and there is a growing body of evidence that the risk of overweight and obesity starts in early life and increases during adolescence and adulthood [8].
UNICEF recently published updated statistics compiling data on nutritional status of children <5 years of age from all countries with available data [9]. From 2000 to 2018, the global prevalence of stunting in children <5 years of age decreased from 32.5 to 21.9% with prevalence decreasing across all regions globally. However, stunting still affected approximately 149 million children <5 years of age in 2018. At the same time, 49.5 million children (7.3%) were wasted, and an additional 40.1 million (5.9%) were overweight. South Asia has the highest prevalence of stunting (34.4%) and wasting (15.2%). While the actual number of children affected is lower in sub-Saharan Africa given population size, stunting prevalence remains very high (≥30%) across most countries. The number of children affected by stunting is decreasing across most regions, but there was a 29% increase in West and Central Africa, due to population growth. Overweight and obesity among children are increasing globally, but with substantial variability by region. For example, in Eastern Europe and Central Asia, overweight in children increased from 8.2% in 2000 to 14.9% in 2018. In the same period, the prevalence decreased in West and Central Africa from 4.2% in 2000 to 2.8% in 2018.
Unlike anthropometric measures of malnutrition, the global burden and trends in micronutrient deficiency in children are not well quantified. For years, the figure of 2 billion people affected by micronutrient deficiency has been quoted (see for example 10). Unfortunately, the empirical evidence that underpins this estimate is weak, likely based on anemia, iodine, and vitamin A deficiency prevalence from the early 1990 [11]. Thus, it is not useful for tracking progress. Given changes in dietary patterns [12], food fortification [13], and other interventions, its appropriateness even for advocacy purposes 20+ years on is questionable.
More recent data provide national, regional, and global estimates [14]. These have been incorporated in high-quality data visualization tools [10], but data limitations persist. First, the use of anemia as a proxy for micronutrient deficiency may not be appropriate in all regions given its complex etiology (discussed further below). Second, for some nutrients there are considerable limitations to existing biomarkers of status (e.g., blood or urine levels). In the case of zinc, for example, national food balance sheets have been used to estimate the proportion of the population with inadequate intakes as a proxy for deficiency [15]. There are several limitations to this approach including the very low quality of national food balance sheets to represent actual food intake of individuals [16]. Finally, infection and inflammation affect many biomarkers of nutritional status, and until recently, there has been little consensus on how to adjust prevalence estimates to account for this. For example, in one study in Indonesia, the prevalence of iron deficiency was substantially underestimated, while vitamin A and zinc prevalence were overestimated without adjustment for inflammation [17]. The recent Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia project has now published adjustment factors for several micronutrients that should help overcome this challenge [18].
Where high-quality recent data do exist, there is substantial evidence of a high burden of micronutrient malnutrition among children. A recent review identified 14 countries with vitamin A deficiency data in children collected after 2010 and an additional 13 countries with data from 2006 to 2010, 7 from 2001 to 2005, 16 before 2000, and 32 with no nationally representative data [19]. Of the 50 countries with data, only 7 reported mild deficiency and 2 no deficiency (Guatemala, Indonesia); all others had a prevalence considered to be a moderate or severe level of vitamin A deficiency among children. Unfortunately, discrepancies between reviews of survey data where data quality can be verified and global tracking mechanisms have not been resolved. For example, the most recent national survey in Kenya (cited by Wirth et al. [19]) reported a prevalence of 9.2% (mild deficiency), compared to 84.4% (severe deficiency) reported through global tracking data [10].
Fig. 1. Coexistence of childhood stunting and anemia and overweight in adult women, 2017, reproduced from [7].
Other nutrients with evidence of high deficiency prevalence in children in low- and middle-income countries include iron, zinc, and iodine [20]. While data are scarce, vitamin D deficiency is likely also high among children in several countries [21]. There is also evidence that intake of essential micronutrients, specifically vitamins A, D, E and calcium, is insufficient among many children in high-income countries [22]. There is an urgent need for greater investment in the collection and use of high-quality survey data to quantify the magnitude and distribution of micronutrient malnutrition and ensure that programmatic responses are appropriately targeted.
There is growing evidence that many countries are affected by both undernutrition (stunting, wasting, micronutrient malnutrition) and overweight/obesity, often referred to as the double burden of malnutrition. For example, the 2018 Global Nutrition Report [7] mapped the coexistence of stunting in children <5 years of age and anemia and overweight in adult women (Fig. 1). This pattern has important policy implications as countries must deal with undernutrition simultaneously with overweight, obesity, and related noncommunicable diseases. Such national-level data, however, provide no indication whether the issues are concentrated in the same subsets of the population (e.g., geographical regions, urban vs. rural areas), the same households (e.g., overweight mother and undernourished child), or whether the same individuals are affected by >1 forms of malnutrition (e.g., stunted, overweight, micronutrient-deficient child). Much survey data exist that, if analyzed, would allow for more comprehensive estimates of the co-existence of nutritional issues in population subgroups, households, and individuals to inform policy and programs. There is also a need for more comparable approaches to quantifying and studying the determinants of the double burden of malnutrition among countries to permit global tracking [23].
Progress Toward Addressing Malnutrition: Some Examples from the Literature
Several countries have made substantial progress addressing malnutrition in children. In Nepal, for example, stunting declined from 57.1% (2000) to 36.0% (2017), and in Lesotho from 52.7 to 33.4% over the same period [24]. In Brazil, stunting declined from 19% in 1990 to 7% by 2006 [2]. Several of the likely determinants to progress in Brazil include significant increase in exclusive breastfeeding prevalence (2% in the
