Randomized trials
Did India’s toilet campaign improve child health?
Open access · cc by · source: Europe PMC
A village-randomized sanitation program raised latrine access and cut open defecation modestly but did not improve child diarrhea, parasites, anemia, or growth.
Study at a glance
- Design
- RCT — Cluster RCT of 80 villages: Total Sanitation Campaign support vs control
- N
- N=3390 · Baseline 3,390 children <5 from 1,954 households in 80 villages; ~21-month follow-up
- Population
- Children under five in rural Madhya Pradesh villages
- Outcome
- Sanitation facility use, open defecation, and child health endpoints
Structured fields used in claim comparison tables when every cited study has a complete layer.
Key findings
Intervention villages gained about 19 percentage points in improved sanitation facilities and a smaller (~10%) reduction in open defecation, but child health endpoints did not improve.
Methodology
Investigators cluster-randomized 80 villages in rural Madhya Pradesh to India’s Total Sanitation Campaign support versus control, surveying thousands of children under five at baseline and about 21 months later for sanitation behaviours and health outcomes.
Limitations
Short follow-up and some control-group contamination mean the trial cannot prove sanitation never helps—only that this scaled TSC package did not move health markers enough here.
How this study connects
Role on claims
Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.
Multiple studies in this library examine randomized trials with empirical patient or population outcomes rather than opinion alone.
Evidence for the claim as stated.
Intervention villages gained about 19 percentage points in improved sanitation facilities and a smaller (~10%) reduction in open defecation, but child health endpoints did not improve.
Evidence for the claim as stated.
Effect sizes and settings differ across randomized trials studies — digital vs clinic, trial vs observational — so results should not be pooled casually.
Evidence for the claim as stated.
India's Total Sanitation Campaign was cluster-randomised across 80 villages in rural Madhya Pradesh. Intervention villages gained about 19 percentage points in improved sanitation facilities and a smaller (~10%) reduction in open defecation, but child-health endpoints did not improve after about 21 months. Short follow-up and some control-group contamination mean the trial shows this scaled package did not move health markers here, not that sanitation can never help.
Evidence for the claim as stated.
Process success and health success are not the same cluster-trial result. TSC moved toilets (~19 percentage points) without moving child health; TB medication monitors moved adherence 40–50% without, in this report, proving cure; SASA! moved IPV and concurrency by about 50% as the intended social outcome. A student who treats 'the cluster intervention worked' as one sentence will misread at least one of these papers.
Evidence for the claim as stated.
Open questions
Tensions this paper is part of
From concept pages' “where studies disagree.” Disagreement means the same question; scope means different assays, populations, or outcomes.
Effect sizes and settings differ across randomized trials studies — digital vs clinic, trial vs observational — so results should not be pooled casually.
- Supports · Same-day ART start improves HIV outcomes
- Supports · Interdisciplinary frailty RCT
Process success and health success are not the same cluster-trial result. TSC moved toilets (~19 percentage points) without moving child health; TB medication monitors moved adherence 40–50% without, in this report, proving cure; SASA! moved IPV and concurrency by about 50% as the intended social outcome. A student who treats 'the cluster intervention worked' as one sentence will misread at least one of these papers.
- Supports · TB pillbox reminders beat SMS for doses
- Supports · SASA! cut partner violence in Kampala
Related papers in this topic
Same topic cluster — not a recommendation engine.