Making WASH behaviour stick

| September 21, 2015

Photo by Jim Holmes for AusAIDSmall scale trials have repeatedly demonstrated the health benefits of water, sanitation and hygiene (WASH) programmes. But the benefits disappear when many of these programmes are taken to scale. Large scale handwashing, sanitation and water treatment campaigns in India, Bangladesh, Peru and Vietnam have not led to a change in the behaviour of most of the people they target. So, the health impacts of these programmes are insignificant.

As we gear up for the Sustainable Development Goals, it is time to take stock of the evidence we have on sustained adoption of WASH programmes. Recent systematic reviews and impact evaluations provide some insights on what does or does not make WASH behaviour stick when programmes are taken to scale.

Why just building toilets doesn’t work

A 3ie-supported impact evaluation showed that providing subsidies and mobilising communities for toilet construction led to a substantial increase in latrine coverage in Odisha, India. But the increase in toilets did not lead to a reduction in the exposure to faecal matter. There was hence no reduction in child diarrhoea.

So, why didn’t building toilets have an impact? Many households did not use the toilets even though they had access to them. Men and children in particular continued to defecate in the open. Changing people’s attitudes about open defecation in India requires overcoming significant cultural barriers. Open defecation amongst particular groups in India is seen as a practice that preserves ‘purity’ since it prevents the accumulation of faeces inside or near the house. Just providing subsidies to households for building toilets does not therefore change cultural norms.

In contrast, a community-led total sanitation (CLTS) campaign in Mali did not provide any financial assistance for building toilets or purchasing any other hardware. An impact evaluation showed that the programme was successful in increasing access to private latrines and in reducing self-reported open defecation among men and women. This resulted in a reduction in child stunting and diarrhoea-related deaths of children under age five.

What made the difference? The approach to community engagement was stronger in Mali than in India. The achievement of open-defecation free status in Mali was also acknowledged in a celebratory public ceremony. But it may well have been that there were no cultural barriers to overcome in getting people to use toilets in Mali.

The lesson here: WASH programmes clearly need to understand the barriers to and facilitators of behaviour change. The ‘just build it and they will come’ philosophy misses the fact that human behaviour and the norms that drive it also need to be addressed and changed.

Many factors affect WASH behaviour

©UNICEF Ethiopia/2014/AyeneA recent 3ie-supported review mines through 44 studies to come up with a host of contextual, psychosocial and technological factors that can affect sustained use of WASH practices and technologies.

As is the case with toilets, just providing water treatment technologies does not automatically lead to a change in behaviour. Making people pay even a very small fee puts them off of adopting the technology. The smell and taste of chlorine can also be a deterrent in adopting water treatment.

Age and gender are other crucial factors in determining WASH practice, particularly in latrine use and handwashing. Safety is clearly a key factor that influences the use of toilets for women. In Mali, women in the CLTS villages found it safer and more private to use toilets. But the evidence on these factors is scant. 3ie’s WASH evidence gap map shows that very few impact evaluations examine gendered impacts.

Children are unable to use toilets if they are very small. Even households that have access to toilets dispose child faeces in an unsafe way. Effective disposal of child faeces is therefore a priority area for improved sanitation and hygiene.

Barriers to behaviour change over time

Barriers to behaviour change depend on the stage of the project. Many studies assess the health benefits of initial uptake of safe water, hygiene and sanitation technologies and practices. But few studies consider sustained use. The early project period may be characterised by enthusiasm over the new technology or promotional activities. Enthusiasm may diminish in the late project period but project staff are still around to resolve issues with respect to cost and availability of the hardware supplies. Although external support ends during the early post-project period, the promotional messages may still be fresh in people’s minds. However, influential household members who were sceptical may reassert their domination during this phase. And finally, in the late post-project period stockouts, technology failure or poor maintenance systems can pose a serious threat to sustained adoption.

Incentives for sustained behaviour behaviour change thus differ from what influences initial uptake. A 3ie-supported study in Kenya shows that local government officials valued the opportunity of taking decisions on how and where chlorine dispensers should be installed within their constituencies. But they did not want to take on the responsibility for maintaining chlorine dispensers.

The nature and frequency of behaviour change communication affects uptake

The effects of one-to-one interactions differ from those of group-based approaches and mass media campaigns. A community programme in Peru combined with a mass media campaign to promote handwashing, was more effective at improving handwashing behaviour, compared to the mass media campaign alone.

The 3ie-supported systematic review finds that frequent, personal contact with a health promoter over a period of time is associated with long-term behaviour change. The review suggests that personal follow-up in conjunction with other measures like mass media advertisements or group meetings may further increase sustained adoption.

What next?

Sustained adoption of WASH technologies and practices can lead to lasting health benefits. What we know now is that just focusing on technological and hardware fixes does not produce behaviour change. Multiple factors affect behaviour but not enough attention is being given to these factors and how they change over time.

The recent synthesised evidence throws up a few important pointers about where we need to prioritise getting more evidence:

  • We need more formative research on the cultural, social and psychological factors that affect WASH behaviour. This can in turn inform WASH programme design.
  • Impact evaluations should analyse the impact of WASH programmes on addressing the multiple barriers to and facilitators of sustained behaviour change.
  • WASH programmes need to address the particular needs of women, children, older and differently-abled people. We need to fill the evidence gap on how age and gender affect uptake and outcomes specific to different population groups.

Tags: , , , ,

4 Comments on “Making WASH behaviour stick

  1. carlos levinton

    we are really inetrested in joinng your programs with our innovative solutions for dry sanitation and potabilization of water.-Also sustainabel solutions to eco managment of all kind of waste.Our model called PACHA was selected by UNIted Nations as one of the 100 bes social practices in the world Dubai 2012.-
    We want to expand this model
    national cordinator of RIVS Helath in housing net.-PAHO-WHO

  2. Peter McIntyre

    Yes evidence does matter – and it is important not only to look at bad news. From Bangladesh you cite a 2012 report of a survey of behaviours six months and 18 months after the (SHEWA-B) programme began and found no measurable reduction in childhood diarrhoea or respiratory illness.
    But give credit where it is due. Overall open defecation has reduced in Bangladesh from more than 30% a decade ago to 3% today – an astonishing figure.
    Changes in behaviour take time to translate into changes in disease patterns and childhood diarrhoea will flourish until water sources are well protected and all members of the family use toilets and wash their hands.
    The BRAC WASH programme which began in 2006 covers half of rural Bangladesh and has been praised by The Embassy of the Kingdom of the Netherlands, as “hugely present and effective” and by the Bill & Melinda Gates Foundation for results described as “beyond excellent”, achieving 98% sustained use of toilets by all family members.
    Independent monitoring suggests that the BRAC approach of loans for poor families and grants for ultra-poor families has been especially successful in supporting community efforts.
    Change takes time. The percentage of households in the programme area that had soap and water in or near the latrine rose from 45% one year after the programme began to 69% after eight years.
    No behaviour change is complete (watch how many people in your organisation leave the toilet without washing their hands). But real change is happening.
    The BRAC WASH programme in Bangladesh is a globally significant example of benefits that have NOT disappeared when taken to scale. Research on the impact on disease is being discussed by BRAC.
    BRAC’s own insights (developed with support by IRC from the Netherlands) have been that sustained community efforts over a long period of time are needed to create generational changes in hygiene behaviour, and that the provision of subsidies and loans has been important in extending gains to very poorest families.
    A report is available at

  3. jindra

    Thanks, it is so important to examine evidence rather than believe it will all turn out as we expected, especially with something as culturally-embedded as WASH behavior.
    You write – and we at ValuingVoices firmly agree — that “few studies consider sustained use….Although external support ends during the early post-project period, the promotional messages may still be fresh in people’s minds. However, influential household members who were sceptical may reassert their domination during this phase. And finally, in the late post-project period stockouts, technology failure or poor maintenance systems can pose a serious threat to sustained adoption.” To what degree do 3ie impact evaluations go back after close-out? Many impact evaluations I’ve come across are projected impact, evaluated during the project. Thanks! Jindra, founder of

Comments are closed.