Household finished flooring and soil-transmitted helminth and Giardia infections among children in rural Bangladesh and Kenya: a prospective cohort study

Abstract

Background: Soil-transmitted helminths and Giardia duodenalis are responsible for a large burden of disease globally. In low-resource settings, household finished floors (eg, concrete floors) might reduce transmission of soil-transmitted helminths and G duodenalis.

Methods: In a prospective cohort of children nested within two randomised trials in rural Bangladesh and Kenya, we estimated associations between household finished flooring and soil-transmitted helminths and G duodenalis prevalence. In 2015-16, we collected stool samples from children aged 2-16 years in rural Bangladesh and Kenya. We detected soil-transmitted helminth infection using quantitative PCR (qPCR; Bangladesh n=2800; Kenya n=3094), and G duodenalis using qPCR in Bangladesh (n=6894) and ELISA in Kenya (n=8899). We estimated adjusted prevalence ratios (aPRs) using log-linear models adjusted for potential confounders.

Findings: 7187 (92·2%) of 7795 children in Bangladesh and 9077 (93·7%) of 9686 children in Kenya provided stool specimens that were analysed by qPCR. At enrolment, 691 (10%) households in Bangladesh and 471 (5%) households in Kenya had finished floors. In both countries, household finished flooring was associated with lower Ascaris lumbricoides prevalence (Bangladesh aPR 0·33, 95% CI 0·14-0·78; Kenya 0·62, 0·39-0·98) and any soil-transmitted helminths (Bangladesh 0·73, 0·52-1·01; Kenya 0·57, 0·37-0·88). Household finished floors were also associated with lower Necator americanus prevalence in Bangladesh (0·52, 0·29-0·94) and G duodenalis prevalence in both countries (Bangladesh 0·78, 0·64-0·95; Kenya 0·82, 0·70-0·97).

Interpretation: In low-resource settings, living in households with finished floors over a 2-year period was associated with lower prevalence of G duodenalis and some soil-transmitted helminths in children.

Funding: Bill & Melinda Gates Foundation and Task Force for Global Health.

Figure 1

Soil-transmitted helminth and Giardia duodenalis prevalence and unadjusted and adjusted prevalence ratios comparing infection at 2 year follow-up by household flooring status at enrolment Finished floor includes majority wood, concrete, or tile household floors; unfinished floors were made of soil or earth. In the plots, error bars indicate 95% CIs. PRs compared the prevalence of infection at 2 year follow-up in children with improved household flooring at enrolment with those with unimproved household flooring at enrolment. Adjusted PRs control for potential confounders associated with the outcome. CIs were adjusted for clustering at the village level. Ancylostoma duodenale was only detected in one sample in each country, so it is excluded from this figure. PR=prevalence ratio.

Figure 2

Soil-transmitted helminth and Giardia duodenalis Cq value and unadjusted and adjusted relative difference in Cq value among infected individuals at 2 year follow-up by household flooring status at enrolment Finished floor includes majority wood, concrete, or tile household floors; unfinished floors were made of soil or earth. Box plots show median (IQR), and whiskers show 1·5 times the IQR. Relative differences in Cq values were estimated as the ratio of the arithmetic mean Cq value at 2 year follow-up in children with improved household flooring at enrolment to those with unimproved household flooring at enrolment minus 1. Adjusted relative differences in Cq values control for potential confounders associated with the outcome. CIs were adjusted for clustering at the village level. Ancylostoma duodenale was only detected in one sample in each country, so it is excluded from this figure. Cq=cycle quantification.