Prevalence of undernourishment in adults in 195 countries and territories

Abstract

The prevalence of undernourishment (POU) is one of two primary indicators used to track progress towards sustainable development goal 2, which aims to end hunger and bolster food security by 2030. The United Nations Food and Agriculture Organization has historically been responsible for producing global estimates of the PoU in order to track progress towards the SDG target; however, the established methodology of PoU estimation is flawed across several dimensions, including no disaggregation by age or sex, a lack of estimated uncertainty, and, most importantly, the definition of an energy requirement that excludes a significant proportion of the population from being considered undernourished. In this study, we utilized copula methods to reconstruct the joint distribution of energy intake and the energy requirement for 195 countries by age and sex from 1990 to 2017 and compute the PoU accounting for undernourished individuals across the entire distribution of energy intake. We found that globally PoU in adults consistently declined from 9.9 (95% UI 9.5, 10.4)% in 1990 to 6.60 (95% UI 6.28, 6.81)% in 2017, though the effects of population growth have caused the number of undernourished adults to significantly increase over that period. We also found that by 2017 no countries to date had achieved SDG targets of 0% PoU and just under half were not trending downward. In comparing our estimates to those of the FAO, we found that ours were generally lower but show high correlation (ρ= 0.807). Sensitivity analyses showed that differences in inputs are a likely explanation and that our copula based method indeed accounted for a substantial number of additional undernourished individuals compared to the existing cutoff-based methods, holding inputs constant.