In his blog, AZM Anas, writing under the caption How Dhaka won the fight against open defecation and improved public health

citiscope.org/story/2017/how-dhaka-won-f...886ae27d03-118055621

says:

NGOs worked closely with the government and used their grassroots networks to build sanitary latrines in infomal settlements like the Sattola area of Dhaka.

Volunteers work with community members one-on-one and in groups to encourage healthy hygiene.

The efforts have helped reduce incidence of diarrhoea and mortality for children under age 5.

While in Pakistan, OD is rampant, it is extremely encouraging to note that Bangladesh is now OD free.

Please see this UNICEF report:

www.unicef.org/publications/index_96611.html


F H Mughal

As it is somewhat related, in this interview two field researchers claim that the effect of CLTS in Bangladesh might have been very minimal:
www.firstpost.com/india/casteism-is-the-...-coffey-3925663.html

CLTS was invented in 2000 and people talk about how it was an important part of what happened in Bangladesh. But open defecation in Bangladesh in 1991 was much less than what it is in India today. It was already going down in Bangladesh and if you look at open defecation before and after CLTS, it is going down at the same rate. That doesn’t definitely prove that CLTS doesn’t work but we need to think about it carefully.

Anyone got an idea where this data could be found in a scientific publication?

Hi All,

Here is another excellent study published in Science magazine (www.sciencemag.org/content/348/6237/903.abstract). I can only access the abstract, but that is enough. *

What they have found is:
1. Community motivation did not raise toilet ownership.
2. Subsidies by themselves did not increase toilet ownership.
3. Subsidies to the landless poor increased ownership among the subsidized.
4. Subsidies to the landless poor increased ownership among unsubsidized households.
5. These subsidies greatly reduced open defecation.

Hence the way to get the poor to use a toilet is to buy them a toilet. And if you buy enough, people will go out of their way to get one. Scope for positive government intervention here.

++++++++++
Note by moderator:
* Encouraging sanitation investment in the developing world: A cluster-randomized trial
Raymond Guiteras1, James Levinsohn2, Ahmed Mushfiq Mobarak2

I have to agree with Blake and say that I wouldn't consider the approach taken in the supply side intervention "improving the functioning of the sanitation market." I echo all of Blake's concerns about the desirability and quality of the product and to what extent it actually responded to the needs and desires of the consumers.

I also question the LSA model used, but not necessarily the use of salaries to compensate LSAs (see iDE's awesome summary of the HYSTRA report attached--having a full-time, dedicated sales force can reduce turnover and increase efficiency). My real concern is that they weren't use effectively. Their roles are described in the supplemental document as: 1) providing information about where to buy a latrine; 2) enabling households to assess the quality of latrines being sold; 3) Assist with delivery and installation; and 4) Provide technical support post-purchase. But they were also instructed not to provide information about the benefits of latrine use!

There was no intervention tested that combined demand creation (either through village level sanitation promotion like LPP, or through IPC or DCC via sales agents) with supply. Just as the success of CLTS is limited without a functioning supply chain, the supply chain cannot function without a sufficient level of demand. I wonder what the results would have been if LSAs were able to have conversations about the benefits of latrine use and the difficulties of OD AND connect people with aspirational and affordable products?

I'd also be interested to see what the use rates are in, say, six months to a year. Did the size of the subsidy make the product cheap enough to buy, whether or not the purchaser really intended to change OD habits? Maybe. But I think there's room to explore how subsidies could actually benefit the private sector, if used in the right way. For example, could these subsidies, in the form of vouchers, actually support the private sector by giving them access to a portion of the market that they wouldn't have without the help of subsidies? Could those vouchers give the poorest the ability (and the dignity) to purchase a latrine that they can be proud to use, AND help make sanitation a more viable business (which would not only help current sanitation businesses survive and grow, but also encourage more entrepreneurship in the sector)?

Fascinating study, fascinating discussion. The comments about precise methodology and extent of generalisability are interesting, but hey, even the best studies in our context all have significant issues: it's not clinical research! [Having said that, self-reported OD seems likely to be subject to substantial shame-related bias, so I'm surprised the researchers selected this as a proxy for latrine use.] But actually I'm more interested in implications than precise methodology. There are two principal arguments against latrine subsidy: a) that it demotivates people who don't get a subsidy (why should they invest, better to wait for next subsidy?); and b) that users won't look after a subsidised latrine as well as if they'd paid for it with their own money. Intriguingly, this study seems to provide evidence against (a): near-neighbours of people who got a subsidy were MORE likely to invest than controls. As regards (b), this study doesn't appear to present any evidence either way: latrine use was assessed only once, presumably soon after the intervention. It would be useful for the researchers or others to go back and see how well latrines are being used now in the different treatment groups (and that'd also be an opportunity to apply some better metrics of latrine use/care). That would be important to properly interpret these findings. (PhD/masters project anyone?) Independently of this, on the other side of the equation from the arguments against, there is of course a clear argument for well-designed targeted subsidy: it provides a hygienic facility for people who wouldn't otherwise be able or willing to invest (with benefits for that household, and also probably for other people in the community). I certainly welcome an evidence-based study which encourages us to question the "latrine subsidy is bad" dogma.

Guy Norman

of Water & Sanitation for the Urban Poor (WSUP) and Public Finance for WASH initiative (www.publicfinanceforwash.com), opinions are my own

Hi All,

Quite an interesting and lively discussion. I can weigh in and provide a couple answers/reflections from the SanMark point of view, as iDE brings experience implementing SanMark in 6 countries to date. Please note that all of these points, and my reflections, are in regards to the 'supply side interventions' mentioned and not a reflection on the whole study (thats not my place).

The study mentions "These results counter the concern among many development practitioners that subsidies undermine intrinsic motivation. Rather, this research shows price is a primary barrier, which is consistent with a growing body of research on adoption of health products." Of course price is a major concern for these households. However, this research concludes that subsidy is the only way to deal with price barriers. The other option is that the latrine itself was not deemed valuable enough by customers to be worth the price. So instead of concluding that subsidy is the only way to tackle this, I would argue that making investments in the latrine itself to make it more desirable by customers and/or professionalizing the sales force to make them effective sales agents could also over come the price barrier.

The Study mentions "The subsidy vouchers could be redeemed for a 75 percent discount on available models of latrines, priced (after subsidy) from $5 to $12. The households were responsible for their own transportation and installation costs." Building off the point above, in SanMark the essential piece is that the product you are selling is truly desirable. These latrines seem pretty cheap and if they don't meet the needs/wants of the customers (aspirational, comfortable, etc.), then its not surprising you can't sell them. I also don't buy things I dont want. I can't assess if the latrines do or don't meet customers' needs, but an important thing to consider with any SanMark approach is designing/selling a product that people want and are thus willing to pay for.

The last point is simply that the supply side intervention seemed very, very light touch - primarily engaging masons for a couple weeks to provide technical and sales support (i.e. convince people to buy a latrine). The study also mentions these LSAs were hired for 12 weeks and were paid a full salary. In iDE's experience, this is extremely light touch and not something I would classify as Sanitation Marketing. SanMark is about creating a market for a product/service, and that inherently takes multiple years (not multiple weeks) and adoption follows the typical product adoption curve. SanMark is far bigger than pointing someone in the direction of a latrine supplier. Its about identifying customers needs, designing products that satisfy them, building suppliers capacity to produce them at a good price, and helping households understand the consequences of not having a latrine so they are motivated to buy one. Often this motivation is NOT health, but dignity, pride, and 'keeping up with the Jones'. Sales agent training is also a major component requiring recruiting specific people, weeks of classroom & field training, and months of ongoing coaching and support. These sales agents are also paid on commission, motivating them to actually close sales, as when sales agents receive a salary they have no motivation to actually follow up/close deals. Plus the product has to be right, the price right, and the sales messaging right. All in all, I don't think its fair to extrapolate the results of this study on more robust SanMark efforts.

It seems there are some important learnings from the study and things not to be discredited. It also seems like a complicated undertaking trying to compare 3 different approaches in an RCT. In the end, I think the results are valuable for those projects studied but we should be careful extrapolating them more widely as they are heavily dependent on the effectiveness of the original approaches in the first place.

Best,
Blake


* abbreviations:
LSA = Latrine Supply Agent
RCT = Randomized controlled trial

++++++++
Note by moderators: This post was made by a former user with the login name BlakeMcK who is no longer a member of this discussion forum.

Dear All
Problem with single pit is that after the pit gets filled (in 3-5 years, depending on number of family members) toilets become non-functional. Such Households have no option except to go for defecation in open. There is no mechanical device readily available to clean faecal sludge in many cases. Even it is cleaned, there is no infrastructure available to treat it, in most of the cases.
For high ground water table areas leach pit is not suitable at all due to chance ground water. In fact this technology in such areas may cause irreversible affects on environment and community health.

pawan

Seems to me that they mean "sealed towards the surface" which is a rather uncommon way to use "sealed pit". But well, with high groundwater tables it is indeed debatable how effective a pit-latrine is in containing pathogens.

Krischan, there are diagrams in the supplementary materials pdf, and they also say:

This definition is based on the Government of Bangladesh’s (GoB) 2005 National Sanitation Strategy, which defines a hygienic latrine as “a sanitation facility the use of which effectively breaks the cycle of disease transmission”

This requires the confinement of feces (i.e. a sealed pit) and a sealed passage between the squat hole and the pit, blocking insect vectors.

Fig S2 (a) "Model 1" seems to imply that the latrine was sealed, but the text also implies that vent was not compulsory.



The other models have more rings, but it is not clear whether they are supposed to be sealed at the bottom.

A bit oft topic, but do they really build fully sealed pit-latrines in Bangladesh (See their Definition)? Seems rather unlikely and would void the usual "benefit" of pit latrines that they don't fill up quickly with flush water, at least in the off-set case.

Just to answer some of my own questions from the supplementary materials:

Hygienic latrines:

A hygienic latrine safely confines feces in order to prevent disease transmission. To accomplish this in our context requires a seal to block flies and insects and a sealed pit to store fecal matter for safe disposal. We classify a latrine as hygienic if it (1) has an intact slab, (2) has a functional seal, and (3) conveys feces to a sealed pit. The following types of latrines are included, provided they satisfy the three criteria above: direct pit latrine (slab directly above the pit); offset single pit latrine (pit is offset from the slab); offset double pit latrine.

On the differences between this intervention and CLTS (in addition to the point made about latrines):

The timeline for “ignition” in CLTS is open-ended, and typically takes up to 5 days ofmeetings. In LPP, we fixed the length of the intervention, such that the initial period leading to ignition lasted 2 days, with a fixed schedule of follow-up visits thereafter. This was to maintain comparability between the LPP components of the LPP Only
and LPP + Subsidy treatments. Without a fixed timeline, the intervention might go
on for longer in LPP Only villages, meaning that the comparison between LPP Only
and LPP + Subsidy would no longer represent the marginal effect of subsidies.

On the supply side:

It was not feasible to randomize an intervention directly with latrine suppliers. First, therewere only 11 in the project area. Second, there was no clear correspondence between communities and suppliers. Instead, we designed a community-level intervention intended to improve the functioning of the sanitation market. VERC identified, trained and hired individuals in randomly chosen neighborhoods to work as Latrine Supply Agents (LSAs) in that neighborhood. VERC recruited residents who worked in fields such as masonry, construction or carpentry, and therefore were likely to have adequate technical ability and knowledge.

VERC trained LSAs to act as technical and sales agents, in particular to:
1.Provide information about where an improved latrine can be purchased
2.Enable households to assess the quality of latrines offered for sale
3.Assist with delivery and installation
4.Provide technical support after installation, e.g., for maintenance, repairs and improvements.
LSA assistance was available to all residents of designated neighborhoods, regardless of
voucher eligibility or winning status. LSAs were hired for 12 weeks, including the 6 weeks
during which households could redeem their vouchers, plus additional time to assist with
installation. LSAs were instructed to provide information only about purchase, installation and maintenance, and not about the benefits of latrine use. VERC paid LSAs a flat salary.

Hi everyone, Dr Guiteras says that some of our (my) questions are answered in the supplementary information, which are freely available here:

www.sciencemag.org/content/suppl/2015/04....DC1/Gutieras.SM.pdf