Iván Mejía-Guevara1,2, Daniel J Corsi3, Jessica M Perkins4,5, Rockli Kim6, S V Subramanian7,8. 1. Department of Biology, Stanford University, Palo Alto, California. 2. Stanford Center for Population Health Sciences, School of Medicine, Stanford University, Stanford, California. 3. OMNI Research Group, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada. 4. Department of Human and Organizational Development, Peabody College, Vanderbilt University, Nashville, Tennessee. 5. Vanderbilt Institute of Global Health, Vanderbilt University Medical Center, Nashville, Tennessee. 6. Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston, Massachusetts; and. 7. Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston, Massachusetts; and svsubram@hsph.harvard.edu. 8. Harvard Center for Population and Development Studies, Harvard University, Cambridge, Massachusetts.
Abstract
BACKGROUND: Addressing anthropometric failure in low- and middle-income countries can have 2 targets of inference: addressing differences between individuals within populations (Wpop) or differences between populations (Bpop). We present a multilevel framework to apply both targets of inference simultaneously and quantify the extent to which variation in anthropometric status and growth failure is reflective of undernourished children or undernourished populations. METHODS: Cross-sectional data originated from the Demographic and Health Surveys program, covering children under age 5 from 57 countries surveyed between 2001 and 2015. RESULTS: A majority of variation in child anthropometric status and growth failure was attributable to Wpop-associated differences, accounting for 89%, 83%, and 85% of the variability in z scores for height for age, weight for age, and weight for height. Bpop-associated differences (communities, regions, and countries combined) were associated with 11%, 17%, and 15% of the variation in height-for-age z score, weight-for-age z score, and weight-for-height z score. Prevalence of anthropometric failure was closely correlated with mean levels of height and weight. Approximately 1% of Wpop variability, compared with 30% to 50% of the Bpop variability, was explained by mean values of maternal correlates of anthropometric status and failure. Although there is greater explanatory power Bpop, this varied because of modifiability of what constitutes population. CONCLUSIONS: Our results suggest that universal strategies to prevent future anthropometric failure in populations combined with targeted strategies to address both the impending and existing burden among children are needed.
BACKGROUND: Addressing anthropometric failure in low- and middle-income countries can have 2 targets of inference: addressing differences between individuals within populations (Wpop) or differences between populations (Bpop). We present a multilevel framework to apply both targets of inference simultaneously and quantify the extent to which variation in anthropometric status and growth failure is reflective of undernourished children or undernourished populations. METHODS: Cross-sectional data originated from the Demographic and Health Surveys program, covering children under age 5 from 57 countries surveyed between 2001 and 2015. RESULTS: A majority of variation in child anthropometric status and growth failure was attributable to Wpop-associated differences, accounting for 89%, 83%, and 85% of the variability in z scores for height for age, weight for age, and weight for height. Bpop-associated differences (communities, regions, and countries combined) were associated with 11%, 17%, and 15% of the variation in height-for-age z score, weight-for-age z score, and weight-for-height z score. Prevalence of anthropometric failure was closely correlated with mean levels of height and weight. Approximately 1% of Wpop variability, compared with 30% to 50% of the Bpop variability, was explained by mean values of maternal correlates of anthropometric status and failure. Although there is greater explanatory power Bpop, this varied because of modifiability of what constitutes population. CONCLUSIONS: Our results suggest that universal strategies to prevent future anthropometric failure in populations combined with targeted strategies to address both the impending and existing burden among children are needed.
Authors: Iván Mejía-Guevara; Beniamino Cislaghi; Ann Weber; Emma Hallgren; Valerie Meausoone; Mark R Cullen; Gary L Darmstadt Journal: J Glob Health Date: 2020-06 Impact factor: 4.413
Authors: Santu Ghosh; Nirupama Shivakumar; Sulagna Bandyopadhyay; Harshpal S Sachdev; Anura V Kurpad; Tinku Thomas Journal: BMC Public Health Date: 2020-11-01 Impact factor: 3.295
Authors: Margit Steinholt; Sam Ol Ha; Chandy Houy; Jon Øyvind Odland; Maria Lisa Odland Journal: Int J Environ Res Public Health Date: 2019-10-29 Impact factor: 3.390