Differences between recumbent length and stature measurement in groups of 2- and 3-y-old children and its relevance for the use of European body mass index references.

OBJECTIVE: To compare length and stature measurements of young children and to examine the relevance of any difference for comparison with body mass index (BMI) references designed for use from birth to adulthood.
SUBJECTS: A total of 426 2-y-old and 525 3-y-old children included in the Dortmund Nutritional and Anthropometrical Longitudinally Designed (DONALD) Study. DATA ANALYSIS: Length and stature were measured to the nearest millimetre using a stadiometre. Agreement between both measurements at age 2 and 3 y, respectively, was determined by mean differences and by comparison with the German BMI reference.
RESULTS: The average length of 2-y-old girls and boys was 88.3 (3.1) and 89.9 (3.2) cm, mean differences (stature minus length) were -0.47 (0.65) and -0.45 (0.64) cm. The corresponding BMI values were 16.18 (1.3) and 16.46 (1.2) kg/m2, with mean differences of +0.17 (0.24) and +0.16 (0.23). According to stature, 9.4% of the girls and 10.8% of the boys were overweight (>90th percentile), while length classified 7.1 and 9.4% as overweight. Similar mean differences between length and stature were observed at age 3 y: -0.53 (0.62) and -0.47 (0.65) cm in height and +0.17 (0.20) and +0.14 (0.20) kg/m2 in the BMI of girls and boys, respectively. According to stature, 7.6 and 7.3% were overweight as opposed to 5.4 and 4.8% using length. The observed differences increased with higher BMI levels.
CONCLUSION: Changing measurements from length to stature results in an upward shift of BMI, not reflected in current European BMI references. This small but systematic error may result in misinterpretation of individual BMI levels or trend observations.