Anthropometry as a tool for measuring malnutrition: impact of the new WHO growth standards and reference.

Anthropometry is a useful tool, both for monitoring growth and for nutritional assessment. The publication by WHO of internationally agreed growth standards in 1983 facilitated comparative nutritional assessment and the grading of childhood malnutrition. New growth standards for children under 5 years and growth reference for children aged 5-19 years have recently (2006 and 2007) been published by WHO. Growth of children <5 years was recorded in a multi-centre growth reference study involving children from six countries, selected for optimal child-rearing practices (breastfeeding, non-smoking mothers). They therefore constitute a growth standard. Growth data for older children were drawn from existing US studies, and upward skewing was avoided by excluding overweight subjects. These constitute a reference. More indicators are now included to describe optimal early childhood growth and development, e.g. BMI for age and MUAC for age. The growth reference for older children (2007) focuses on linear growth and BMI; weight-for-age data are age-limited and weight-for-height is omitted. Differences in the 2006 growth pattern from the previous reference for children <5 are attributed to differences in infant feeding. The 2006 'reference infant' is at first heavier and taller than his/her 1983 counterpart, but is then lighter until the age of 5. Being taller in the 2nd year, he/she is less bulky (lighter for height) than the 1983 reference toddler. The spread of values for height and weight for height is narrower in the 2006 dataset, such that the lower limit of the normal range for both indices is set higher than in the previous dataset. This means that a child will be identified as moderately or severely underweight for height (severe acute malnutrition) at a greater weight for height than previously. The implications for services for malnourished children have been recognised and strategies devised. The emphasis on BMI-for-age as the indicator for thinness and obesity in older children is discussed. The complexity of calculations for health cadres without mathematical backgrounds or access to calculation software is also an issue. It is possible that the required charts and tables may not be accessible to those working in traditional health/nutrition services which are often poorly equipped.