The double burden of malnutrition: aetiological pathways and consequences for health

Abstract

Malnutrition has historically been researched and addressed within two distinct silos, focusing either on undernutrition, food insecurity, and micronutrient deficiencies, or on overweight, obesity, and dietary excess. However, through rapid global nutrition transition, an increasing proportion of individuals are exposed to different forms of malnutrition during the life course and have the double burden of malnutrition (DBM) directly. Long-lasting effects of malnutrition in early life can be attributed to interconnected biological pathways, involving imbalance of the gut microbiome, inflammation, metabolic dysregulation, and impaired insulin signalling. Life-course exposure to early undernutrition followed by later overweight increases the risk of non-communicable disease, by imposing a high metabolic load on a depleted capacity for homoeostasis, and in women increases the risk of childbirth complications. These life-course trajectories are shaped both by societal driving factors-ie, rapidly changing diets, norms of eating, and physical activity patterns-and by broader ecological factors such as pathogen burden and extrinsic mortality risk. Mitigation of the DBM will require major societal shifts regarding nutrition and public health, to implement comprehensive change that is sustained over decades, and scaled up into the entire global food system.

Figure 1. Complex interconnections between intergenerational cycles of undernutrition and nutritional excess and the effect of nutrition transition

The intergenerational cycle of undernutrition (blue) associated with energy-inadequate diets and micronutrient deficiencies constrains growth and reduces the metabolic capacity for homoeostasis. The intergenerational cycle of overnutrition (red) associated with energy-dense diets is characterised by excess metabolic fuel and elevated adiposity, each of which challenges homoeostasis. Both cycles of malnutrition contribute to a wide range of adverse health outcomes (grey boxes), and specific diseases also increase the risk of malnutrition (black arrows). Through nutrition transition, individuals shift between these cycles within the life course, both increasing the risk and exacerbating the magnitude of health consequences. This framework helps identify how nutrition transition generates biological connections between many different forms of ill health (eg, low birthweight, stunting, central obesity, diabetes, and caesarean delivery).

Figure 2. Risk of stunting and caesarean delivery according to maternal phenotype in selected LMIC populations relative to mothers of normal height and normal BMI

(A) Associations between stunting and maternal nutritional phenotype in 12 populations. (B) Associations between caesarean section and maternal nutritional phenotype in 11 populations. Based on Demographic and Health Survey data. All models adjust for wealth, parity, and offspring sex. Full details are given in the appendix (p 21). BMI=body-mass index. LMIC=low-income and middle-income countries. OR=odds ratio.

Figure 3. Rapid transition and evolution of the double burden of malnutrition in Pune, India, over the past 40 years, based on data from rural and urban cohorts

Average age in index pregnancy, height, and BMI of mothers and fathers are shown for three cohorts set up in Pune, India in the 1980−90s.^– Rural mothers (F_o generation, cohort 1: PMNS, recruited 1993−96) show the legacy of multigenerational undernutrition (stunted and underweight, low energy intake and excess physical activity from subsistence farming, multiple micronutrient deficiencies, and low prevalence of gestational diabetes). Parents in the contemporary urban cohort (cohort 2: PCS, recruited 1987−89) were somewhat taller and had a higher BMI. Parents in the diabetic pregnancy cohort (cohort 3: GDM, recruited 1990−2010) were the tallest and heaviest, almost half being overweight or obese (BMI>25 kg/m²), reflecting the effects of socioeconomic transition. F₁ babies born to cohort 1 mothers had low average weight and a characteristic thin-fat composition (ie, low lean mass but high fat mass compared with European babies). F₁ babies in cohort 2 had somewhat higher birthweight but 35% were still SGA by INTERGROWTH criteria. Babies born to mothers in cohort 3 were the heaviest (11% LGA), but 22% were still SGA. F₁ young adults in cohort 1 were taller and heavier than their parents, although still thin (low BMI) but adipose (high body fat). F₂ babies were 200 g heavier at birth than their mothers’ birthweight, highlighting an intergenerational effect of the DBM. In cohort 2, F₁ children were similarly taller and heavier than their parents, with 75 (21%) of 357 being overweight or obese. In cohort 3, these intergenerational effects were more marked, with 48 (24%) of 200 F₁ children being overweight or obese. BMI=body-mass index. DBM=double burden of malnutrition. GDM=gestational diabetes. LGA=large for gestational age. NA=not available. PCS=Pune Children’s Study. PMNS=Pune Maternal Nutrition Study. SGA=small for gestational age. *Husbands of women with gestational diabetes.