Community Health Workers, Subsidies and Safe Drinking Water: Experimental Evidence from Malawi

Last registered on April 17, 2018


Trial Information

General Information

Community Health Workers, Subsidies and Safe Drinking Water: Experimental Evidence from Malawi
Initial registration date
April 13, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
April 17, 2018, 2:06 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.


Primary Investigator

Princeton University

Other Primary Investigator(s)

PI Affiliation
Stanford University
PI Affiliation
Partners in Health

Additional Trial Information

On going
Start date
End date
Secondary IDs
Diarrheal disease is the second-leading cause of childhood mortality. In countries that cannot afford to provide piped water to dispersed, rural households, increasing access to safe drinking water remains a challenge. A recent paper in Science (Dupas et al., 2016) suggests that targeting coupons for water-treatment solution to mothers provides a particularly cost effective way to deliver the solution to households which need and will use the product but cannot afford it. In this study, we run a randomized controlled trial (RCT) to test the impact of the coupon system over time, test whether embedding the coupon system within a Community Health Work (CHW) program can enhance its impact over time, and compare the coupon system to a free delivery system implemented through trained CHW. We are partnering with Partners in Health which is running a CHW program in Neno district in Malawi.
External Link(s)

Registration Citation

Dupas, Pascaline, Zachary Wagner and Emily Wroe. 2018. "Community Health Workers, Subsidies and Safe Drinking Water: Experimental Evidence from Malawi." AEA RCT Registry. April 17.
Former Citation
Dupas, Pascaline, Zachary Wagner and Emily Wroe. 2018. "Community Health Workers, Subsidies and Safe Drinking Water: Experimental Evidence from Malawi." AEA RCT Registry. April 17.
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details


There will be three distinct interventions:
1) provision of monthly access to a free bottle of WaterGuard, a point-of-use water treatment, through vouchers (18 months)
2) Training CHWs to promote use of clean water.
3) Training CHWs to promote use of clean water and provision one free bottle of WaterGuard monthly, delivered at home by the CHW (for 18 months)

The details are as follows:

1) Coupons
Households assigned to this intervention will be provided with 18 coupons, each of which can be redeemed for one 150 ml bottle of WaterGuard. Only one coupon can be redeemed per month and each coupon corresponds to a unique month (i.e. 18 months’ worth of coupons). If the coupon is not used in the designated month, it expires. Coupons can be redeemed from local shops empaneled by the study.
Coupons will attached to a large, wall-poster type of calendar indicating which month the coupon can be redeemed. Households not sampled for coupons will also receive the same calendar, but there will be no coupons attached.

2) CHW Training Intervention
Some CHWs will be trained to promote clean water use. This intervention will take place at the community/CHW level, since CHWs are responsible for an entire community (roughly 40 households). PIH’s CHWs, which are only located in Neno district, will be asked to visit all households in their catchment area each month and to demonstrate how to use the water treatment solution and to encourage use of clean water.

3) A subset of trained CHWs will be instructed to make monthly deliveries of WaterGuard to all of the households in their catchment area.
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Use of chlorine to treat drinking water: This will be an objective binary measure captured during both baseline and follow-up. We will use a chlorine test kit to test for whether there are traces of chlorine present in each household's supply of drinking water.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
1. Coupon redemption: We will collect coupon redemption records from the shop owners who were enrolled to distribute WaterGuard in exchange for the coupons. Shop owners are instructed and paid to keep detailed record of each coupon redemption
2. Prevalence of child diarrhea: We will measure prevalence of child diarrhea using a survey question that asks caretakers if a child had a case of diarrhea in the last 4-weeks.
Secondary Outcomes (explanation)
For the diarrhea prevalence measure, we will test for robustness to different recall durations by constructing a 14-day recall measure, and 7-day recall measure.

Experimental Design

Experimental Design
Sampling, Enrollment, and Randomization
There are two study sites: Neno and Mwanza district.
Neno is where the PIH CHW program is running. Mwanza has no such program. As a result the sampling and enrollment processes will be different in the two sites.

Sampling frame: The sampling frame in Neno consists of all households with a child under 7 located in the catchment area of a PIH CHW. The sampling frame in Mwanza consists of all households in region villages of Mwanza that boarder Neno. A listing exercise was conducted to identify all households in the sampling frames.

Sample Selection: In Neno, 4 Households per CHW will be randomly selected for enrollment in the study. In Mwanza, villages will be separated into geographic clusters that represent a potential CHW catchment area. 4 households will then be selected from each cluster.

CHW Randomization: In Neno, each of the CHWs will be randomly assigned to one of the three groups prior to the baseline survey (free deliver+training, training, control).

Household Randomization: Households in Mwanza, and household in Neno whose CHW was not assigned to free delivery will be randomly assigned to one of the two WaterGuard interventions prior to the baseline survey (coupons vs. no coupons).

Caretakers of children under-7 will be enrolled by enumerators through household visits. If the household was assigned to the coupon intervention, all 18 coupons will be provided immediately during the baseline visit.
If a selected household refuses to participate or cannot be found, the enumerator will be prompted to enroll a pre-selected back-up household in the same CHW's catchment area to ensure we meet our sample size requirements.

Data Collection
Baseline: Upon enrollment, each household will be asked to complete a baseline survey, which will record information on water treatment practices and other health and demographic characteristics. In addition, we will test the main source of drinking water of the household for chlorine content during these visits.

We will conduct follow-up visits once per month for no less than 18 months and as long as funding permits following enrollment. Each month, a random subset of households will be visited and the chlorine content of their main drinking source will be tested, and they will complete a follow-up survey. This will give us time series data, which will allow for assessment of how the treatment effect changes over time .
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
The coupon intervention will be randomized at the household level. The free delivery+training and CHW training interventions will be randomized at the CHW level.
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
Neno: 436 CHWs
Mwanza: 424 households
Sample size: planned number of observations
2132 Households
Sample size (or number of clusters) by treatment arms
Average of 4 households per CHW in Neno
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Institutional Review Boards (IRBs)

IRB Name
Malawi National Health Sciences Research committee
IRB Approval Date
IRB Approval Number
IRB Name
Stanford IRB
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials