Sustainability in NGO Programming: A Case Study of Working Equid Welfare Organizations

The sustainability of NGO initiatives has become a criterion by which to judge the success of programming. This study explores the conditions needed in order to achieve sustainable change by focusing on a sector that has recently experienced this shift towards more sustainable, interdisciplinary programming: that of working equid (animal welfare) NGOs. Semi-structured, in-depth interviews with 32 NGO staff working for 8 different animal welfare NGOs across 13 countries were conducted. These interviews covered potential barriers and avenues to success in producing sustainable working animal welfare improvements. Thematic analysis revealed consistent themes across contexts and NGOs. Approaches perceived as creating sustainable outcomes were ones that tailored initiatives to local contexts, focused on changing human behaviour, worked with key community contacts, built capacity in local infrastructure, combined approaches, and had comprehensive exit strategies. Barriers to sustainability included lack of continuity, short funding cycles and unpredictable external factors. Consideration of these conditions for designing effective future initiatives is recommended across the wider not-for-profit sector. Plain Language Summary Sustainability in NGO Programming The sustainability of NGO initiatives has become an aspect on which the success of programming is judged. This study explores the conditions needed to achieve sustainable change by focusing on a sector that has recently experienced a shift towards more sustainable, interdisciplinary programming: that of working equine (animal welfare) NGOs. Semi-structured, in-depth interviews with 32 NGO staff working for 8 different animal welfare NGOs across 13 countries were conducted. These interviews covered potential barriers and avenues to success in producing sustainable working animal welfare improvements. Thematic analysis revealed consistent themes across contexts and organizations. Approaches perceived as creating sustainable outcomes were ones that tailored initiatives to local contexts, focused on changing human behaviour, worked with key community contacts, built capacity in local infrastructure, combined approaches and had comprehensive exit strategies. Barriers to sustainability included lack of continuity, short funding cycles and unpredictable external factors. Consideration of these conditions for designing effective future initiatives is recommended across the wider not-for-profit sector.


Introduction
Across not for profit sectors there is increasing need for sustainability in non-governmental organizations' (NGOs) programming in order for them to be considered effective (Devine, 2003).Sustainability requires the creation of lasting benefits beyond the lifespan of specific initiatives.The cost of initiating programs for NGOs is usually high and failure or premature termination of programs can leave beneficiary needs unmet, while progress on a target problem may revert back to pre-program levels, and trust between NGOs and beneficiaries will be eroded which may affect engagement in any future initiatives (Shediac-Rizkallah & Bone, 1998).
Historically, the remit of many NGOs consisted of a specialist topic within their sector.Their programming reflected this focus and each organization worked independently within their own silo.However, alongside an increased focus on sustainability, many NGOs now acknowledge the connections between different sectors and disciplines, and work more holistically to aid the durability of initiatives (Harris & Lyon, 2013;Moreno-Rivero, 2018).For example the Jane Goodall Institute originally focused solely on chimpanzee conservation but now engages in a variety of projects including microcredit, agroforestry, and youth engagement programs that indirectly impact conservation but are thought to produce lasting change (Johnson & Johnson-Pynn, 2007).While there are positive examples of the integration of conservation with development agendas, reviews of programs caution over optimistic promotions when in reality ''win-win outcomes have proved elusive in conservation/poverty programmes'' (Adams, 2020).Integrated programs have tended to be too ambitious, failed to undertake feasibility studies that highlighted the complexity of linkages, were inadequately financed, lacked technical expertise, and were slow to produce tangible results (Adams, 2020).How widely the mixed results have been disseminated beyond the immediate confines of the conservation sector is interesting to consider given publication of studies over the challenges of holistic programming dating back to the 1990s.This shift is reflected in the new One Health and One Welfare approaches which emphasize the links between humans, animals, and the environment in terms of health & welfare, promoting multidisciplinary working and collaboration between government officials, researchers, and professionals in health, veterinary, environmental, and social sciences (One Health, 2021;Pinillos et al., 2016).
One sector that is well suited to a One Welfare approach and where an increased focus on program sustainability has seen a recent shift towards broader, interdisciplinary working is working equid NGOs.Working equids (donkeys, horses, and mules) fulfill a variety of working roles on a domestic and commercial scale (Pritchard, 2010) and are vital to the livelihoods of their owners.Most working equids are found in low and middle income countries and are owned by some of the poorest members of society, many in the so called ''bottom billion'' (Stringer, 2014).The welfare of these equids is often poor, with wounds, lameness and poor body condition prevalent (Burn et al., 2010;Pritchard et al., 2005;Reix et al., 2014).Working equid NGOs aim to improve levels of equid welfare and in turn strengthen livelihood support for their owners.Traditionally organizations utilized approaches that focused directly on the animal, such as providing veterinary treatment, farriery, or feed supplementation (Upjohn et al., 2014).However, based on independent research and programming evaluations, doubts over the long-term sustainability of these types of approaches have been expressed.Concerns exist over the creation of a culture of dependency upon NGO service provision, the financial viability of these services, the limited numbers of animals reached by programs, the treatment of symptoms rather than engagement with the root causes of welfare issues, and the unintended negative consequences of offering free services on the local animal health infrastructure (such as decline in local business) (Mohite et al., 2019;Pritchard, 2010;Rogers, 2010;Upjohn et al., 2014).
As a result, working equid NGOs are undergoing significant programming changes to work more holistically in order to target the root causes of welfare problems and realize more lasting equid welfare benefits.As they work alongside people, socio-cultural factors have a large influence on working equid welfare (Luna & Tadich, 2019).This has led to the adoption of the One Welfare Framework whereby equid welfare is positioned within the wider context of the communities, livelihoods, and the environment in which they work.This more holistic approach to welfare improvement has resulted in greater interdisciplinary working within animal welfare NGOs, where social scientists now regularly work alongside veterinarians and biologists and links are made to humanitarian and environmental NGO sectors (Johnson & Johnson-Pynn, 2007).
Realizing these new programmatic approaches, with sustainability as a driving factor, is not a straightforward task; many NGOs are delivering their programs in complex real-world environments where multiple dynamic factors interact to influence the target group.There is widespread lack of long-term impact resulting from NGO initiatives across many sectors including food security, community social initiatives and poverty alleviation programs (Rodrı´guez et al., 2016;Savaya et al., 2008;Wabwoba & Wakhungu, 2013).Reviews and analyses have been conducted across NGO sectors in order to identify the factors associated with program sustainability (Ceptureanu et al., 2018;Scheirer, 2005;Wabwoba & Wakhungu, 2013).These have highlighted factors that can influence the likelihood of sustainability within a program at different levels including those relating to the program itself, to the implementing organization and to the community in which the program takes place (Shediac-Rizkallah & Bone, 1998).For programs, a strong understanding of the target community, flexibility, and responsiveness in meeting target group needs and the presence of a program ''champion'' were key to the likelihood of program sustainability (Ceptureanu et al., 2018;Scheirer, 2005).On an organizational level, organizational stability, partnering, and receiving appropriate support from partner organizations were particularly noted (Ceptureanu et al., 2018;Scheirer, 2005).Finally for community factors, community context (which can represent a sustainability risk, particularly where multiple and competing problems are involved), community support, and a sense of ownership were important, cultivated through community involvement in the design, implementation, and monitoring of initiatives (Ceptureanu et al., 2018;Wabwoba & Wakhungu, 2013).
Although these factors are broadly applicable to many NGO sectors, there is little research regarding the factors associated with program sustainability in animal welfare and specifically, working equid initiatives.The objectives of the study were to examine, from the perspective of NGO staff, the conditions that need to be in place at three levels: NGO, community, and programmatic levels in order to achieve programming that creates sustained welfare improvement.The recent shifts in priorities and programmatic approaches seen within the equid welfare sector make it a particularly interesting model in which to do this.

Methods
A key informant interview study design was used.A total of 32 in-depth online interviews were conducted between February and July 2021 with individuals working in 13 countries for 8 different animal welfare NGOs.The inclusion criterion for organizations was that they regularly developed initiatives for working equids and their owners.Initially researchers contacted two known staff members across two international working equid NGOS.These key contacts identified potential participants within each organization and invitations to participate were distributed via email to these individuals by the research team.The key contacts also identified key contacts in other relevant national and international NGOs who then facilitated participant recruitment within their organization.Thus a volunteer sample of participants were recruited through a form of snowball sampling, neither participants nor organizations received an incentive to participate.Organizations were not specifically chosen based on their country of work and no limits were applied to the number of participants interviewed by geographical area.Participants from a range of positions within NGOs were sought, including Directors of Research and Operations, Managers of Regional Operations, Researchers, and Welfare Officers working on the ground in equid owning communities.This range of roles was felt to be representative of the welfare initiative process.Inclusion criteria for participants were that the individual was over 18 and was a staff member of an animal welfare NGO whose role encompassed direct involvement within the welfare initiative process, from the direction and design of initiatives to their implementation in equid-owning communities.On average interviews lasted for 55 minutes (range 30-108 minutes).
Interviews took place over Zoom video conferencing software and were recorded for later transcription.Interviews were semi-structured and were based on a set of interview themes (Supplemental Appendix 1).The themes covered participants' previous experience of the design and implementation of equid welfare initiatives, including which initiatives participants' felt had been most (or least) successful in creating lasting change and any potential reasons for this result.Participants were also asked to discuss the most commonly utilized approaches within their organization and any changes they had witnessed in the types of initiatives employed by their organization.Thematic analysis was applied to interview transcripts to identify common themes; these were examined across both different organizations and job roles.Frequently occurring themes in relation to elements that participants perceived as 1) contributing towards the sustainability of projects or alternatively 2) were barriers to long-term program success.Ethical approval for the research was granted by the University of Portsmouth's Ethics Committee for the Faculty of Science and Health.

Strengths and Limitations
A key strength of the study is the exploration of shifts in priorities and programmatic approaches seen within the equid welfare sector while these are still ongoing.The information remains valuable and useful to other organizations who are undertaking a similar process at a different rate or even to sectors beyond animal welfare whose programming may be affected by similar barriers and opportunities.
Another strength of the study is the breadth of perspectives and experiences presented, of NGO staff with a variety of roles and responsibilities at different levels as well as lengths of time with the organizations involved.The use of organizational gatekeepers in the form of key contacts represents both a strength and a potential limitation.The longstanding collaborative relationships fostered by the authors with key contacts within the NGOs and partner organizations involved enabled trusted access to participants whose opinions are rarely externally documented.However, using gatekeepers to select potential participants has an inherent risk of bias.Also, although participants were recruited from a range of roles and countries all participants, except for one, spoke English during the interview.The lack of participants undertaking the interview in their own language could reflect a selection bias, with non-English-speaking field staff underrepresented.The total number of participants interviewed was 32; as qualitative data provides contextual information in a high level of detail from the perspective of the participant (Hammersley, 2012), the sample size was sufficient.However fewer staff from small, national NGOs took part in comparison to international NGOs and the two types of organization can face different types of challenges with regards to funding, programming and staff capacity.A larger number of participants from smaller NGOs would have helped to further explore these differences.

Results and Discussion
A change of direction: Sustainability was described as being central driver of observed changes to welfare improvement initiatives across organizations.The longevity of the impacts of welfare initiatives now form a key component of the evaluation of their success and sustainability needs to be considered from the outset.''As soon as the funding stops what will happen, is there going to be a sustainable change there, what will we have achieved?''P10.Although the consensus was that sustainability was a factor crucial to successful initiatives, the best way to achieve sustainability remains a source of contention.''[T]he organisation is working towards some kind of sustainable change on some level.And usually everyone agrees that they want to do that, how that happens in reality is another story!''P15.
The features of initiatives that participants commonly raised as important in creating sustainable outcomes and commonly encountered challenges to sustainability are discussed below.

Tailoring the Approach to the Local Context
Program development: At the program level, the need to tailor approaches to each context in which they are implemented was frequently raised in interviews.Due to their versatility, working equids are utilized around the globe across a vast range of environments, cultures and working roles.Many NGOs in the sector work across multiple countries and there was a consensus that transplanting an approach from one place to another without modification was likely to result in failure.''It's just so different, every single community, which I think is important .you don't go out with a model that you're going to fit to every situation'' P31.Particular differences were emphasized in regard to the welfare problems encountered in rural and urban areas within a region (Burn et al., 2010), the roles that equids fulfilled, availability of resources such as grazing, the social dynamics of communities and owner attitudes towards equids.''[W]e try to give room for the community as per the context, we have different approach in the urban intervention approach and the rural intervention approach'' P30.
Community-resources: It was discussed that, to be sustainable, any materials used during education and skills initiatives need to be materials that communities can easily source.Using specialist equipment that was locally unavailable or expensive to purchase reduced the success of initiatives as equipment was later unable to be replaced.''I went back to do a search looking for those people .A few of the tools are there, a lot of them have sold the tools'' P22.''people have hung the shiny farrier tools on the wall as an ornament rather than actually using them'' P5.Similarly, demonstrations of management and treatment using specialist veterinary or commercial products were described as being ineffective, and impossible to replicate.One participant described trialing different ways of demonstrating wound management, with the most effective being wholly adapted to use resources available in the target village.''[W]e bring water from wells around .we bring clean clothes .No, we don't bring stainless steel, clinical grade scissors and forceps'' P29.It was acknowledged that initiatives aimed at veterinary professionals also needed to take into account the feasibility of treatment options based on local resource and supply chain limitation.''It's no good going into a country and training them what drugs to use if they can't get hold of the drugs'' P14.
Community dynamics: Those working in the field noted that successful initiatives had to take into account community dynamics, cultural norms, and beliefs, and to gain an understanding of the motivations behind current management practices.''[T]he cultural diversity in Africa is one of the things that is fundamental .learn the cultures around you and you understand how to approach people.In Africa you don't tell the elders .So how to approach this is a bit tricky'' P22.Cultural norms can require the adaptation of elements of an initiative ranging from the manner in which information is delivered to participants to the detail of technical aspects of harness design.
[W]e were doing a hip strap over the back and I said to the guy ok we are going to cut this 4cm wide and he said ''we can't do that!''I said why not, and he said ''if I put a narrow strap across the back of my donkey everybody will know he's the devil''.It's a well-known thing if a donkey has a thin strap over its back it's a devil donkey, it's got a demon in it.P4 Detailed knowledge and an understanding of the target community is required, described as being best delivered by a local team member or partner who works on the ground, ideally someone from the community itself.
Community needs: Based on experience, participants discussed the need for approaches that were focused on the priorities of the target community.Priority welfare issues identified through veterinary assessments of animals and the issues that owners believe to be priorities can be very different.For example, ''we ranked respiratory condition ninth, the last one.They (community owners) prioritised .respiratory conditions, which we .already recognised the last.They ranked it first'' P29.Participatory methods such as scoring and ranking exercises can be useful to ascertain both the issues that the community deem to be a priority and generate discussion of the reasons behind these choices.Both quantitative and qualitative approaches are recommended at the needs assessment stage to optimize the potential longterm success of an initiative (Upjohn et al., 2013).Owners may prioritize common welfare issues that lead to ongoing loss of productivity as less important than rarer but potentially fatal occurrences such as infectious diseases (Hadrill & Haroon, 1994).Some welfare issues, despite causing chronic pain are so commonly encountered that they are considered to be normal: ''often the kind of reaction I would get from an owner would be well, that's just what happens .all horses or lots of horses here have these kinds of injuries .it was somehow normalized'' P32.In these cases, initiatives designed to target these ''normalised'' issues are unlikely to be perceived as useful by owners and engagement will be low.Similarly, in order to improve welfare, the need for demand creation when training service providers was a frequently discussed topic, especially when creating a service that requires payment.Owners need to understand the associated welfare (and economic) benefits in order to be prepared to invest resources into it.''[I]f you've got these wonderfully trained service providers, but owners that don't see the importance of using them, because Bob up the road is .[X currency] cheaper, then who are they going to go to?''P13.Service providers are often unsure if changes (improvements) will result in a loss of business.''[E]ven if they know they want to do something differently, if it's their main livelihood source, and they're scared of losing it, and the owners don't want them to do it differently to how they did it beforehand, then they're not going to'' P27.

Focusing on Changing People
Human behaviour change: An organizational shift from the more traditional service provision approaches to more human-focused initiatives utilizing human behaviour change (HBC) models was often noted, ''a lot of the problems that .animal welfare organisations have, is that they forget that they're working with human beings'' P21.Realistically any initiative aiming to achieve sustainably improved animal welfare will rely on the facilitation of HBC (Reed & Upjohn, 2018).It was felt that behavioral science approaches can aid with understanding the root causes of welfare problems and matching initiatives to identified community needs.A number of participants utilized models such as COM-B, a behaviour system which proposes that behaviour is generated by three components: capability, motivation, and opportunity; with a range of potential influences between components in the system (Michie et al., 2011).HBC approaches have been extensively utilized in the development and human health sectors (Carney et al., 2016;Gould et al., 2017;Michie et al., 2011) and participants described their flexibility to be applied and adapted to different contexts.''I would say, and this is more from the development and humanitarian sectors, is that increasingly what is being seen as replicable is not the product but the process.So the process of developing participatory community-led work is a thing that can be replicated everywhere.How you go and do it has to be specific to that context and culture'' P7.
Facilitators: At NGO level, one caveat to the success of HBC and participatory approaches was the personal quality of the facilitator and the rapport that they can create with their audience.''It's the practitioner that turns [the approach] into something that makes change happen or it doesn't, whether it's mediocre or amazing.And I just don't think that's given enough credence within the whole .change process'' P6.

The Importance of Key Contacts Within Communities
The importance individual contributions can bring to a project was again highlighted in the context of working with equid owning communities.''You have to have people in the community that are from the communities, that are trusted in the communities with that information, with that understanding of why welfare is important, and take those concepts and .adapt it to their community'' P21.This encompassed roles such as service providers, community facilitators and local individuals employed within NGO projects.It was particularly highlighted that working with key contacts within the community was vital for a number of reasons, discussed below.''[O]ne of the things that [an internal report] found to be repeatedly successful was working with the change agent or community lead or .a notable community person'' P27.
Key contacts: Although time consuming, establishing relationships with contacts enabled NGOs to gain access to communities, to build trust, and was seen as essential for creating successful long-term change.''[W]e try to find somebody that is influential to the community, and that can actually open the community's doors for us .and for them to explain to the community, why we want to go and what we want to do'' P24.Individuals from within a target community were those best placed to enable NGOs to gain a detailed understanding of community context, perspectives and priorities, a sentiment echoed in other NGO sectors (Holmes et al., 2015).It has been highlighted that caution is needed when identifying people with the skills set to foster support for initiatives otherwise programs can inadvertently cause tensions within communities where differences in incomes and education levels can breed resentment for individuals singled out by NGOs (Brown & van den Broek, 2020).However key contacts can encourage community participation and subsequent ownership of decisions regarding initiatives, highlighted as invaluable for programming sustainability.
It also really helped that our community leader in that place .came from the community himself.He was very much a local and he had the full respect of all of the people there .I think the guys in the community were very honest with him because he wasn't an outsider at all.They weren't scared of suggesting things .they didn't feel that they couldn't say ''hey that's not going to work, there is no way,'' so I think that relationship also really helped have those easy discussions.Probably having that sole person in our team was one of the reasons that project was successful because it was their idea and it was him at the front of it and therefore everyone supported it.P2.
Within any community welfare intervention setting (nationally or internationally run) NGO workers are likely to be perceived as outsiders (Donini, 2007).Participants working in the field had noted the difference in relationships between community members and key contacts and community members and NGO workers.''It was almost like when the Westerners turn up everyone was on their best behaviour but when it was the people from [local organisation] there that they have had these longstanding working relationships with, it was a more honest conversation that they were having'' P3.
Local expertise: Although the initial contact person was key in terms of accessing communities, community leaders, and/or authority figures were not the only (or necessarily best) options for making long-term community contacts.''[Y]ou have to .get them on board.But they are not necessarily going to be your community service providers, you have to look elsewhere for that'' P21.Those with a more specific knowledge of equids were discussed such as leaders of local livestock associations or people acting as local equid experts, for example animal healers or farriers.Local experts are in a position of respect and are already used by the community, meaning that setting up local infrastructure and demand for services is not a barrier.Small changes adopted by these 'champions' can therefore have a widespread effect within the community.''[W]hen we upskilled them just a little bit .that did have quite an immediate impact and it was also quite feel-good because we involved people who were already engaged and wanted to be involved and they were really up for learning'' P5.Another group mentioned were enthusiastic, interested community members willing to engage in education and skills training or to cascade welfare messages within the community.This can be particularly important in communities where there are no existing equid related networks.''Sometimes it just really works that you have an enthusiast .the teams work with them to sort of build their knowledge and capacity, and then you instantly have someone within the community that can provide that care'' P31.
It was noted that working in a less cohesive community was significantly more difficult.''[A] lot of the communities we were working with .didn't have a strong community group where I guess normally you'd train someone and they'd have a very good network around them of people that they were going to help.But in [country] that community link wasn't so strong, that was what I think we really struggled to get round'' P2.A lack of community cooperation or cohesion has been previously flagged as a contributing factor to ''hard win'' situations (Haddy, Burden, Fernando-Martinez, et al., 2021;Pritchard et al., 2018).

Capacity Building
In-country infrastructure: In line with the creation of local networks of individuals, organizations described a shift at program level to focus on capacity building of incountry infrastructure.Paralleling the international development sector, rather than relying on external inputs, improving existing animal health services is seen as a strategy to increase the likelihood of creating sustainable levels of welfare (Catley et al., 2002;Pritchard, 2010).Many participants welcomed this change, describing the contrast with previous delivery methods where UK nationals were flown in to deliver training for short periods of time before leaving again.''I think one of the big problems is that a lot of people rush in, do it and then rush out again'' P4.Participants underscored the need for support to be provided over a longer timescale in order to see long-term changes, which is easier to provide though an in-country presence, findings echoed in the environmental sector (Adams, 2020).This approach aims to create a viable network of equid services, such as vets, farriers and saddlers, reducing future reliance on input from NGOs.''Because in the long-term, you then move out and do much less work.In the long-term it's self-sustaining.''P27.To achieve this, international NGOs described an increase in partnering with local organizations.
Train the trainer: Some NGOs focused specifically on building the capacity of in-country animal health infrastructure, through avenues such providing funds for drugs and vaccines and having discussions with governments surrounding systemic issues (such as lack of access in some areas to medicines such as pain relief and antibiotics).Mentoring programs for government vets and nominated individuals within communities were designed to be delivered by local animal health professionals.Utilizing a ''train the trainer'' approach, these networks were designed to cascade information to communities regarding preventative healthcare, welfare friendly management practices and positive human-animal interaction.

Combinations of Approaches Are Most Effective
A holistic approach: When asked about the type of approach that had been most successful in their area, many participants answered that it was not one specific approach that had been successful, as each has its limitations, and so a combination of initiatives was needed.''I think in terms of our approach, they all have to be there.One cannot exist without the other .I think they're all really important together.It's what makes the system work'' P23.As is seen in humanitarian sectors, approaches on the ground to address immediate welfare needs were now being combined with community priorities and those that targeted national and international policy and disease surveillance programs.Synergy: some approaches were felt to be especially complementary to each other and were frequently used in combination.Results of research were seen to inform the direction of practical programming at ground level and were also key to determining and supporting the advocacy avenues pursued.'' [Y]ou can speak to that connection from a point of view of authority because you've got some data there and it's not just anecdotal, not just the crazy horse lady, it's a credible meaningful contribution'' P15.Presenting scientific evidence at meetings was seen as more likely to result in requests being taken seriously and ultimately being successful, especially if the research demonstrates potential avenues for action.''[I]n terms of your advocacy, you can go with answers, you're not just saying change this law or if this is wrong, do something about it .you go with solutions, which obviously kind of makes .your request much more attractive to whoever you're talking to'' P27.
It was also expressed that certain approaches were best suited to different timescales throughout a project.''[T]he thing that really struck me was that all the approaches that you listed have value, it's knowing when they need to be applied and which other of the approaches they work in either order of or in unison with'' P6.The use of participatory exercises to engage communities regarding the welfare of their animals and to identify priority issues and the use of veterinary treatment to address significant initial health needs were highlighted as two effective approaches at the first stage of a project.''But after we slowly shifted .to more community based, you know so we are doing a veterinary clinic and also the community participation work and also started doing a lot of trainings.And once we started doing all this together, that's when I found great success'' P8.Participatory approaches are not without critics, particularly in the development sector (Cooke & Kothari, 2001) where the impact of power relations within communities and the potential for co-option by more articulate and confident community members needs to be carefully planned for.As such participatory approaches have to be engaged with carefully, recognizing their limitations.
This combinatory approach reflects the complex and inseparable linkages between animal and human welfare and both the physical and social environment which is typified by the One Welfare approach (Burn et al., 2010;Haddy et al., 2021).At NGO level, advocacy coalitions have been shown in the environmental sector to be influential in bringing about policy change (Markard et al., 2016) and partnerships with organizations from different disciplines (humanitarian and environmental) were also described as a potential future avenue for on the ground initiatives.''[O]ne of the connections that is really important to make is moving beyond animal welfare .I think animal welfare organisations tend to work in silos a bit'' P15.''[W]e have a plan to approach those multidisciplinary international development agencies in the future'' P30.

Exiting
How to exit: One of the biggest challenges NGOs face, once an initiative is established, is how to withdraw from a community while ensuring the positive effects remain.One participant working on the ground emphasized the need for a gradual exit strategy where clarification is given to equine owners and other stakeholders about the transition process, to avoid confusion and unfulfilled expectations.Exploring the possibility of transferring some service provision to government entities was also recommended, complementing a capacity building approach.A sudden ceasing of all initiatives was perceived as putting staff in a difficult position, being detrimental to animal welfare and being reputationally damaging for the organization.
[S]o many people actually calling me .I've seen this donkey having a car accident and leg broken, why don't you come in actually take it and treat or if not recoverable euthanise it humanely?What do I answer to that organisation or that person if I have been told not to.This is something which is very, very difficult, and so damaging for the charity organisations.P11 When to exit: while most organizations had exit strategies in place for their projects, choosing when to exit a community was still a hard decision.Interviewees felt that there would always be more that could be done and, often after years of work spent getting to the point where communities were engaged, it was hard to withdraw only to start the process again somewhere else where the same results were not guaranteed.There was also concern that communities may regress after a period of time although organizations wanted to avoid creating dependency on their services.''[T]here have been cases in the past few years now, where we're getting reports that .change is not being sustained.So that's the question now, you know, what do we do?Do we go back in?''P27.However, some interviewees felt that expecting communities (and hence welfare) to be the same many years after an initiative took place was not a realistic expectation.''[P]eople think .communities are stationary, our societies are stationary, like okay we'll leave them with this support then we can go, and then five years from now we can go back, and it's all the same-No.People change, communities change'' P21.To ensure sustainability, follow-up support for communities from afar or in a less frequent capacity was viewed favorably.Differences in opinion were seen regarding exiting communities where making a positive impact had proven to be very challenging (Pritchard et al., 2018).Under these circumstances, some individuals felt that exiting the community and applying the resources in a different location would yield a better chance of success.''[Y]ou can find other communities that need the help, and probably they will take it better than that community?So yeah, it's complicated'' P24.Others expressed reluctance to leave a community due to difficulties, preferring to continue working but with different tactics.''[W]e do try as a kind of cultural ethos to not give up when it's hard .just look for alternative approaches'' P27.
Successful exit strategies: The creation of selfsustaining welfare groups was felt to be a way of building knowledge and skills within a group of individuals that could then use their collective power and capital to achieve positive welfare changes in the long term.''[T]hey have a large group of people.So sometimes .they call the doctor 'please come to our village and treat or do a health check-up examination of our animals and we will pay for you'.So they collectively do this.And so this is how they become empowered'' P28.One particular example was women's groups who, once established, could apply for a government grant to assist their activities.The groups focused not only on equine based skills but livelihood diversification in the form of a collective business.''I know that doesn't happen in all the sites, but most of the sites become really sustainable.And we can see changes even after we have exited the group for five years'' P28.Another type of exit was exit from NGO or government partnerships.This was felt to be advantageous in the fact that once an exit was achieved, an organization remained in-country that had positive animal welfare integrated into their work and agenda.''[T]hen we potentially exit from that partnership, hoping that then, you know, the partnership continues its work wherever it's working, and that's sustained'' P27.

Lack of Continuity
One of the attributes that participants identified as needing to be in place for the achievement of sustainable welfare change was continuity within communities, within NGOs and within government.
Within communities: NGOs running programs that required community animal health workers (CAHWs) to run educational and skills initiatives such as farriery, saddler, and vet training, described difficulties retaining trainees both during and after the course.''[I]f we go out there, we try to find somebody that's interested in learning.But .from 10 people that we start training, maybe we are left with one or two, that keep going through the training'' P24.A significant amount of resources are invested in the individuals trained and this high turnover of people can hamper the success of initiatives looking to build capacity within a community.Similar issues have been described in some countries when offering training to government vets as they have little control over where they are subsequently posted and turnover is high.In order to try and maximize retention of trainees within communities some initiatives now take a predetermined ratio of trainees.''[N]ow we need to choose 50/50.If we choose 10 people, five needs to be there from the village and five from the ministry in case if we lose from the ministry .then five are still there in the village'' P22.Others are increasingly selective over the type of individual that they will accept and seek certain qualities that they believe will minimize the chance of losing candidates including specifying an age range, literacy level, equine ownership and demonstration of a genuine interest in equines before being accepted as a trainee.Other organizations focus solely on strengthening the capacity of already existing service providers, which has the advantage of established demand for a service and utilizing reputable individuals.
Within NGOs: Turnover of staff can also disrupt the continuity and implementation of projects across not for profit sectors (Bayalieva-Jailobaeva, 2014;Wiltsey Stirman et al., 2012).Changes in leadership were discussed in relation to the adoption of new directions for future programming; these (explored in more detail in a subsequent article) had the potential to cause tension among staff who were not aligned with new strategies.''Things sometimes nullify the previous work because one CEO comes, then completely changes the way it works'' P11.The addition of individuals with new skill sets was also described as influencing the types of techniques adopted.''There was a change in some of the technical teams, in terms of personnel, and I think the new personnel bought community development expertise in'' P21.
Within governments: when working to make policy changes, continuity in relationships with government officials were seen as necessary to ensure that lengthy processes were followed to their conclusion.''[Y]ou can be working with somebody for three or four years, then comes another administration and you're working with other people in that ministry.And then you're starting all over again'' P21.Changes in government can quickly reverse positive progress and dictates the likelihood of successfully influencing policy, ''[A]nimal welfare was just not very high in the agenda there and you're just not going to get very far when you're trying to do some action in that political landscape.Whereas now we might be looking at more of a receptive administration .you choose and pick your moments as to when you move that forward'' P10.

Short Funding Cycles
Some interviewees also expressed frustration at the length of funding cycles available to undertake projects.''[W]hen you're trying to get funding, I mean, most of the projects are one year, two years, three years .a large project is five years'' P21.Participants acknowledged that creating sustainable change in communities takes time, forming strong relationships, and building trust was felt to be necessary in order to work effectively.It was felt that short funding cycles merely scratched the surface of a problem and would result in little lasting impact.Some interviewees considered that the length of time a project ran was directly proportional to the amount of time that positive impacts would last after the project ended.''What you're talking about is the development model isn't it?10 or 15 years of partnership work and you may be able to show that things are going to carry on for the 10 or 15 years after that.You do something for 6 months you might get 6 months of impact after it.That's just the reality of it'' P7.The difficulty of maintaining projects beyond short external funding cycles has been documented across other NGO sectors, with concerns about program survival after funding cessation a preoccupation for staff (Scheirer, 2005).

Unpredictable External Factors
A fragile process: implementing welfare initiatives in real-world environments means that there will always be uncontrollable factors that impact upon their potential success.Although the factors mentioned varied across organizations, countries, and contexts, the impact of these unforeseen issues on equid welfare was a common theme.''[I]t's not that you're doing anything bad.I mean .it is that there are things that are outside of your control, that you cannot change'' P21.Even projects with fairly established improvements in welfare could see those changes reversed as a consequence of one adverse event.
Variety of factors: natural disasters were described as one factor that could cause an unexpected and widespread downturn in the welfare state of animals, as resources such as feed are destroyed and owners struggle financially.The frequency of natural disasters and extreme climatic events is set to increase and is predicted to disproportionally impact groups such as livestock keepers and those reliant on agriculture (IPCC, 2007;Thornton et al., 2007).It is therefore likely that NGOs working in communities prone to natural disasters will face increased disruption of welfare initiatives and subsequent further need for emergency programming such as feed relief schemes.NGOs with programs in politically unstable states can also encounter barriers such as restrictions on the transport of materials, staff safety concerns and breakouts of violence which can lead to owners being unable to engage with NGO programming.''[T]he areas that we work in are quite dangerous in terms of gang violence.So there was one horrible incident where the inspectors were caught almost in the midst of it all.So that's when we decided we can't actually risk it anymore to go into the areas and hold workshops'' P23.These abrupt changes in welfare can present problems for the evaluation of projects, especially if funding relies on fixed measures of success.Recognition of the context in which initiatives are taking place is important in order to draw accurate conclusions about the progress of an initiative.

Comparison to Factors Identified in Other Fields
The identified factors associated with both success in creating sustainable initiatives and barriers to sustainability in this study spanned multiple levels: NGO, community and programmatic levels.Organizational factors, referred to as ''institutional factors'' or ''factors within the organisational setting'' have also been identified across other fields as impacting initiative sustainability, examples include the maturity and stability of the implementing organization (Bamberger & Cheema, 1990;Wiltsey Stirman et al., 2012).A lack of continuity both in funding cycles and in the form of staff turnover were highlighted by participants in this study as disrupting the sustainability of current projects.Integration of projects with existing programs was also noted as important (Scheirer, 2005;Shediac-Rizkallah & Bone, 1998) which, as working equid NGOs begin to explore more multidisciplinary collaborative projects may need particular attention.
The most striking similarities emerged between sectors when considering community-based factors.Regardless of the type of initiative run (health, development, or environmental) community participation and engagement is vital to initiative success (Bamberger & Cheema, 1990;Ceptureanu et al., 2018;Shediac-Rizkallah & Bone, 1998).Factors such as tailoring the approach to the local context in the form of available resources, understanding community needs and dynamics, and working with key local contacts were factors mentioned by participants across organizations as successful strategies in creating sustainability.Conversely community-based factors including socio-economic inequalities and poor relationships with agencies such as government can reduce initiative success (Ceptureanu et al., 2018) and this was echoed in participant's discussion of difficulties with engaging less cohesive communities.It was also highlighted that external factors operating locally and at national and international levels have the capability to impact project sustainability with politics and natural disasters specifically mentioned (Bamberger & Cheema, 1990), in parallel with points raised by participants in this study.
In terms of the programmatic level, the aspect of program effectiveness (Ceptureanu et al., 2018;Shediac-Rizkallah & Bone, 1998) is one at the forefront of changes towards more sustainable long-term programming for equid welfare NGOs.It is clear that organizations are moving to understand the root causes of welfare issues, utilizing approaches such as theories of human behaviour change in order to avoid making assumptions about the needs of target communities which would reduce the effectiveness of programming.Another aspect addressed that was highlighted in the field of international development is the provision of training within programming (Bamberger & Cheema, 1990), seen here in this study reflected in initiatives that foster skills key to equid welfare within communities and within veterinary training centers in order to build local capacity and strengthen services in a way that promotes long-term viability.

Conclusion
Using working equid NGOs as a case study, conditions needed for the achievement of sustainable welfare change initiatives have been identified.These highlight the need to take into account the human factors that can influence the success of initiatives.Specifically, the need for long-term funding in order to invest in understanding communities and identifying people with whom beneficiaries can have honest conversations; the utility of multiapproach initiatives and the need for tailoring initiatives to local conditions.The consideration of these conditions is recommended for future program planning in order to ensure that welfare improvement initiatives are maximally effective.Due to the broad nature of the barriers and conditions identified in this study, which draw parallels with findings from other fields (Ceptureanu et al., 2018), many of the issues raised are likely to be relevant to other NGO sectors.As such this type of exercise, documenting the opinions of highly experienced staff from a range of positions across related NGO organizations, could be useful in understanding the factors associated with sustainable initiative success in other not-for-profit fields.