Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
The effect of public handwashing stations on health behaviour and outcomes during COVID-19
Last registered on April 15, 2021


Trial Information
General Information
The effect of public handwashing stations on health behaviour and outcomes during COVID-19
Initial registration date
December 22, 2020
Last updated
April 15, 2021 5:46 AM EDT

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
BRAC Institute of Governance and Development
Other Primary Investigator(s)
PI Affiliation
The Behavioural Insights Team
PI Affiliation
BRAC Institute of Governance and Development
PI Affiliation
BRAC Institute of Governance and Development
PI Affiliation
BRAC Institute of Governance and Development
PI Affiliation
BRAC Institute of Governance and Development
Additional Trial Information
On going
Start date
End date
Secondary IDs
In line with BRAC's overall goals to limit the COVID19 transmission, BRAC has constructed 1,000 handwashing stations coupled with other supports i.e. in-person demonstration, hygiene meetings, and soap distribution in Bangladesh. The purpose of this study is to assess the impact of public handwashing stations (HWSs) on people’s hygiene behavior and health outcomes in response to the COVID-19 situation. For this, at the household level, we will collect information on self-reported handwashing practices, income, health status, and HWSs along with other relevant socio-demographic variables in detail. In addition, we will observe people’s mobility, mask usage at public places, and the HWSs usage; and will associate the observed HWSs usage with their self-reported health outcomes. This study will conduct surveys on households (7,760) and public (5,820) in 20 sub-districts of Mymensingh, Khulna, and Dhaka divisions.
Overall, this study is expected to directly benefit study participants and their communities by unveiling new evidence to inform BRAC’s existing hygiene behavioral change programming in these communities. In addition, this information will yield future societal benefits by generating evidence on how to effectively implement and disseminate infrastructure and information that people can trust, believe, and use to form healthy hygiene habits.
Registration Citation
Afrin, Sonia et al. 2021. "The effect of public handwashing stations on health behaviour and outcomes during COVID-19." AEA RCT Registry. April 15. https://doi.org/10.1257/rct.6962-4.0.
Sponsors & Partners
Experimental Details
In line with the overall goals of HBCC for reducing the transmission of COVID-19 via rapid response interventions to raise awareness and enacting long-term behaviour change regarding people’s personal and environmental hygiene habits, BRAC has taken an adaptive approach under the Hygiene and Behaviour Change Collation (HBCC) platform. The aim of this project is to systematically change the behaviour of people across communities in Bangladesh via interventions such as setting up handwashing stations and broadcasting behaviourally-informed messages to nudge people to adopt healthier prevention and response practices.
To achieve the overall goals of the project and positively impact the lives of over 10 million people in communities across Bangladesh, BRAC primarily chose 20 sub-districts from 3 divisions (out of a total of 8 divisions) for implementing the interventions, namely Dhaka, Mymensingh and Khulna. The prime activity is to install 1,000 handwashing stations across these sub-districts to increase the access of communities to handwashing facilities. The BRAC handwashing stations have been designed to enhance both availability and inclusiveness, fitted with foot-operated pedals for dispensing water and liquid soap, to reduce contact with sink surfaces and thereby reduce probability of transmission from the device surface. They have been fitted with posters, promoting handwashing with proper technique, and have also been designed to allow hand-washers to maintain social distance while using the devices.

In addition to these handwashing stations, there are other supports available in the intervention areas i.e. in-person demonstration, hygiene meetings, and soap distribution. After the economy opened up in Bangladesh after a long lockdown, the community engagement activities in the form of in-person communication and forum discussions, to motivate people to uptake safe hygiene behaviours including but not limited to regular handwashing. In these demonstrations and meetings, the participants are motivated to adopt recommended hygiene behaviours such as handwashing with soap, mask usage and disposal, sneezing etiquette, social distancing, etc. Along with the BRAC staff, the WASH entrepreneurs who are members of the local community whom BRAC traditionally supports with free training in entrepreneurship and market development to improve the communities' access and knowledge of hygiene products. The HBCC project empowers them with responsibilities to disseminate the messages of the recommended behaviours and distribute the soap in the communities across the intervention areas, especially the poorest of the poor community and the participants who attend the meetings regularly.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
The main outcome variables of interest in this study are:
i) daily handwashing frequency
ii) likelihood of using soap when handwashing
iii) prevalence of transmissible diseases in the past 15 days
Primary Outcomes (explanation)
i) constructed using the self-reported daily frequency of handwashing and typical handwashing times
ii) constructed using self-report (times out of 10), rapid observation of the presence of soap near handwashing facilities, and self-reported expenditure on soap
iii) constructed using self-report of symptoms and specific diseases
Secondary Outcomes
Secondary Outcomes (end points)
i) knowledge of handwashing practices
ii) attitudes towards handwashing
iii) use of public handwashing stations and alternatives to public handwashing stations
iii) frequency of mask-wearing
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
To understand whether access to public handwashing stations improves people’s hygiene practices, we collected a list of potential villages from BRAC-WASH programme where BRAC can install the public handwashing stations in their selected 20 sub-districts from three divisions-Dhaka, Mymensingh and Khulna. Then, from each of these 20 sub-districts we randomly selected 10 villages, of which 5 villages are assigned to the control group, which will not receive HBCC interventions (i.e., no new public handwashing stations or related community demonstrations/communications). The rest of the 5 villages are assigned to the treatment group where BRAC’s WASH team must implement the HBCC interventions. We have 98 villages (from 20 sub-districts) to control and 96 to treatment. Because of an insufficient number of village selection in Khulna city (one of the 20 sub-districts), we had to reduce the number of treatment and control villages from this sub-district. In the rest of the villages listed, the program may/may not be implemented.

Re-randomization for soap distribution
The first round household survey revealed that soap use significantly increased while handwashing frequency and soap expenditure decreased but not statistically significant for treatment villages. Around one-fourth of the treatment households received the soap from BRAC. From the results, it can be hypothesized that handwashing with soap reduces the repeated handwashing frequencies and people who use soap are more likely to wash their hands less frequently to save soap expenditure. In order to test these hypotheses, we would like to provide the soap among randomly selected 50% of our surveyed households. This re-randomization of soap distribution is stratified at the sub-district level. The numbers of soap-recipient households are 3,842 (1,909 households from treatment villages and 1,933 households from control villages. The research team distributes the soaps in their capacity.

Re-randomization for the phone survey
The second wave of COVID-19 showed up at the end of March in 2021 worldwide. This wave drastically increased the number of cases in Bangladesh since its outbreak and the government announced lockdown nationwide in early April 2021 to control the spread. During that situation, we planned to carry out a phone survey with a short questionnaire to understand the situation at that time. We conducted the phone survey on those households who were successfully surveyed in the first round of survey. On top of it, Ramadan was about to start. Thus, we randomly divided the sample into two groups- to be surveyed before and during Ramadan. This randomization is also stratified at the sub-district level. 44% of the households were interviewed before Ramadan and the rest of them were interviewed during Ramadan.
Experimental Design Details
Not available
Randomization Method
We collected the list of potential villages from the BRAC-WASH programme. They selected these villages based on some selection criteria (i.e., presence of BRAC’s WASH programme, the scope of collaboration with BRAC’s other programmes, the existence of WASH entrepreneurs, potential partnership and collaboration with stakeholders such as Pourashavas, Upazila, and Union parishads, educational institutions, etc.). Then, we randomized the villages using STATA.
Randomization Unit
Our randomization unit is village. For soap-distribution, we randomized at the household level.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
194 villages
Sample size: planned number of observations
7,760 households using the household level instrument and 5,820 individuals in the public places.
Sample size (or number of clusters) by treatment arms
96 treatment villages and 98 control villages

3,842 households received the soap of which 1909 are from treatment villages and 1933 are from control villages. And 3845 households did not receive any soap of which 1898 are from treatment villages and 1947 are from control villages.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
IRB Approval Date
IRB Approval Number