Estimating the Double Burden of Malnutrition among 595,975 Children in 65 Low- and Middle-Income Countries: A Meta-Analysis of Demographic and Health Surveys

Introduction: Given the changing global nutrition landscape, the double burden of malnutrition is a major public health challenge in many developing countries. The main aim of this study is to estimate the double burden of malnutrition among children in low- and middle-income countries (LMICs). Methods: This study used cross-sectional data from Demographic and Health Surveys (2001–2016). A meta-analysis was conducted to estimate the prevalence of malnutrition indicators in 595,975 children under five years from 65 LMICs. Significant heterogeneity was detected among the various surveys (I2 >50%), hence a random-effect model was used. Sensitivity analysis was also performed, to examine the effects of outliers. Results: The pooled estimate for stunting, wasting, underweight, and overweight/obesity was 29.0%, 7.5%, 15.5%, and 5.3% respectively. Countries with the highest coexistence of undernutrition and overweight/obesity were: South Africa (stunting 27.4% (95% CI: 25.1, 29.8); overweight/obesity 13.3% (95% CI: 11.5, 15.2)), Sao Tome and Principe (stunting 29.0% (95% CI: 26.8, 31.4); overweight/obesity 10.5% (95% CI: 9.0, 12.1)), Swaziland (stunting 28.9% (95% CI: 27.3, 30.6); overweight/obesity 10.8% (95% CI: 9.7, 12.0)), Comoros (stunting 30.0% (95% CI: 28.3, 31.8); overweight/obesity 9.3% (95% CI: 8.3, 10.5)), and Equatorial Guinea (stunting 25.9% (95% CI: 23.4, 28.7); overweight/obesity 9.7% (95% CI: 8.0, 11.6)). Conclusions: There is an urgent need to strengthen existing policies on child malnutrition to integrate and scale up opportunities for innovative approaches which address the double burden of malnutrition in children under five years in LMICs.


Introduction
Child malnutrition in all its forms is a major public health challenge in low-and middle-income countries (LMICs). Malnutrition refers to a lack of proper nutrition either from insufficient nutrient and/or energy intake (undernutrition) or excessive nutrients and/or energy intake (overnutrition). Undernutrition and overnutrition have different effects on the child's development and cognitive performance. Child undernutrition including stunting, wasting, and underweight adversely affects survival, physical and cognitive development, and reproductive and economic productivity of individuals as well as increasing susceptibility to acute and chronic diseases in adulthood [1,2]. However, childhood undernutrition is closely linked to overweight and obesity later in life. According to the World Health Organisation (WHO), children who have suffered from undernutrition and were born with low birthweight or stunted are at far greater risk of being overweight and obese when faced with energy-dense diets and a sedentary lifestyle later in life [3]. Childhood overnutrition negatively affects a child's physical health, and social and emotional well-being resulting in low self-esteem, depression, and social isolation as well as obesity-related medical conditions including liver disease, sleep apnea, type 2 diabetes mellitus, cardiovascular disease, and asthma and other respiratory problems [4]. Most LMICs experience the co-existence of both undernutrition and overnutrition in the same population. This phenomenon, popularly known as the "double burden" of malnutrition, is rapidly gaining global attention, especially in LMICs [5,6], hence, the need for action plans that target both undernutrition and overnutrition simultaneously.
The WHO double-duty actions aim to integrate and harness the synergetic effect of multiple interventions, programs, and policies to simultaneously address the burden of both undernutrition and overnutrition as well as diet-related noncommunicable diseases [7]. The approach recommends the adoption of multisectoral approaches combining nutrition-specific (addressing immediate causes of undernutrition) and nutrition-sensitive (addressing underlying causes of undernutrition) interventions to effect a more holistic sustainable response to mitigate the short-and long-term effects of the double burden of undernutrition [8,9]. Therefore, to properly operationalize the WHO double-duty actions and maximize pre-existing policies linked to nutrition in the Sustainable Development Goals (SDGs), baseline data on the double burden of child malnutrition are urgently needed, especially in LMICs. Hence, the aim of this study is to estimate the double burden of malnutrition among children under five years in LMICs to identify high-priority countries for implication of the WHO double-duty actions, thus making proper use of limited resources.

Data Sources
This study used data from the most recent Demographic and Health Survey (DHS) of 65 LMICs. The datasets are publicly available from the DHS website [10]. DHS are population-based and nationally representative surveys with high response rates (>90%). The surveys are comparable across countries and have large sample sizes (usually between 5000 and 30,000 households). The households were selected using a multistage cluster sampling method. Three core questionnaires are used in DHS surveys: A household questionnaire, a women's questionnaire, and a men's questionnaire. Eligible participants in all households were women and men aged 15 ± 49 years. Details on data collection and sampling methodology employed by DHS are described elsewhere [10]. Only countries with DHS post-2000 AD were included in the analysis in order to capture the double burden of malnutrition from the introduction of the millennium development goals (MDGs).

Outcome Variable
Child malnutrition was measured by stunting, wasting, underweight, and overweight/obesity in children under five years. The DHS manual [11] indicates that the age of the child was determined from the date of interview and the date of birth of each child. Weight was measured using SECA digital scales and recorded in kilograms (kg) to the nearest 0.1 kg. The height was measured using a wooden support board (Shorr Board) and recorded in centimeter (cm) to the nearest 0.1 cm. Children <2 years or <85 cm were measured lying down (length) and those ≥2 years ≥85 cm were measured standing up (height). Stunting, underweight, and wasting were defined as height-for-age, weight-for-age, and weight-for-height <−2 standard deviations of the 2006 WHO Standard Growth Reference. Overweight/obesity was defined as weight-for-height >2 standard deviations [12]. For countries with DHS data prior to 2006, the anthropometric indicators were recalculated using WHO growth standards [12]. Stunting was used as the key indicator for monitoring child undernutrition as endorsed by the World Health Assembly (WHA) [13].

Discussion
This study estimated the double burden of child malnutrition in LMICs. Countries with the highest burden of child undernutrition were Timor-Leste, Burundi, Yemen, Nigeria, Madagascar, India, Guatemala, Niger, and Eritrea. Child undernutrition has consequences on morbidity, mortality, human capital development, and economic productivity [16]. Research has shown that most LMICs experience shortage in food supply [17][18][19], environmental challenges such as drought and climate change [20], harmful economic systems, conflict, and limited access to land for agricultural purposes [21] and these could adversely affect agricultural productivity, food security, and child nutrition [22]. Child nutritional status could also be adversely influenced by socio-economic, demographic and agro-ecological factors such as rising cost of living, rapid population growth and desertification [23][24][25], which are prevalent in most LMICs.
Countries with the highest burden of child overweight/obesity were Albania, Azerbaijan, Sao Tomo and Principe, South Africa, and Armenia. The food environment has been a primary driver of the rise in childhood overweight/obesity worldwide over the past half-century [26,27]. Studies have shown that changes in global diets and the global food system due to globalization and trade liberalization could have an adverse effect on nutrition in LMICs [27,28]. Global trade could influence food systems, particularly the availability, accessibility, and affordability of nutritious food, which may lead to the uptake of nutritionally deficient food [27,28].
Countries experiencing the highest coexistence of child undernutrition and overweight/obesity were South Africa, Sao Tome and Principe, Swaziland, Comoros, and Equatorial Guinea.
Optimal child nutrition is a key determinant in achieving global health targets. The growing double burden of malnutrition in LMICs presents a major global public health concern, which, if not properly addressed, might prevent the SDGs from being met. Existing nutrition interventions mostly target single forms of malnutrition [29] and this may lead to slow achievement of the global nutrition target as observed within most LMICs. With limited fiscal and human resources, it becomes necessary to identify opportunities where reduction in multiple forms of malnutrition could be achieved with single interventions. Therefore, policy, programs, and interventions which open effective channels that focus on the simultaneous reduction of both undernutrition and overweight can help accelerate progress in achieving the SDGs. This is the potential of "double-duty actions" [7].
As highlighted in the United Nations Decade of Action on Nutrition 2016-2025 [7], the integrated approach adopted in the double-duty actions leverage on the coexistence of multiple forms of malnutrition and has the potential to address the double burden of child malnutrition via three pathways. First, the "do no harm" approach ensures that current policies, programs and interventions addressing one form of child malnutrition are not inadvertently increasing the risk of other forms of malnutrition. This is mostly observed when interventions targeted at reducing acute child undernutrition through improvement of child feeding practices inadvertently increases a child's susceptibility to long-term risks of overweight. The primary concern this approach aims to address is the imbalance in the impact of nutrition interventions. Second, the "retrofit" approach which leverages on existing actions addressing single forms of malnutrition to simultaneously influence other forms of malnutrition. For instance, initiatives which target exclusive breastfeeding in the first 6 months of life could also address excessive maternal weight gain in the postpartum period, which in turn protects against obesity and some noncommunicable diseases later in life for both infant and mother [30]. This approach is cost-effective with a huge potential for double returns on existing initiatives. It also provides policymakers and program managers with more efficient and integrated action plans to address multiple forms of malnutrition. Finally, the third approach involves development of de-novo initiatives to proactively address the double burden of malnutrition building from identified shared drivers of different forms of malnutrition. This could involve identifying new initiatives targeting the most significant shared drivers of malnutrition in a particular country. Considering that the type of initiative deployed and its effectiveness will vary between and within countries depending on each country's unique local epidemiology, policy, cultural, environmental, and food contexts, therefore, to advance the potential of double-duty actions in countries, it is vital to assess existing national policies and evaluate their impact on outcomes associated with both undernutrition and overnutrition.
This study had some strength. First, the DHS are nationally representative and population-based surveys with large sample sizes. Second, the variables used by DHS are comparable across all countries [10]. However, this study also had some limitations. First, some LMICs were not included because they either had no DHS data or had no comprehensive data on the malnutrition indicators or their DHS was post-2000 AD. Second, the estimation of child nutritional status was limited to weight and height, with limited data on more detailed components of nutritional status, such as body composition or biochemical or metabolic status. Third, due to the difference in years of survey, the results are not comparable for present conditions in some nations.

Conclusions
The burden of child malnutrition varies within LMICs; therefore, interventions to address the double burden of malnutrition should be appropriate to the country setting. Such interventions should simultaneously target shared drivers of both undernutrition and overweight from shared platforms. Additionally, there is an urgent need to assess existing policies and their alignments with the Sustainable Development Goals 2 (Zero Hunger) and 3 (Good Health and Well-Being), and to evaluate impact, dissemination strategies, and scaling up opportunities of innovative approaches to prevent the double burden of child malnutrition in LMICs.