Double Burden of Malnutrition: A Population Level Comparative Cross-Sectional Study across Three Sub-Saharan African Countries—Malawi, Namibia and Zimbabwe

Background: The double burden of malnutrition in sub-Saharan African countries at different levels of economic development was not extensively explored. This study investigated prevalence, trends, and correlates of undernutrition and overnutrition among children under 5 years and women aged 15–49 years in Malawi, Namibia, and Zimbabwe with differing socio- economic status. Methods: Prevalence of underweight, overweight, and obesity were determined and compared across the countries using demographic and health surveys data. Multivariable logistic regression was used to ascertain any relationships between selected demographic and socio-economic variables and overnutrition and undernutrition. Results: An increasing trend in overweight/obesity in children and women was observed across all countries. Zimbabwe had the highest prevalence of overweight/obesity among women (35.13%) and children (5.9%). A decreasing trend in undernutrition among children was observed across all countries, but the prevalence of stunting was still very high compared to the worldwide average level (22%). Malawi had the highest stunting rate (37.1%). Urban residence, maternal age, and household wealth status influenced maternal nutritional status. The likelihood of undernutrition in children was significantly higher with low wealth status, being a boy, and low level of maternal education. Conclusions: Economic development and urbanization can result in nutritional status shifts.


Introduction
One in three persons globally suffer from one or more forms of malnutrition: wasting, stunting, vitamin and mineral deficiency, overweight or obesity, and diet-related non-communicable diseases (NCDs) [1]. The coexistence of undernutrition (wasting and stunting) alongside overnutrition (overweight or obesity) at any population level: country, city, community, household, and individual is defined as the double burden of malnutrition [2]. In 2018, more than 149 million children were stunted, and 40 million were overweight worldwide, with more than a third of the stunted and a quarter of the overweight children living in sub-Saharan Africa [3]. Sub-Saharan Africa is experiencing the double burden of malnutrition with high levels of undernutrition and a growing burden of overweight or obesity and diet-related non-communicable diseases [4]. As a result, the region is not on track to achieve the Sustainable Development Goals of ending hunger and all forms of malnutrition by 2030 [5].    Gender differences in undernutrition and overnutrition were observed. Data from the latest surveys for all countries revealed that male children were more likely to experience undernutrition than female children (Table 3). There was a marked significant gender difference for Namibia in wasting, with boys having a higher odds of wasting than girls (OR = 2.35, 95% CI: 1.51-3.64, p < 0.001). A similar result was also observed for Zimbabwe, with male children being 1.33 (95% CI: 1.15-1.54, p < 0.001) times more likely of being stunted than female children. Malawian boys were likely to be underweight (OR = 1.29, 95% CI: 1.06-1.57, p = 0.013), stunted (OR = 1.19, 95% CI: 1.03-1.38, p = 0.016), or overweight/obese (OR = 1.82, 95% CI: 1.29-2.56, p = 0.001) compared to female children.
For children under five years old, maternal education influenced the nutritional status of children born in Zimbabwe and Namibia and not in Malawi. For Zimbabwe, the odds ratio of having stunted children were 5.42 (95% CI: 2.54-11.57, p < 0.001) times higher for women with no education, 3.21 (95% CI: 1.83-5.65, p < 0.001) times higher for women with primary education, and 2.59 (95% CI: 1.5-4.46, p = 0.001) times higher for women with secondary education than women who had higher education.
Household wealth status had a significant influence on stunting in children under five years. For all countries, children from the poorest, poorer, and middle quintiles were more likely to be stunted compared to the richest quintile; for Zimbabwe, even those in the rich quintile were also likely to be stunted (Table 3). Malawian children in the poorest quintile were also more likely to be underweight than those in the richest quintile, whereas Namibian children from poor and poorest quintiles were also likely to be underweight (Table 3). Table 3 shows increasing trends of overweight and obesity among women of child bearing age in all the three countries regardless of residency, with the highest prevalence of overweight/obesity observed in Zimbabwe (34.9%), while Malawi had the lowest prevalence (20.7%) in the latest surveys. Declining trends in the prevalence of underweight was observed for all countries, regardless of residency (Table 4). Namibia had the highest prevalence of underweight women (15.25%). A comparison of the overweight-tounderweight ratio for the earliest and latest survey years of all three countries revealed a higher prevalence of overweight/obesity in women of childbearing age was higher than that of underweight, Zimbabwe had a larger difference than other countries, having over five times more women with overweight/obesity than underweight (Table 4).   For Malawi, maternal education did not influence the nutritional status of the women. In contrast, for Zimbabwe, women whose highest educational attainment was primary or secondary education were less likely to experience obesity than the reference group of higher education attainment (Table 5). Namibian women with no education, primary education, or secondary education were more likely to have wasted than those with higher education.

Trends and Prevalence of Undernutrition and Overnutrition in Women of Childbearing Age
For all countries, household wealth status had an influence on overnutrition and undernutrition in women. For both Malawi and Zimbabwe women, experiencing overweight or obesity was less likely among the poorest, poorer, middle, and richer quintiles than the reference group of richest quintiles (Table 5). In Namibia, being overweight or obese was less likely among women from the poorest, poorer, and middle quintiles compared to the richest quintile. The wealth status did not influence the likelihood of being underweight among Malawian women of childbearing age. However, for Zimbabwean women, the odds ratio of being underweight was 3.5 (95% CI: 1.99-6.15, p < 0.001) times higher among those who are in the poorest quintile compared to those who are in the richest quintile. Namibian women of childbearing age in the poorest quintile were 1.81 (95% CI: 1.14-2.88, p = 0.013) times more likely to be underweight than those in the richest quintile.

Food Availability
Per capita energy supply remained stable, with an increase from 2000 to 2017 for all countries ( Table 6). The availability of cereals in Malawi increased by 50%, while the supply decreased by 50% in Zimbabwe. There was a significant increase in meat availability in Malawi. Fruit and vegetable availability in Malawi more than quadrupled, followed by Namibia, which had a 50% increase.

Physical Activity Levels
Countries with higher economic status had an increased prevalence of inadequate physical activity. Namibia had the highest prevalence of inadequate physical activity (33.4%), followed by Zimbabwe (26.8%). Malawi had the lowest prevalence of physical inactivity (15.6%) ( Table 6).

Discussion
We compared the prevalence of undernutrition and overnutrition among children under five years and women of childbearing age in three selected sub-Saharan African countries with varying levels of socio-economic development. Findings were based on data from national surveys, World Bank development indicators, food balance sheets (FAO), and WHO physical activity levels data. Large disparities in nutritional status among young children and women were observed across the three countries. Rural-urban and gender differences were also observed in the prevalence of overnutrition and undernutrition. Overweight/obesity was more common in urban areas than in rural areas.
Many studies showed that socio-economic development and increased urbanization resulted in diets shifts, leading to the increasing burdens of overweight or obesity and diet-related noncommunicable diseases, such as diabetes and heart disease [27]. With increasing socio-economic development, there is also the growth of the urban population. Urban populations tend to consume more calories due to the availability of foods high in saturated and trans fats, refined carbohydrates, simple sugars, salt, animal source food, and processed foods and reduced consumption of traditional starchy carbohydrates as dietary staples [27]. This study found that countries with higher economic status had a higher prevalence of overweight/obesity in childbearing women. Neuman et al., 2014, found a positive association between GDP and mean BMI, though non-significant [28]. However, Namibia, an upper-middle-income country, had the second-lowest prevalence of overweight/obesity in children and the highest prevalence of underweight women. This highlights that besides socio-economic transformation and urbanization, other factors contribute to overweight/obesity in these countries [27]. The perception of a larger body size as a sign of affluence and desiring women with larger body sizes in some African cultures might explain the observed differences in the prevalence of overweight/obesity in these countries [29]. Physical inactivity is linked to an increased prevalence of overweight or obesity and NCDs. This study found that countries with higher economic status had an increased prevalence of inadequate physical activity. This finding is in line with previous studies. For example, one study in Cameroon observed that urban residence compared with rural residence was associated with lower physical activity energy expenditure and higher prevalence of metabolic syndrome [30].
We found rural-urban differences in the prevalence of overweight/obesity were observed with higher levels in urban areas compared to rural areas. In urban areas, markets are increasingly replacing fresh produce and selling commercially prepared and processed foods from transnational and local industries and street vendors [31]. However, it should be noted that increasing trends of overnutrition in both rural and urban areas were observed in all countries. Bixby et al., 2019, found that particularly in LMICs, BMI is rising at the same rate or faster in rural areas compared to urban areas, except among women in sub-Saharan Africa, highlighting that urban living and urbanization may not be the only key driver of the global epidemic of obesity.
Undernutrition in children aged under five years is highly prevalent in the three selected sub-Saharan countries, with stunting being the most prevalent. Rural-urban differences in stunting, wasting, and underweight were observed in most of the countries. Malawi, the country with the lowest gross domestic product per capita, had the highest prevalence of stunting (37.1%) and infant mortality rate. In sub-Saharan Africa, nutritional status was found to be a central determinant of child mortality [32].
In this study, maternal education had a limited influence on women's nutritional status. However, a significant influence was observed for children under five years old. Maternal education on child health and nutrition was well demonstrated to play a significant role in many studies. Higher maternal educational attainment was shown to improve the socioeconomic status of mothers leading to better children's health and nutritional outcomes [33]. Higher socioeconomic status can result in better feeding practices and the utilization of health services, and maternal education improves the mother's knowledge about child health, including causes, prevention, and treatment of diseases [34]. Our findings show the influence of wealth status on overnutrition and undernutrition, with those in the poorest quintile more likely to be undernourished or have stunted children than those in the richest quintile, and those in the richest quintile were more likely to experience overweight/obese than those in the poorest quintile.
The production of food supply increased in most countries, with a significant increase in meat production in Malawi. Meat availability was found to correlate positively with obesity prevalence [35]. However, general market availability or supply of food does not necessarily translate to consumption.
The study has some important limitations. First, the finding cannot be entirely generalizable to the subcontinent, as the three countries included in the study might not be necessarily like the remaining sub-Saharan African countries. Second, we used data from cross-sectional surveys; therefore, causal inference is not viable, and the available data were from different years, which limited our comparisons. However, this study used the most recent national data and compared selected countries based on their economic status, representing Southern Africa, and the findings may only reflect this region.

Conclusions
The double burden of malnutrition in sub-Saharan Africa is of critical concern. Current data indicate that the world is not on track to achieve the United Nations' Sustainable Development Goal 2: Zero Hunger by 2030, which is concerning. In recent years, many health-related policies and interventions in Africa focused on addressing undernutrition and infectious disease; however, the current nutrition status trends highlight the need to address all forms of malnutrition. Policies and interventions to address malnutrition in sub-Saharan Africa and other transitional societies need to be double-pronged and gender-sensitive. There is no one-size-fits-all solution for countries, and policymakers will need to assess the context-specific barriers. When formulating and implementing national policies or interventions, factors such as national economic development, urbanization, food availability, diet quality, and physical activity levels need to be considered. Institutional Review Board Statement: We conducted analysis for publicly available data, no ethical approval was required. Demographic and Health Surveys (DHS) data collection procedures were approved by the Measure DHS (Calverton, MC) Institutional Review Board and by the national body that approves research studies on humans in Namibia, Malawi, and Zimbabwe.
Informed Consent Statement: Written consent was obtained by the interviewers from each participant. The use of the DHS data for this particular study was approved by Measure DHS, and considered exempt from full review because the study was based on an anonymous, public-use dataset.
Data Availability Statement: DHS data for the three countries are available Measure DHS.

Conflicts of Interest:
The authors declare no conflict of interest.