Measuring child malnutrition: does it matter which population reference is used?

The double burden of malnutrition, defined as the co-occurrence of undernutrition and overnutrition at the population, household or individual level, is evident worldwide, especially among children, and in lower and middle income populations. It has significant implications in terms of adverse health outcomes throughout the life course, and subsequent effects on populations’ economic productivity and health resource needs [1]. Accurate identification of undernutrition and overnutrition in children is essential to inform effective clinical and public health interventions. However, agreeing upon a suitable measure for this purpose has proved to be challenging.

Defining the four main anthropometric measures of child malnutrition – underweight, stunting, overweight and obesity – is a complex task, because (1) “normal” or “extreme” weight or height changes as children grow, and (2) children rarely have adverse outcomes during childhood; thus, cut-offs defining malnutrition cannot be meaningfully linked to disease outcomes, as they are in adults. To address this, a child’s absolute anthropometric measure (e.g. height) is compared against a population reference describing how that measure changes with age in a general population of children. Malnutrition is defined as the child being at the extreme ends of the reference population’s anthropometric distribution for their age and sex; “extreme” is defined arbitrarily (e.g. below 5th percentile) [2].

Currently, four international and multiple local references are used in research and clinical practice globally. Each is based on a different reference population, with its own cut-offs. An important question is: does it matter which reference is used? Research suggests that it does. For example, obesity prevalence was doubled and thinness prevalence approximately halved when using the World Health Organization or Centers for Disease Control references versus the International Obesity Task Force reference in a Pakistan-based study [3]. This would lead to very different conclusions about the extent of malnutrition in this population, and the resources required to address it. So which reference should be used?

For children under five, the widely accepted choice is the WHO 2006 growth standards. The standards are based on a study of healthy children conceived and raised in unconstrained environments in six diverse populations. In this study, children living in such advantaged environments grew very similarly regardless of population, supporting the use of a universal standard for this age group [4]. For older children, there is no clear answer, and references are chosen either arbitrarily or due to ethnic similarities between the reference population and the population of interest. This makes it difficult to meaningfully compare malnutrition estimates between populations, and importantly, highlights a fundamental gap in our understanding of optimal and impaired child growth in older children across populations.

A more coordinated approach towards measuring malnutrition is required, especially among older children. This must include the establishment of a consensus on the universal use of one reference to provide comparable malnutrition estimates across populations and further research characterising unconstrained child growth. Likewise there is a need to assess the accuracy of existing references in predicting later-life adverse outcomes, in diverse populations. These steps will facilitate more effective planning and monitoring of strategies to address the global double burden of malnutrition in children.

Key references

1. WHO. Double burden of malnutrition. 2016 04.10.16]; Available from: http://www.who.int/nutrition/double-burden-malnutrition/en/
2. Cole, T.J., et al., Establishing a standard definition for child overweight and obesity worldwide: international survey. Bmj, 2000. 320(7244): p. 1240-3.
3. Mushtaq, M.U., et al., Height, weight and BMI percentiles and nutritional status relative to the international growth references among Pakistani school-aged children. BMC Pediatr, 2012. 12: p. 31.
4. Borghi, E., et al., Construction of the World Health Organization child growth standards: selection of methods for attained growth curves. Statistics in medicine, 2006. 25(2): p. 247-265.