Lessons from CLTS Implementation in Seven Countries

Learning Brief • February 2016

Purpose

This learning brief shares key findings that emerged from a cross-country synthesis of CLTS projects implemented by Plan International Country Offices (COs) in Cambodia, Nepal, Indonesia, Lao PDR, Uganda, Niger, and Haiti. Specifically, this research aimed to characterize variations in CLTS implementation through the perspectives of stakeholders, and to identify the roles of local actors in implementing CLTS. Several implications are relevant for consideration by Plan International staff across the seven COs, as well as other sanitation practitioners.

The brief is part of the CLTS Learning Series (LS), a collection of case studies on CLTS implementation, prepared by The Water Institute at the University of North Carolina at Chapel Hill (UNC) as part of the Plan International USA project, Testing CLTS Approaches for Scalability. Research activities conducted between 2012 and 2015, sought to better understand CLTS facilitation and mobilization by rigorously evaluating three distinctive strategies to enhance the roles of local actors in CLTS interventions in Kenya, Ghana and Ethiopia. The seven country case studies referenced in this report were undertaken to complement the three pilot evaluations, and to further explore the role and potential of local actors in CLTS.

Methods

UNC researchers collected data between May 2013 and June 2014 with support from Plan International COs. In-depth interviews were conducted with 293 people, including policymakers, Plan International staff, other non-governmental organization (NGO) partners, local government officials, village-level CLTS facilitators, and community leaders. Policy and programmatic documents were also gathered over the course of two to three weeks in each country. Across all 7 countries, thirty-four communities were visited, 44% of which were declared or certified as open-defecation free (ODF) by the time of the visits. Interview transcripts, field notes, and documents were analyzed qualitatively to identify themes pertaining to different stages of CLTS, from which implications and conclusions were drawn for sanitation practitioners as a whole.

Key Findings

CLTS Triggering in Lao PDR
Photo by Plan International

  1. CLTS was widely perceived as a universally applicable approach in rural communities, despite varying success in outcomes.
  2. Local government, in five of the seven case studies, had insufficient resources and motivation to take ownership of CLTS,
  3. Village volunteers were involved throughout CLTS implementation, but required considerable support from Plan International and local government.
  4. Triggering techniques had been adapted in all seven case studies, but were not always designed with the aim of improving outcomes.
  5. Although community-developed sanctions are encouraged, most examples from the case studies were enacted by village or district government.
  6. Affordability and access to durable latrine mate-rials were key obstacles affecting sustainability of outcomes.
  7. Monitoring of CLTS varied widely across programs, with different indicators of success, ODF definitions, verification guidelines.

Implementation Context

Plan International’s implementation arrangements for CLTS varied between each country, ranging from direct implementation to playing a more supportive role of providing technical and capacity building assistance to local government. The complexity of arrangements depended largely on government support for CLTS and sanitation, as well as the capacity of different actors to participate in the process. Where national government support for CLTS was strong and the approach had been implemented for several years (Nepal, Indonesia, Uganda, and Cambodia), local government played a more important role in facilitating CLTS activities, and arrangements were more complex. These governments were directly investing in CLTS activities to train staff local government counterparts. Where national government support was weaker (Lao PDR, Niger, and Haiti), Plan International played a lead role in implementing CLTS. In all seven programs, community leaders played an important role in implementing CLTS, but mostly in the post-triggering stage.

The policy environment in all seven countries was found to be largely positive towards CLTS. All seven governments recognized the need for demand-led sanitation strategies. However, several national policies, such as those in Lao PDR, Cambodia, and Niger, allowed for targeted hardware subsidies for households or public facilities. In all countries where latrine subsidy projects and CLTS overlapped, CLTS practitioners cited considerable challenges.

CLTS Progress

Across all seven LS countries, Plan International has trig-gered nearly 1,000 communities and is one of the major NGOs implementing CLTS. Self-reported monitoring data revealed a wide range in the presence of ODF communities—from 6% of communities in Haiti being declared ODF to 97% of communities in Indonesia being certified as ODF (Table 1). However, these numbers cannot be directly compared across case studies because ODF definitions and verification processes were not consistent. In terms of household latrine coverage, Plan International programs in Indonesia, Uganda, and Lao PDR appeared to have the best end-line results. Baseline data were available for four of the seven case study programs. Of these, the largest ab-solute increase in household latrine coverage after trig-gering activities occurred in Uganda (44%). However, it is not possible to attribute this progress entirely to CLTS activities, as other factors may have also contributed to the increase in latrine coverage, such as campaigns by government or other organizations.

Table 1. Overview of Plan International’s CLTS outcomes in case study programs, 2013–2015

Country No. of communities triggered No. of HH Avg. no. of HH per village Avg. baseline latrine coverage Avg. end-line latrine coverage Increase in latrine coverage No. (%) ODF communities
Cambodia 356 64,562 181 NA 40% NA 38 (11%)
Lao PDR 46 4,027 88 48% 74% 26%  17 (37%)
Nepal 105 171,212 1631 32% 59% 27%  29 (28%)
Indonesia 153 174,426 1140 NA 97% NA  149 (97%)
Uganda 152 14,284 94 51% 95% 44%  67 (44%)
Niger 87 10,968 126 8% 33% 25%  31 (36%)
Haiti 83 NA NA NA NA NA  5 (6%)

Key Findings & Implications

Finding 1: CLTS was widely perceived as a universally applicable approach in rural communities, despite varying success in outcomes.

Most CLTS practitioners and several policymakers who were interviewed believed that CLTS could be implemented in all rural settings. Only a few respondents challenged the notion of CLTS as universally relevant, and believed in the need for alternative strategies in settings where CLTS had not worked. This widely-held conviction meant that even in challenging environments, alternatives to CLTS were less likely to be considered. Many respondents believed that CLTS could at least be a “starting point” to generate demand, even in settings where practitioners themselves felt it was unlikely to succeed in ending open defecation by itself.

Some practitioners did recognize that alternative or complementary strategies to CLTS may be needed, but the more prevalent belief in the universal relevance of CLTS seems to have overshadowed the need for targeting appropriate communities with this approach and seeking alternative sanitation strategies for less receptive communities. Practitioners’ perceptions of CLTS are important because they determine how the approach is implemented, specifically the manner in which communities are selected for triggering. The widespread application of CLTS may help explain slow progress in several programs. By not targeting communities to optimal settings for CLTS, it is probable that villages not appropriate for CLTS may have been triggered, leading to slow increases in latrine coverage and low ODF attainment.

Finding 2: Local government, in five of the seven case studies, had insufficient resources and motivation to take ownership of CLTS.

It is widely acknowledged that local government support and capacity are vital for scaling up social and public health programs. In all case studies except Haiti, Plan In-ternational worked closely with local government actors, even when local government was not mandated to be involved in CLTS. Local government involvement ranged from leadership in CLTS (Nepal) to requiring considerable assistance from Plan International (Lao PDR) to no involvement at all (Haiti).

Government and NGO respondents cited insufficient local government capacity as a key challenge to increasing the scale of activities. Capacity referred to legal responsibility for local government to provide sanitation services; local government budget for CLTS; sufficient staff time avail-able for sanitation; access to transportation to routinely follow-up in remote areas; and experience and skills for facilitating or managing CLTS on their own.

Slow progress in some countries may be partly explained by local government implementing CLTS with limited capacity, including facilitation skills, resources, and motivation for routine follow-up activities. Where local government capacity is insufficient to lead CLTS, NGOs continue to play a dominant role in all stages of implementation until local government is able to take on a leadership role.

Finding 3: Village volunteers were involved throughout CLTS implementation, but required considerable support from Plan International and local government.

As CLTS is a “community-led” process, a crucial component of the strategy is to involve community leaders to take charge of their own sanitation situation. CLTS programs typically refer to “natural leaders”, who emerge from the triggering process and participate in post-triggering activities.

In the seven case studies, a variety of village volunteers were found to be involved in all stages of CLTS. In Nepal and Haiti, volunteers were involved in the pre-triggering stage itself to mobilize communities. In Indonesia, Nepal, and Uganda, village volunteers were recruited as triggerers. While these volunteers were supposed to ultimately lead triggering events, they were not yet able to do so and were supported by local government or Plan International facilitators.

Village volunteers increase community-level engagement, and can also lower the cost burden for practitioners since fewer follow-up visits may be required. However, costs are then transferred to volunteers. Although volunteers in the case studies appeared to be highly motivated, main-taining this level of engagement in the long run remains a concern and may require additional resources, such as increased numbers of training events, and providing mon-etary and non-monetary incentives.

Finding 4: Triggering techniques had been adapted in all seven
case studies, but were not always designed with the aim of improving outcomes.

Plan International used a variety of approaches to trigger communities to change sanitation behavior. The most commonly cited triggering tools were the transect walk/ “walk of shame,” village mapping, shit calculation, water-feces demonstration, and analysis of medical costs. However, several of these triggering tools did not appear to be used routinely in all programs.

For instance, local government facilitators in Cambodia and Lao PDR hesitated to use strong shaming techniques, reportedly due to cultural reasons, and omitted certain steps, such as the water-feces demonstration, because they themselves were too embarrassed to lead these activities. On the other hand, in two triggering events observed in Nepal, LNGO facilitators insisted strongly on community members’ continued participation, even if they were too ashamed or disgusted to stay in place during the transect walk or the water-feces demonstration. Triggering techniques were also adapted in Niger, where facilitators observed that communities were not as aware of the harms of open defecation; therefore, facilitators emphasized health benefits of ending open defecation, rather than techniques that are meant to incite shame and disgust.

Triggering techniques are likely to strongly influence CLTS outcomes. Adaptations in triggering indicate that Plan International does not follow a set template for CLTS and recognizes the need to modify the approach to suit different contexts. However, all adaptations are not equal. Context-specific adaptations that emerge from community- level observations and experiences can be encouraged. However, certain adaptations may compromise the CLTS approach itself and slow progress in communities, and may need to be modified through improved training and selection of facilitators.

Finding 5: Although community-developed sanctions are encouraged, most examples from the case studies were enacted by village or district government.

In international CLTS guidelines, community-innovated sanctions against open defecation are encouraged and listed as a key indicator for monitoring progress in communities. In the seven case studies, truly community- developed sanctions were only reported in a few instances in Uganda and Lao PDR. It was more common to find sanc-tions developed by village or district government, such as in Indonesia and Nepal where social insurance cards or government donations were withheld from households. These sanctions were often informal in nature, as there was no law or guideline authorizing government to with-hold services based on sanitation status.

Sanctions may be a key component in creating and enforcing social norms. The question is not whether the law has a role to play in sanitation, but rather which form is appropriate, at what stage it is introduced, how it is enforced, and how effective it is at ending open defecation and improving safe and equitable sanitation. Local government-imposed sanctions may lead to increased latrine construction, but it is unclear how they influence long-term changes in social norms. They may also inadvertently harm the most vulnerable sections of society who cannot afford to build latrines.

Finding 6: Affordability and access to durable latrine materials were
key obstacles affecting sustainability of outcomes.

In all case studies, practitioners, policymakers, and community leaders cited the poor quality of latrines built as a result of CLTS as the primary challenge for ensuring sustainable use of sanitation. Triggered households constructed a variety of latrine types, but there was a strong preference for water-sealed latrines or durable latrines made of cement. In all seven countries—particularly in Cambodia, Niger, Uganda, Haiti, and Indonesia—Plan International worked to improve access to the supply chain, primarily by training masons or through broader sanitation marketing efforts. However, community leaders cited affordability as the main obstacle for being able to act on behavior change messages. Positive examples of local financing and community support mechanisms were identified in the case studies. However, there were also several examples of different forms of hardware subsidies in triggered communities. While they largely believed that CLTS should be a no-subsidy approach, several policymakers and Plan International staff were in support of targeted financial or material support to vulnerable households. Some acknowledged that CLTS may not be enough to address the supply side of the sanitation problem.

By training masons, implementing sanitation marketing programs, and encouraging local financing mechanisms, Plan International staff have shown that they recognize the importance of improving supply-side conditions in the post- triggering phase. Sanitation marketing in particular may help improve access to the supply chain, serving as a com-plement or as a viable alternative to CLTS. However, the challenge of increasing access to sanitation for vulnerable and marginalized populations may require additional financial or material support, provided that it follows CLTS activities and is targeted to those most in need.

Finding 7: Monitoring of CLTS varied widely across programs, with different indicators of success, ODF definitions, verification guidelines.

CLTS monitoring activities comprised a variety of process and outcome indicators, but ultimately focused on achievement of ODF status, except in Haiti. Most programs had simple monitoring systems that did not systematically capture sufficient data to enable cross-country comparisons. Of the seven case studies, only programs in Lao PDR, Uganda, and Niger consistently captured baseline measurements in communities. Definitions of ODF varied substantially across all seven case studies, underscoring the challenge of measuring behavior change at the community level.

Without baseline measurements and routine assessments, programs cannot measure change or appropriately analyze the effectiveness of their CLTS activities. Furthermore, many programs aimed to measure total sanitation under the definition of ODF by adding indicators on handwashing, safe water practices, and environmental sanitation. This sets a more ambitious goal for ODF attainment than the actual the term ODF would imply. Improved monitoring efforts are especially important when there is no standard ODF definition, thereby making it impossible to compare results across programs.

Recommendations

Role of CLTS

CLTS should be considered as one component of a sanitation strategy. Communities that are more likely to be receptive to CLTS should be targeted systematically so that practitioners can allocate remaining resources to test other approaches, such as sanitation marketing, in communities where CLTS may not be appropriate.

Local government capacity

Where local governments are unable to lead CLTS activities, INGOs could strengthen their capacity through training, mentorship, and targeted technical support, and by engaging local NGOs to trigger communities. They could advocate for increased national government investment in CLTS.

Role of village-level actors

When involving volunteers, resources have to be budgeted for training, financial and in-kind support, recognition, and exchange visits, in order to sustain motivation through the lifetime of the program and beyond.

Adaptations to triggering

Programs could systematically identify adaptations to CLTS and critically analyze whether the adaptations are a result of community context or a result of convenience or logistical constraints. They could attribute results to the actual approach that has been implemented so that rural sanitation stakeholders can better understand the effectiveness of CLTS vis-à-vis other approaches.

Sanctions

CLTS practitioners need to carefully consider which sanctions they actively encourage or passively condone, who enforces the sanctions, and how they are enforced. Sanctions need to be introduced at the right time, in the right manner, and target the right people, so that they do not unevenly affect those who are already marginalized, but rather protect the majority from those who are unwilling to change despite having the ability to do so.

Hardware supply and financing

Plan International could continue improving supply-side conditions in triggered communities, including testing sanitation marketing in more country programs. In countries where government or NGO subsidies are still present, Plan International can help influence the mechanisms by which these subsidies are targeted to ensure that they do not negate CLTS efforts but rather enhance sustainability of outcomes.

Monitoring outcomes

CLTS programs could consider focusing on routinely collecting household-level indicators of sanitation (including baseline measurements) to measure and recognize incremental progress in communities. Improved monitoring of activities will help generate evidence on the potential, the effectiveness, and the limits of CLTS.

Limitations

Findings described in this report are from a qualitative analysis of CLTS implementation and sample sizes are intentionally small to allow in-depth analysis. Although readers may connect these findings to their own CLTS experiences, they should be cautious about generalizing the findings. Furthermore, researchers visited a subset of communities where Plan International implements CLTS, which means the study may not fully capture all aspects of CLTS implementation in Plan International COs.

Acknowledgements

Authors: Vidya Venkataramanan and Alexandra Shannon, The Water Institute at UNC

This learning brief was made possible through support from Plan International USA, which received a grant from the Bill & Melinda Gates Foundation. The findings in this brief are derived from the Cambodia Country Report, available in our Resource Library at waterinstitute.unc.edu/clts. Data were collected by the Water Institute at UNC with logistical support from Plan International Cambodia. The findings and conclusions contained within do not necessarily reflect positions or policies of the funder, Plan International USA, or of The Water Institute at the University of North Carolina at Chapel Hill.

The Testing CLTS Approaches for Scalability project involves The Water Institute at UNC working with Plan International USA to evaluate whether capacity strengthening of local actors influences CLTS outcomes. Our activities span 10 countries in Africa, Asia, and the Caribbean. More information, project resources, and news are available at the project website.