Children's height and weight in rural and urban populations in low-income and middle-income countries: a systematic analysis of population-representative data

Abstract

Background: Urban living affects children's nutrition and growth, which are determinants of their survival, cognitive development, and lifelong health. Little is known about urban-rural differences in children's height and weight, and how these differences have changed over time. We aimed to investigate trends in children's height and weight in rural and urban settings in low-income and middle-income countries, and to assess changes in the urban-rural differentials in height and weight over time.

Methods: We used comprehensive population-based data and a Bayesian hierarchical mixture model to estimate trends in children's height-for-age and weight-for-age Z scores by rural and urban place of residence, and changes in urban-rural differentials in height and weight Z scores, for 141 low-income and middle-income countries between 1985 and 2011. We also estimated the contribution of changes in rural and urban height and weight, and that of urbanisation, to the regional trends in these outcomes.

Findings: Urban children are taller and heavier than their rural counterparts in almost all low-income and middle-income countries. The urban-rural differential is largest in Andean and central Latin America (eg, Peru, Honduras, Bolivia, and Guatemala); in some African countries such as Niger, Burundi, and Burkina Faso; and in Vietnam and China. It is smallest in southern and tropical Latin America (eg, Chile and Brazil). Urban children in China, Chile, and Jamaica are the tallest in low-income and middle-income countries, and children in rural areas of Burundi, Guatemala, and Niger the shortest, with the tallest and shortest more than 10 cm apart at age 5 years. The heaviest children live in cities in Georgia, Chile, and China, and the most underweight in rural areas of Timor-Leste, India, Niger, and Bangladesh. Between 1985 and 2011, the urban advantage in height fell in southern and tropical Latin America and south Asia, but changed little or not at all in most other regions. The urban-rural weight differential also decreased in southern and tropical Latin America, but increased in east and southeast Asia and worldwide, because weight gain of urban children outpaced that of rural children.

Interpretation: Further improvement of child nutrition will require improved access to a stable and affordable food supply and health care for both rural and urban children, and closing of the the urban-rural gap in nutritional status.

Funding: Bill & Melinda Gates Foundation, Grand Challenges Canada, UK Medical Research Council.

Figure 1

Children's mean height-for-age and weight-for-age Z scores in rural and urban areas of low-income and middle-income countries in 2011 A population with a mean height-for-age or weight-for-age Z score of zero would have the same average height or weight as a well-nourished WHO reference population. Negative numbers indicate countries that are on average shorter or more underweight than the reference population, whereas positive numbers indicate those that are taller or heavier than the reference population.

Figure 2

Urban advantage in children's height-for-age and weight-for-age Z scores, by country, in 2011 Larger numbers indicate greater differences in height or weight between urban and rural children. For height, one Z score is equivalent to about 3·2 cm of height at age 2 years and 4·7 cm at age 5 years. For weight, one Z score is equivalent to about 1·4 kg of weight at age 2 years and 2·6 kg at age 5 years. The appendix (pp 42–43) reports urban–rural differences in the prevalences of stunting and underweight.

Figure 3

Change in urban advantage in children's height-for-age and weight-for-age Z scores between 1985 and 2011, by region A point above the diagonal line means a larger improvement in urban areas than in rural areas; a point below the diagonal line indicates the opposite. The vertical and horizontal error bars show the 95% credible intervals. The figure excludes Oceania because its large credible intervals would reduce the visibility of data from other regions. The urban–rural differential in height-for-age Z score in Oceania was 0·56 (0·11–1·05) in 1985 and 0·57 (0·24–0·91) in 2011. The urban–rural differential in weight-for-age Z score in Oceania was 0·38 (0·12–0·68) in 1985 and 0·37 (0·15–0·61) in 2011. The global advantage is larger than all region-specific advantages because regions such as southern and tropical Latin America where children are tallest and heaviest are also those that are more heavily urbanised, whereas regions with the shortest and most underweight child populations are those that are more rural, such as south Asia (ie, the global results encompass both within-region and between-region differences).

Figure 4

Trends in rural and urban mean height-for-age and weight-for-age Z score, by region The solid line represents the posterior mean and the shaded area the 95% uncertainty interval. The appendix reports trends in prevalences of stunting and underweight by region (pp 44–45), and trends in rural and urban mean height-for-age and weight-for-age Z scores and prevalences of stunting and underweight by country (pp 46–188).

Figure 5

Number of stunted and underweight children, by region and rural or urban place of residence, 1985–2011 Stunted is defined as a height-for-age Z score lower than −2. Underweight is defined as weight-for-age Z score lower than −2.