. Future work will explore differences by gender, and will determine whether the inclusion of potential confounders and interactions in predictive models would provide improvement in the validity of RWG as a screening tool for childhood obesity. Support or Funding Information Research Support: R03 HD075841