Abstract
Many water, sanitation, and hygiene (WASH) interventions target improvements in personal hygiene behaviors. Yet measuring personal hygiene behaviors is a challenge due to a lack of reliable, valid, objective, and simple-to-use approaches. The purpose of this study was to examine differences between two types of hygiene outcome measures and their ability to detect relationships between WASH-related behavioral factors and behaviors. We compared hygiene outcomes generated by the Quantitative Personal Hygiene Assessment Tool (qPHAT), which yields objective measures of cleanliness on an 11-point scale, and those generated by conventional, dichotomous indicators of cleanliness. We used cross-sectional data on hygiene outcomes related to facial and hand cleanliness collected during the Andilaye Trial, an impact evaluation of a community-based WASH intervention implemented in Amhara, Ethiopia. We fit multivariable models to examine associations between measures of children's facial and hand cleanliness, via both qPHAT and dichotomous indicators, and 1) household WASH conditions, 2) psychosocial factors, and 3) reported personal hygiene practices. The qPHAT-generated outcomes were able to detect relationships between intermediate behavioral factors and hygiene outcomes that dichotomous indicators were not, including associations with water insecurity and various psychosocial factors. qPHAT-generated outcomes were negatively associated with reported face washing practices, suggesting a bias in reported behaviors. Our study highlights the limitations of reported practices and dichotomous hygiene indicators and indicates that using more quantitative hygiene outcome measures, such as those generated by qPHAT, may reveal important intermediate factors that influence hygiene behavior and support improved monitoring and evaluation of interventions.