Back

Evaluating Swiss growth reference curves: A comparative analysis in Zurich schoolchildren

Leuenberger, L. M.; Belle, F. N.; Mozun, R.; Spycher, B. D.; Mallet, M. C.; Jenni, O. G.; Saner, C.; Latzin, P.; Moeller, A.; Kuehni, C. E.

2024-07-01 pediatrics
10.1101/2024.07.01.24309377 medRxiv
Show abstract

INTRODUCTIONSwitzerland has an ongoing debate about the appropriateness of national growth reference curves. The Swiss Society of Pediatrics currently recommends the growth references of the World Health Organization (WHO), while the Center for Pediatric Endocrinology Zurich has proposed alternative growth references based on local data. Specialists and researchers also use International Obesity Task Force (IOTF) references to define overweight and obesity. We investigated the fit of these three growth references to anthropometric measurements from schoolchildren in the canton of Zurich and assessed the prevalence of overweight, obesity, and short stature across the three references. METHODSWe analyzed data from 3755 children aged 6-17 years of the cross sectional LuftiBus in the school (LUIS) study, collected between 2013-2016 in the canton of Zurich. We calculated z-scores of height, weight, and body mass index (BMI) based on WHO, local, and IOTF references. We compared the mean and distribution of z-scores to the expected standard normal distribution using the Anderson-Darling test. We classified BMI based on cutoff values given by the three references: overweight (WHO: >90.0th percentile; local: >82.9[girls], >78.9[boys]; IOTF: >89.3[girls], >90.5[boys]), and obesity (WHO: >97.0; local: >96.8[girls], >95.5[boys]; IOTF: >98.6[girls], >98.9[boys]). We defined short stature as <3rd percentile of height-for-age. RESULTSThe mean z-scores in LUIS were 0.56 for height, 0.28 for weight, and 0.06 for BMI based on WHO references; 0.15 for height, 0.06 for weight, and -0.01 for BMI based on local references; and 0.19 for BMI based on IOTF references. WHO references provided a worse fit to the LUIS children than local references. Anderson-Darling goodness of fit A2 was 578.1 (WHO) vs. 48.1 (local) for height and 124.0 vs. 10.0 for weight, with lower values indicating better fit. WHO (A2: 24.3) and local references (A2: 0.8) fit the BMI of LUIS children better than IOTF (A2: 64.1). The WHO classified fewer children as overweight than the local and IOTF references (WHO: 9% vs. local: 15% vs. IOTF: 13%) but more children as obese (6% vs. 4% vs. 3%). The WHO defined fewer children as being of short stature than the local references (1% vs. 3%). CONCLUSIONSOur findings suggest that anthropometric data of schoolchildren in Zurich differ notably from WHO and IOTF references potentially leading to misclassification of overweight, obesity, and short stature. Nationally representative and longitudinally collected data are needed to develop new Swiss growth references.

Published in Swiss Medical Weekly (predicted rank #11) · training set

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.