Using structural equation modelling to untangle sanitation, water and hygiene pathways for intervention improvements in height-for-age in children <5 years old

• Type of Material: Article
• Authors: Heather Reese, Emory University; Sheela S Sinharoy, Emory University; Thomas Clasen, Emory University
• Language: English
• Date: 2019-12-01
• Publisher: Oxford University Press (OUP): Policy B - Oxford Open Option D
• Publication Version: Final Published Version
• Copyright Statement: © The Author(s) 2019. Published by Oxford University Press on behalf of the International Epidemiological Association.
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1093/ije/dyz202
• Title of Journal or Parent Work: International Journal of Epidemiology
• ISSN: 0300-5771
• Volume: 48
• Issue: 6
• Start Page: 1992
• End Page: 2000
• Grant/Funding Information: This work was supported by the Bill & Melinda Gates Foundation to the London School of Hygiene & Tropical Medicine [grant number OPP1008048]; and by Emory University [grant number OOP1125067].
• Keywords: stunting; India; hygiene; path analysis; Water and sanitation
• Research Categories: Environmental Sciences; Health Sciences, Public Health

Abstract

BACKGROUND: Despite a strong theoretical rationale for combining water, sanitation and hygiene (WaSH) interventions to improve child health, study findings are heterogeneous with little understanding of the mechanisms for these effects. Our study objective was to demonstrate the utility of structural equation modeling (SEM) to assess intervention effects on height-for-age z score (HAZ) through the complex system of WaSH pathways. METHODS: We used data from a matched cohort effectiveness evaluation of a combined on-premise piped water and improved sanitation intervention in rural Odisha, India. Height/length was measured in children 0-59 months old (n = 1826) from 90 matched villages in February-June 2016. WaSH behaviours and infrastructure were assessed through household surveys and observation, respectively. We used SEM to calculate the standardized path coefficients and the total contributions of WaSH pathways to HAZ. RESULTS: Intervention improvements on HAZ were through the sanitation pathway (coverage → use β: 0.722; use → HAZ β: 0.116), with piped water coverage indirectly affecting HAZ through improved sanitation use (β: 0.148). Although the intervention had a positive association with handwashing station coverage, there was no evidence of a total hygiene pathway effect on HAZ or further direct effects through the water pathways. CONCLUSIONS: This study demonstrates the utility of SEM to assess the mechanisms through which combined WaSH interventions impact HAZ as a system of pathways, providing a more nuanced assessment than estimation of the total intervention effect. Our finding, that water impacts HAZ through the sanitation pathway, is an important and actionable insight for WaSH programming.