A1 – Both HK2020 and WHO 2006 are considered suitable for local use for 0-5 years. Healthcare providers and organisations can decide whether or not to prioritise a particular set of charts for children 0-5 years. If WHO 2006 is used from 0-5 years then the step at 5 years (transition to HK2020) may be considered too big, and there will be 2 breaks (one from 0-2 year chart (supine length) to 2 to 5.0 years, and the second break at 5.0 years when transitioning to 5-18 year HK2020 chart) whereas the steps to HK2020 at 2 years are smaller and there will be only one break at 2 years. However some users may wish to use WHO2006 0-5 years for non- Chinese and some users may wish to use HK2020 0-5 years for routine use.
A2 – The Hong Kong Growth Study 2020 revealed that the mean gestational age of infants born in Hong Kong is 38-39 weeks. As such, “Age 0” in HK2020 growth charts represents the birth parameters of infants born at 38-39 gestational weeks instead of 40 weeks. The centile width at age 0 were also narrower as they were derived from a more homogenous Chinese population (i.e. smaller standard deviation).
Therefore healthcare professionals need to pay attention the following and should bear in mind when using the birth weight percentile from HK2020 as a reference point:
- Birth weight centile - Hong Kong infants born at full term (40 week gestation) likely have higher birth weight centile at birth (because the average gestational age has become 38-39 weeks).
- Extreme birth weight – More extreme birth weight would be classified (4-5 % <2nd and 4-5%> 98th) than WHO2006 (2-3% <2nd and 0.3-0.4% >98th) based on routine MCHC data.
- Growth faltering – More infants have downward crossing centile from birth compared to WHO2006 partly due to the narrower difference between centile lines. This has an impact on interpretation from the HK2020 Growth References on the birth weight centile, and growth trajectories for Hong Kong Chinese infants.
A3 – Toddlers, particularly girls, are slightly shorter compared to WHO2006. A relatively large change in centile is expected when swapping WHO2006 to HK2020 at age 5 years.
A4 – Based on the assessment undertaken by the Hong Kong Growth Study, WHO2006 and HK2020 are both suitable for contemporary Chinese children living in Hong Kong under 5 years of age. No growth charts are perfect but an understanding of the population where the growth charts were derived would allow appropriate use of them. Family physicians are welcome to choose between them or use both for growth monitoring for patients under age of 5 years with special attention of the differences stated in this User Guide.
A5 – Given that HK2020 was developed from an ethnic Chinese Hong Kong population and that WHO was developed from a sample from 6 countries, it can be argued that WHO2006 could be a better choice for non-ethnic Chinese children aged 0-5 years. Some countries have adopted WHO2006 totally or partially. US, Canada, Australia adopted the WHO2006 from birth to 24 months. UK adopted the part from 2 weeks to 48 months. Korea adopted the WHO2006 from 0 to 36 months. India and Pakistan adopted for 0-60m. Mainland China also adopted WHO2006 for 0 to 36 months until 2023 when they changed to use national growth reference.
A6 – Hong Kong Growth Study 2020 did not update the gestational-age specific birthweight references for Hong Kong children. Healthcare professionals need to be aware that the mean gestation of births in Hong Kong are 38-39 weeks as the study population deriving the HK2020 growth charts. Our study group does not recommend a fixed gestation for transitioning all preterm babies as the gestation-specific growth curves have different characteristics from our HK2020 growth charts. Growth trajectories should be interpreted cautiously when crossing between growth charts at post conceptional age of 38-40 week gestation.
A7(a) – In Hong Kong, overweight (including obesity) for individuals under 18 years has been defined as body weight > 120% of the median weight-for-height for males with height between 55 and 175 cm and for females with height between 55 and 165 cm; and BMI ≥25 for males with height >175cm and for females with height >165cm. The review in this study noted that the relation between ratio of weight-for-height and adiposity varies by age and thus it is not an ideal measure of childhood overweight.
BMI-for-age instead of weight-for-height is recommended for use to assess overweight and obesity in children. The change is consistent with the practice in many countries. As obesity is less common among young children than those of older age. It has been shown that using the same BMI z-score cut off as older children overdiagnoses obesity under 6 years old, it is therefore recommended to adopt the WHO approach to have higher BMI cut-offs to screen for overweight and obesity for children aged 0 to 5.0 years than for those >5.0 to < 18.0 years.
A7(b) Using the routine data from Student Health Service Centres (SHSCs) from 2016 to 2018, and compared to the current definition of overweight (including obesity), the new definition i.e. BMI > 91% classified similar proportion of overweight (including obesity) among children > 5.0-13 years old but lower proportion at 14-17years old group. Using the routine data of Maternal and Child Health Centre (MCHC) from 2016 to 2021, there were very small differences in proportion of overweight and proportion of obesity by different BMI cut-offs when compared to the WHO definition for children under 60 months (98% vs 2SD; 99.6% vs 3SD)
A7(c) – Changes in definition of overweight and obesity will inevitably change the weight status for some children and the prevalence of childhood overweight and obesity. It is important for healthcare professionals to explain to children and the parents the limitation of using BMI to define adiposity and strengthen the importance of healthy diet and lifestyle for all children regardless of weight status.
A8 – The methodology for Hong Kong growth study 2020 was similar to that used in the 1993 and 2005/6 local growth surveys and was based on advice from overseas experts. In brief, it was a population-based cross-sectional study of a minimum of 175-200 male participants and 175-200 female participants for each age category of 0, 1, 2, 4, 6, 12 months, 6-monthly from 1-18 year(s) and 19-20.5 years. They were recruited from a public birthing hospital, a private birthing hospital, 6 Maternal and Child Health Centres, 5 nurseries, 31 kindergartens, 19 primary schools, 14 secondary schools, one all-through school, 5 vocational schools, 3 universities and a non-governmental organisation. Recruitment sites were geographically distributed throughout Hong Kong.
We did not exert selection criteria at recruitment but we excluded data from the following children when constructing growth charts:
- children of any age with both parents of non-Chinese ethnicity
- children aged < 2.0 years who were born preterm (< 37 week gestation) or multiple pregnancies
A9 – The change in lifestyle and diet among school children during COVID-19 might result in transient increase in rates of obesity. However, mean weight, length/height and BMI from the HK growth survey were comparable with routine growth data from MCHCs and SHSCs, except that boys were taller at 10 years onwards; girls were taller at 10-11 & 17 years and 6-year-old children were slightly thinner.
We did not update the HK1993 BMI-for-age charts (3) above age 2 years so as to avoid normalizing childhood obesity. Therefore the obesogenic effect of COVID-19 and the slightly thinner 6-year-old in HK growth survey, if any, would not have affected the BMI-for-age charts and their function in indicating obesity.
A10 – The new growth charts were constructed with 9 centiles (99.6th, 98th, 91st, 75th, 50th, 25th, 9th, 2nd and 0.4th). The centiles are equally spaced with a width correspondent to two thirds (0.67) of an SD for that age and included two extreme centiles (0.4th and 99.6th). In general, the closer the growth parameters are to the extremes, the more likely they are abnormal. The two extreme centile lines (0.4th and 99.6th) will help screen out the more likely abnormal growth for further evaluation.