The Double Burden of Malnutrition: When Hunger and Obesity Coexist

In 2022, the South African National Income Dynamics Study - Coronavirus Rapid Mobile Survey recorded that 23.9 per cent of children under five in the rural Eastern Cape were stunted - a marker of chronic undernutrition - whilst 43 per cent of adult women in the same province were classified as overweight or obese.¹ These figures do not describe two crises unfolding in separate populations. They describe a single community, frequently a single household, in which a grandmother living with overweight-associated hypertension prepares meals for a grandchild whose linear growth has been permanently compromised by early dietary insufficiency. This is not a contradiction. It is the defining nutritional signature of the contemporary low- and middle-income country, and South Africa represents one of its starkest expressions.

South Africa is far from unique. A nationally representative household survey conducted in Ghana between 2014 and 2018 found that 19 per cent of under-five children exhibited stunting, whilst the prevalence of overweight and obesity among women aged 15–49 had climbed to 40 per cent - a figure nearly double that recorded two decades earlier.² Across Sub-Saharan Africa (SSA) more broadly, the epidemiological portrait is one of unresolved undernutrition fusing with an accelerating overnutrition epidemic - not sequentially, as linear developmental models once projected, but concurrently, within the same social and geographic spaces. This article examines the mechanisms driving that convergence, the evidence base for its household-level manifestations, and the structural deficiencies in the policy responses mounted against it.

Defining the Double Burden of Malnutrition

The term double burden of malnutrition refers to the coexistence of undernutrition - encompassing stunting, wasting, micronutrient deficiency, and inadequate energy intake - alongside overnutrition, expressed as overweight, obesity, and diet-related non-communicable diseases (NCDs). The concept operates at multiple scales: it can describe a national population in which both conditions are statistically prevalent; a community in which undernourished and overnourished individuals reside in proximity; or a single household in which members at different life stages present with opposing nutritional pathologies.

The World Health Organization’s 2017 policy brief was among the first authoritative statements to acknowledge explicitly that the traditional framing of malnutrition - as synonymous with undernutrition, with overnutrition treated as a separate “affluence” problem - had become operationally untenable.³ The WHO articulated a fourfold typology: undernutrition and overnutrition coexisting within countries, within communities, within households, and within individuals across the life course - as in the well-documented phenomenon by which adults stunted in childhood carry heightened risk of abdominal obesity and metabolic syndrome.⁴

The double burden is not only a spatial coincidence of opposites; it is, in part, a temporal one. Victora and colleagues’ 2008 Lancet cohort analysis - drawing on data from Brazil, Guatemala, India, the Philippines, and South Africa - demonstrated that individuals who experienced early-life growth faltering carried significantly elevated risks of type 2 diabetes, cardiovascular disease, and obesity in adulthood, particularly when they subsequently encountered energy-dense dietary environments.⁴ Undernutrition does not merely precede overnutrition in a developmental sequence; it can actively programme metabolic vulnerability to it.

Mechanisms Driving the Double Burden

The Nutrition Transition

The double burden cannot be understood without reference to Popkin’s nutrition transition framework, which describes the shift from traditional, largely plant-based dietary patterns associated with physical labour towards diets high in refined carbohydrates, saturated fats, and added sugars - accompanied by progressively more sedentary lifestyles.⁵ Most low- and middle-income countries now occupy a stage characterised by rapid dietary change driven by urbanisation, income growth, and the globalisation of food systems.

The transition is neither smooth nor uniform within populations. In countries undergoing rapid economic transformation, a dietary bifurcation frequently occurs: wealthier urban households shift quickly towards energy-dense, nutrient-poor diets and accumulate overweight, whilst poorer households retain insufficient energy intake alongside persistent micronutrient deficiencies. Both pathologies come to concentrate disproportionately in low-income groups, expressing themselves differently by age, sex, and geography - but no longer mapping onto a simple “poor = underweight, wealthy = overweight” gradient.

Urban Migration and the Restructuring of Dietary Environments

Urban migration is one of the primary social vectors of the double burden. When rural households move to urban centres in SSA, they enter profoundly altered food environments. Traditional systems that supported dietary diversity through small-scale production are abandoned. Urban food access becomes mediated by cash income and the architecture of periurban retail - dominated, in the settlements of Nairobi, Lagos, Accra, and Johannesburg, by informal vendors selling energy-dense prepared foods, packaged snacks, and sweetened beverages.

Ultra-processed foods (UPFs) are central to this shift. UPFs are industrially formulated products - containing emulsifiers, hydrogenated fats, flavour enhancers, and colourants absent from domestic kitchens - designed for palatability, convenience, and long shelf life. Their nutritional profile is notably problematic: high in refined starches and added sugars, low in fibre, low in most micronutrients. Dietary surveillance data across SSA document rapid increases in UPF consumption over the past two decades, particularly in urban populations. Their consumption produces caloric adequacy with nutritional poverty - sufficient energy to forestall classical undernutrition, but insufficient micronutrient density to eliminate hidden hunger, alongside a caloric surplus that accumulates as adipose tissue.

This dynamic is especially consequential for adolescent girls and women of reproductive age. A young woman consuming predominantly UPFs may carry excess adiposity by BMI whilst remaining profoundly deficient in iron, folate, zinc, and vitamin A. Her children may be born with compromised intrauterine growth, become stunted in early childhood, and subsequently encounter the same obesogenic environment - perpetuating the burden across generations.

The Household-Level Double Burden

The household is perhaps the most analytically revealing unit at which the double burden manifests. Doak and colleagues’ pioneering 2005 analysis of nationally representative data from eight low- and middle-income countries - including Brazil, Russia, China, and South Africa - demonstrated that households containing at least one underweight member alongside at least one overweight member were not a statistical anomaly but a quantitatively significant phenomenon, accounting for 14 to 20 per cent of households in several study countries.⁶ Critically, such households were not concentrated in the poorest income quintile; those in the middle of the income distribution were disproportionately represented, suggesting that neither extreme poverty nor relative affluence was a reliable protective factor against the simultaneous presence of both conditions.

Tzioumis and Adair’s 2014 systematic review synthesised data from 42 studies across 31 countries, confirming that the within-household double burden was a consistent, cross-contextually validated phenomenon rather than an artefact of particular national datasets.⁷ Their analysis identified several structural correlates: the urban–rural interface, where one household member may retain traditional dietary patterns whilst another has adopted a more Westernised diet; the age differential between members; and maternal overweight alongside child stunting as the modal expression of the within-household double burden across SSA contexts.

The maternal–child dyad is particularly significant. Ruel and colleagues’ 2013 Lancet analysis of nutrition-sensitive agriculture programmes noted that maternal nutritional status at conception is among the strongest determinants of child birth outcomes - yet child undernutrition interventions and those addressing maternal overweight have historically been architecturally separate, administered by different agencies, with different monitoring indicators and different political constituencies.⁸ A child’s stunting is measured by a community health worker; the mother’s overweight may never be systematically addressed at all.

Sub-Saharan Africa: The Epidemiological Particulars

The burden of both undernutrition and overnutrition in SSA is substantial and, in the case of overweight and obesity, rapidly expanding. Black and colleagues’ 2013 Lancet series estimated that stunting, severe wasting, and intrauterine growth restriction collectively contributed to approximately 3.1 million child deaths per year globally, with SSA carrying a disproportionate share of mortality and morbidity attributable to undernutrition.⁹ Yet the NCD Risk Factor Collaboration’s 2017 analysis - drawing on data from 186 countries and 128.9 million individuals - documented a tripling of global obesity prevalence between 1975 and 2016, with some of the steepest absolute BMI increases occurring in Sub-Saharan African populations.¹⁰

In SSA specifically, overnutrition has a notable demographic structure. Overweight and obesity rates are consistently higher among women than men, higher in urban than rural settings, and increasingly concentrated in middle-income rather than exclusively high-income households. The 2019 UNICEF nutritional status data for Eastern and Southern Africa indicated that whilst 36 per cent of children under five remained stunted, overweight prevalence among women aged 20–49 exceeded 35 per cent in several countries, including Lesotho, Eswatini, and South Africa.

Ethiopia is instructive. Historically regarded as a quintessential undernutrition context - and rightly so, with national child stunting rates above 37 per cent - its 2019 Demographic and Health Survey revealed that overweight prevalence among urban women had reached 27 per cent, against a national average just above 8 per cent. The urban–rural differential illustrates how rapidly dietary environments diverge even within a single country during economic transition, manifesting not as a resolution of undernutrition but as its stratification alongside an emerging overnutrition burden.

For a broader analytical context on how food security frameworks interact with these data patterns, see our comparative analysis of food security frameworks across regions .

The Policy Paradox: Structural Blindness to Overnutrition

Conventional malnutrition programmes in SSA are, by and large, structurally blind to the overnutrition arm of the double burden. This is not merely an observation about resource allocation, though the resource differential is real and striking; it is a claim about institutional design, measurement architecture, and the political economy of international nutrition funding.

The dominant paradigms of international nutrition programming have, for most of the past three decades, been organised around reducing child stunting, wasting, and micronutrient deficiency. The World Bank’s Stunting Reduction Dashboard, the UNICEF nutrition country profiles, and the Scaling Up Nutrition (SUN) Movement’s country commitments all prioritise undernutrition indicators. Community-based management of acute malnutrition (CMAM), therapeutic feeding protocols, vitamin A supplementation, and iron-fortification initiatives are genuinely valuable, evidence-based interventions - but they address structurally only one pole of the double burden.

The overnutrition arm is either absent from these frameworks or treated as an appended concern managed by a different sector - typically the NCD division of ministries of health, where funding is chronically insufficient and primary care capacity to manage obesity-related conditions is limited. In practice, a stunted child who recovers to adequate height-for-age and subsequently enters an obesogenic food environment in adolescence falls out of the malnutrition surveillance apparatus at the precise moment when a new vulnerability is being created. The very success metric of the intervention - recovery from stunting - terminates the surveillance relationship too early.

This structural blindness is reproduced at the level of dietary assessment. Most community-level nutrition surveys in SSA are designed to capture dietary diversity as a proxy for micronutrient adequacy - an entirely legitimate goal - but are not equipped to assess UPF intake, added sugar consumption, or the energy density of the overall dietary pattern. A household achieving a satisfactory dietary diversity score through processed maize flour, tinned fish, fortified biscuits, and sweetened tea has met the measurement criterion for nutritional adequacy whilst consuming a diet that carries significant overnutrition risk over the medium term. The instrument does not see what it was not designed to measure.

The implications are programmatic. Nutrition-sensitive agriculture programmes - such as those reviewed by Ruel and colleagues - show meaningful reductions in child stunting and maternal micronutrient deficiency when well implemented.⁸ Yet their logic models are constructed around undernutrition pathways and typically incorporate no dietary guidance on limiting UPF consumption or managing energy density. As long as governing logic models treat overnutrition as external to the malnutrition agenda, interventions will systematically fail to address the arm of the double burden that is growing fastest.

For a discussion of micronutrient intervention frameworks that begin to grapple with this architecture, see our analysis of micronutrient intervention strategies and their limitations .

Adolescents and the Under-Measured Population

Adolescents - aged 10–19 - represent a population of particular salience for the double burden in SSA, yet they are chronically under-measured in national nutrition surveillance. Most national surveys collect anthropometric data on children under five and, to varying degrees, on women of reproductive age; adolescent boys and girls outside those categories are frequently absent from systematic monitoring.

Adolescence is a period of rapid physical growth with elevated nutritional requirements, and it is also a life stage of intense dietary transition in SSA’s urban environments. Adolescents are avid consumers of UPFs, street foods, and sweetened beverages - partly for palatability and affordability, partly for the social meanings attached to “modern” foods in peer contexts. The simultaneous presence of stunting (a legacy of early childhood undernutrition) alongside rapid weight gain creates precisely the metabolic configuration - short stature, elevated abdominal adiposity, insulin resistance - that Victora and colleagues identified as carrying the highest cardiovascular and metabolic risk in later life.⁴

Monitoring systems that fail to capture adolescent nutritional status are failing to detect the population in which the double burden’s next generational wave is forming. For a broader examination of how public health monitoring systems have evolved in response to such surveillance gaps, see our analysis of the evolution of public health monitoring in low-income settings .

Limitations and Methodological Considerations

Several important caveats attend the evidence base reviewed in this article. First, the within-household double burden literature relies heavily on cross-sectional survey data, which precludes causal inference about the direction of nutritional change within households. Co-occurrence of undernutrition and overnutrition does not, in itself, establish that the two conditions share a common aetiology or that a single intervention can address both.

Second, measuring overnutrition in SSA populations raises substantive challenges not always acknowledged in the epidemiological literature. BMI is a crude proxy for adiposity, particularly in populations where stunting is prevalent: a stunted individual may have a BMI in the “normal” range whilst carrying excess visceral adiposity relative to lean mass. Waist circumference and waist-to-height ratio are more sensitive indicators of cardiometabolic risk, but their deployment in large-scale SSA surveys remains inconsistent. Published obesity prevalence figures may therefore systematically underestimate population-level cardiometabolic risk.

Third, UPF intake data across SSA are sparse and methodologically inconsistent. The NOVA classification system, whilst increasingly standard in research contexts, requires detailed dietary recall data that most national surveys in the region do not collect. Estimates of UPF contribution to total energy intake in SSA are extrapolated from relatively small, subnational, or context-specific datasets - a limitation that should temper confidence in specific claims about the pace of the UPF transition in particular country contexts.

Fourth, attribution of the double burden to the nutrition transition, urban migration, and UPF exposure does not adequately account for the independent roles of infectious disease burden, psychosocial stress, and physical activity environments. The epidemiological framework is necessarily simplifying; it should not be mistaken for a complete causal account.

Towards an Integrated Policy Response

The evidence reviewed above points towards a set of programmatic redirections, none of which require abandoning the hard-won achievements of existing undernutrition programming. The task is integration, not substitution. National nutrition strategies should incorporate explicit targets and monitoring indicators for overweight and obesity alongside undernutrition metrics, enabling health systems to track the double burden’s full dimensionality. Dietary guidance within community nutrition programmes should address UPF consumption as a risk factor - not merely dietary diversity as a protective one.

At the level of food systems, fiscal and regulatory instruments - including taxes on sugar-sweetened beverages, restrictions on UPF marketing to children, and front-of-pack labelling - represent evidence-informed policy levers whose deployment in SSA lags significantly behind high-income country practice. South Africa’s 2018 health promotion levy on sugar-sweetened beverages is an encouraging precedent, with early evaluation data cautiously promising. But regulatory reform of food environments is politically contested and requires sustained institutional commitment that donor-funded, project-based nutrition programming cannot reliably generate.

Adolescent nutrition must be moved from the margins of national surveillance and programming to the centre. If the double burden’s next cohort is forming in the bodies of adolescents navigating obesogenic urban environments whilst carrying the metabolic legacy of early undernutrition, surveillance systems that systematically exclude this age group are not merely incomplete - they are structurally guaranteeing that the epidemiological problem they purport to address will persist into the next generation.

Frequently Asked Questions

What is the double burden of malnutrition? The double burden of malnutrition describes the simultaneous coexistence of undernutrition - including stunting, wasting, and micronutrient deficiency - and overnutrition, expressed as overweight, obesity, and diet-related non-communicable diseases, within the same country, community, household, or individual across the life course. It is particularly prevalent in low- and middle-income countries undergoing rapid economic and dietary transition.

How can stunting and obesity occur within the same household? Within-household double burden typically occurs when household members at different life stages face different nutritional exposures. Young children may be stunted due to early childhood dietary insufficiency, inadequate breastfeeding, or recurrent infection, whilst adults - particularly mothers - accumulate excess adiposity through consumption of energy-dense, nutrient-poor diets. The phenomenon has been documented in nationally representative data from across Sub-Saharan Africa, South Asia, and Latin America, and is not a statistical curiosity but a quantitatively significant population-level pattern.

Why is Sub-Saharan Africa particularly affected by the double burden of malnutrition? SSA is experiencing a rapid nutrition transition in which traditional, predominantly plant-based diets are being displaced by ultra-processed foods and energy-dense dietary patterns - particularly in urban and periurban settings - whilst undernutrition, driven by poverty, poor water and sanitation infrastructure, and inadequate dietary diversity, remains unresolved in rural areas and among the poorest urban households. The resulting dietary bifurcation, combined with demographic structures in which young children and adult women are concentrated in the same households, makes SSA particularly prone to within-household double burden.

What should nutrition programmes do differently to address the double burden? Programmes need to integrate monitoring and intervention capacity across the full spectrum of malnutrition rather than treating undernutrition and overnutrition as separate mandates managed by separate sectors. This requires incorporating overweight and obesity indicators into national surveillance systems, redesigning dietary guidance to address ultra-processed food consumption as well as dietary diversity, extending systematic nutritional monitoring to adolescent populations, and supporting regulatory reform of food environments through fiscal and labelling instruments. Donor frameworks must create incentives for this integration rather than rewarding narrow programme performance on single-pole malnutrition metrics.

References