YM Tam, PE Karlberg, YW Kwan, MC Tsang, HP Sheng, Q He, CK Lo

Objective: The body mass index (BMI) between Chinese and Caucasian infants was compared so as to assess the usefulness of the NCHS growth reference in the assessment of nutritional status of Chinese children.

Methods: Mean weight, length and BMI were compared between 6 cohorts of Chinese children (from Taiwan, the urban and rural areas of China, and from Hong Kong in 1984, 1993 and 1994) and 5 cohorts of Caucasian children (from Canada, Switzerland, Sweden, Britain and the Netherlands). These data of the two ethnic groups were also compared with the growth reference of the National Center for Health Statistics (NCHS). All BMI values were derived from weight/height*height in this study. Individual BMI values of the Chinese children in the Hong Kong 1994 study were analysed and compared with the mean BMI values derived from the NCHS growth reference.

Results: The changes in the mean BMI curves during the first two years of life in the two ethnic groups were entirely different but the different cohorts in the same ethnic groups displayed a similar pattern of change with age. As compared to the Caucasian infants, the BMI in the Chinese infants increased earlier in life with a peak value at 4-6 months and followed by a decline between 6 and 24 months of age. The change in BMI in children of the Caucasian cohorts was very similar to the BMJ values derived from NCHS growth reference values. In the Hong Kong 1994 study, 23% of the children had their mean BMI higher than the NCHS reference at 3 months of age and 90% of the children had their mean BMI lower than the NCHS reference at 15 months. The difference in change in BMI was related to the difference in change in mean weight in children of the Chinese cohorts as compared to the NCHS weight-for-age reference. In contrast, the change in mean height of the well-nourished present Hong Kong Chinese children followed the mean NCHS height-for-age values.

Conclusions: We speculate that the difference in BMI between Chinese and Caucasian children could be genetic in nature in the first 6 months of life while infections, cultural difference in the weaning diet and child care and other socioeconomic factors may be operative between 6 and 24 months of life. The results of this study suggest that linear growth would be better for the assessment of health and nutrition in infancy and early childhood. If BMI and weight for height standards were to be used, then an ethnic specific and population based reference should be used.