Ethics statement

Ethical approval was obtained from the Brighton and Sussex Medical School Research Governance and Ethics Committee, UK (ref. ER/BSMS9D79/1) and the Institutional Review Board of the College of Health Sciences at Addis Ababa University, Ethiopia (ref. 061/18/CDT). All participants were given a participant information sheet (PIS) in Amharic and were required to give written informed consent before taking part, if literacy allowed. If they were unable to read the PIS or sign the consent form due to illiteracy, then the form was read out, and they were asked to give witnessed oral consent. Participants were able to withdraw from the study at any time without detriment or reason up until the point at which data had been aggregated. In the event that any participant was deemed to need immediate assistance or support, for instance because they disclosed mental distress, appropriate action was established according to local resources and capacities.

Study setting

The EnDPoINT study was conducted in three districts (of 12) in Awi zone, Amhara Region, north-west Ethiopia. The Awi zone in the central highlands of Amhara has the red clay soil, seasonal rainfall and altitude that are necessary for the development of podoconiosis,[14] and 87.5% of its population live rurally [15]. The EnDPoINT research team together with the Ethiopian FMOH selected the particular districts for the study based on a combination of their accessibility, previous or ongoing work in the area and their co-endemicity of podoconiosis, LF and leprosy. The pilot study was conducted in Guagusa Shikudad woreda (district).

Study design

This study was guided by both the Medical Research Council (MRC) framework for complex interventions and the Context and Implementation of Complex Interventions (CICI) framework [16,17], building through three phases with an iterative approach, and with “Theory of Change” embedded for the development and feasibility/piloting phases [18]. Further details on these frameworks and how the EnDPoINT project maps onto them are provided elsewhere [19].

Phase 1 of EnDPoINT, which this study was part of, entailed a number of research activities to inform the development of the holistic care package. This care package was based on an established MMDP (physical) self-care package for podoconiosis and LF, but building in mental health and psychosocial components, and including leprosy care for lower limb disorders. This paper reports on one element of these Phase 1 activities: key informant interviews and focus group discussions with stakeholders. These were intended to assess feasibility, acceptability and appropriateness of the draft care package as viewed by the stakeholders, as well as assessing key aspects of the Theory of Change, such as assumptions made, for example the willingness and availability of key stakeholders to participate in and engage with the various care package interventions.

Three focus group discussions (FGD) and 11 individual key informant interviews (KII) were conducted in January and February 2019 by researchers MK, OA and AM, involving 34 participants in total (25 male and 9 female). Three participant groups were recruited: regional health office NTD staff, staff members at Injibara hospital working on NTDs, and community members including affected persons. Purposive sampling with snowballing was used to identify and recruit key stakeholders based on their role/position in the community; the sample size was guided by the number of key stakeholders who were identified to take part and available, as per the original study protocol [19]. Participants had to be at least 18 years of age.

Each of the three focus groups were separated according to a different stakeholder group, as follows: 1) community representatives (kebele administrators, religious leaders, affected persons, health extension workers); 2) health professionals (nurses, health officers, general practitioners and specialists); and 3) decision-makers (NTD focal, NTD officer, TB-Leprosy officer, programme managers). There were three affected persons (two female and one male) in the community representative groups. The sample size during focus groups was limited to ensure that all participants have the opportunity to speak.

The KIIs included participants from the FMOH (officers for leprosy and podoconiosis), staff members from International Orthodox Christian Charities (IOCC), a dermatologist from Felegehiwot hospital, and several focal team members from Awi zone (NTD focal, leprosy focal, mental health focal), staff from the Guagusa Shikudad health office, the head and vice head of the woreda office, as well as the woreda NTD focal person.

Open-ended questions with follow-up clarification and further enquiry were used [see S1 Text for the topic guide]. Questions were based around participants’ experiences regarding affected persons with lower limb disorders secondary to podoconiosis, LF or leprosy. All participants were encouraged to contribute and be heard; the discussion facilitators set ground rules for behaviour to support participants to feel safe; they also actively tried to include all participants in the discussion and to make the participants feel their contribution was welcome and worthwhile. Those participants who were not highly verbal were encouraged to share their views. The interviews and focus groups were conducted in Amharic, audio recorded, transcribed and checked for accuracy. Unique identifying numbers were assigned during transcription, and names of specific people removed if mentioned. A meaning-based translation of transcripts into English was made before coding.

Data analysis

Data were thematically analysed using a combination of manual coding and nVivo software. Two researchers (BD and MK) independently reviewed the data from the focus groups and the first few interviews initially, ensuring detailed knowledge before using within-case analysis, close text analysis and cross-case analysis to develop initial themes. These were then compared and integrated to form a coding framework which was then applied to the remainder of the data. Examples of disconfirming evidence were also sought, as well as comparing responses about different stakeholders. Data collection continued until no new themes emerged.

Confidentiality and anonymity of data were ensured by using allocated identifiers instead of names, and with identifying data kept separately. Data were fully anonymized before data analysis. Data were stored in a secure OneDrive folder with access limited to members of the study team for whom it was essential.

Quotations below are identified as follows: Key informant interviews were labelled K1 –K11 in chronological order of their interviews; focus group participants are identified by the focus group attended and their allocated number at the start of the session. For example, [P2, FG3] was participant number two in the third focus group.

Establishing a need for the new integrated care package

Solutions: integration into routine care

Solutions: integration of mental health and psychosocial care

Historically, routine care has not included psychological aspects of care: “The social, psychological, physical and financial therapy is somewhat forgotten. The biggest gap of the country for not achieving the 2020 goal is also these.” (K10). This lack of psychological care was viewed as a critical deficit, due to an intrinsic link: “Physical impairment brings social impact that leads to psychological impairment. These are linked and working together is very important and effective.” (FG3, P2, 37-year-old male decision-maker). This applies to all three diseases in the integrated care package: “Stigma is so common and bad among all the three diseases” (FG3, P2, 37-year-old male decision-maker); “It is clear that the three diseases need psychosocial support.” (FG3, P7, 49-year-old male decision-maker). Participants supported combining psychosocial care with routine medical care: “While doing the integrated treatment of those three diseases it will be even better to include mental health on it because mental health is mostly neglected.” (FG2, P8, 38-year-old male health professional)

Participants thought that this could be one way of better addressing the problems experienced due to stigma,

“They need to have psychological support in order to avoid the stigma and discrimination they face due to their physical disability.” (K5)

“The disease by itself can bring social discrimination. But counselling and developing their mental capacity during rehabilitation sessions can be effective. So this integration idea is a good opportunity for policy makers.” (FG2, P4, 36-year-old female health professional)

Other ideas on how to specifically address stigma mirror the themes already discussed: collaborative work, “We can control the stigma and discrimination if all of us including those from different sectors like health centres, woreda office or hospitals can take full commitment.” (FG2, P7, 33-year-old female health professional); involving influential community leaders, “if we can use schools and religious organizations” (K8); addressing the knowledge gap: “the stigma and discrimination will change if their knowledge about the thing is advanced.” (K11) and using expert patients: “It is better if those who are cured and saved give witness and awareness about stigma on the stage face to face to the people.” (FG3, P7, 49-year-old male decision-maker).

Solutions: integration of care across the causes of lower limb disorders

The EnDPoINT care package was seen as a way to improve the services delivered for people living with lower limb disorders and consequently their outcomes. These would benefit the affected persons themselves, their families, and wider society. “People were isolated then, we brought them […], and after six months they came to know the disease is not communicable and they [are] able to have family, close friends and even life partners.” (FG3, P2, 37-year-old male decision-maker). Healthcare providers also gained through professional satisfaction in achieving good care outcomes: “Some providers [who] get [made] aware and start treatment and management, [and] start seeing improvement within three months are considering their jobs as blessings.” (FG3, P5, 42-year-old male decision-maker)

The majority of participants were in favour of integrating services across the three diseases due to the shared elements of the disease and treatment process:

“Though there are different positive agents of the three diseases they have many things in common. For example, the three diseases can result in disability, foot care can reduce their burden, all are prone to stigmatization. Therefore, providing integrated foot care is very important” (FG3, P4, 38-year-old male decision-maker)

“Both lymphatic filariasis and podoconiosis manifest with leg swelling and the case management is similar. That means it’s necessary to have integrated management.” (K6)

An integrated service may be easier for people living with lower limb disorders in gaining initial access to foot care services, at a stage when they may not know what is causing their disease: “We told them to accept all cases of foot swelling […] if the cases are not LF or Podo, we will not tell the patients to return to their home but the health professional will give the treatment that he can.” (K4); and may encourage greater engagement: “If three of them are integrated, the health seeking behaviour of the community will also increase.” (K11); “when we are able to give common service for those people without differentiating one from another” (K2).

Most participants thought that integration would allow a more efficient and cost-effective service: “There is nothing to lose by working in an integrated way. We can reach the community with an integrated package of services, this will improve staff efficacy, reduce wastage and redundancy. Generally, for me it is a great way.” (K9). “If we apply these packages together, this is a process by which we can bring big changes. Leprosy, podoconiosis and lymphatic filariasis are foot-related problems, therefore, by providing more integrated service, there is the possibility that integrated care will make us more effective.” (FG3, P1, 32-year-old male decision-maker) Such outcomes would make the integrated care package an attractive option for service provision:

“Decision makers will not oppose if you mobilize and integrate resources and man power because you are going to minimize loss of resources and give care service for many patients.” (K2)

Uniting the separate patient associations may also benefit in the same way: “If the two or three come together, it will play a crucial role in terms of human resource utilization and knowledge transfer.” (P6, FG3); “If patients of all the three diseases come together and establish one large association, it will benefit all patients […] If we integrate it, considering the resource, there is a chance we can get more things.” (FG3, P1, 32-year-old male decision-maker).

Not all were optimistic or supportive of integration across the three diseases. Some were worried that integration may be detrimental in some regards, due to competing priorities when resources are limited, “they may think there will be resource sharing …I also think that way.” (K5). There were concerns that integrating care across the three diseases may negatively affect uptake of care by some with podoconiosis or LF, due to the stigma associated with leprosy.

“If leprosy is integrated with the other disease, those patients taking podoconiosis and LF treatment could dropout.” (K1);

“From what I have seen they may refuse to take the treatment together because they think leprosy is communicable even by sitting together.” (FG1, P2, 59-year-old male community representative)

Others were concerned that the current government health care strands could be problematic for successful integration: “The problem in integrating leprosy treatment with others is the overall government structure problem and it is not designed in such a way.” (K1). This was mostly related to separating leprosy from its current integration with tuberculosis: “We shouldn’t lose our experts…because, we have already integrated TB and leprosy and it has a good outcome… the one who is trained on TB is also trained on leprosy and it demands extra resources if we are trying to separate those providers or experts from the TB program.” (K5). Another participant was worried that there were aspects of leprosy care that were not covered by the integrated package, such as ulcer management and specific medical treatment: “In leprosy, the most probable disability is going to be neurological ulceration rather than lymphoedema.” (K10)

This same participant expressed anxiety that integrated services rather than vertical ones reduced expertise: “There are no vertical clinics in this village. These days every service is considered to be an integrated service […] The health professionals don’t have the ability to detect these diseases.” (K10); and that they were already struggling to provide health care due to lack of resources and infrastructure: “For treating podoconiosis patients, we even beg for materials from the cleaning staff. We can’t even accommodate this kind of minor materials.” As a result: “I don’t see the feasibility of the integration.” (K10), even though they supported the ideal: “Wound care clinics linked with psychiatric care and social work issues done by social workers need to be established”.

Is it necessary and appropriate?

Our findings describe a significant burden of disease related to lower limb disorders including lymphoedema in the endemic districts of Ethiopia studied. This study found that individuals with lower limb disorders suffer a great deal, from the physical aspects of the disease as well as the psychosocial consequences with stigma and discrimination resulting in social isolation for affected persons and their families [20]. These lead to further consequences for their finances and health, in a vicious cycle. This burden–whether physical, psychological or financial–was similar regardless of which of the three diseases had caused the lower limb disorder, providing support for an integrated approach to providing care across the three diseases.

Like other studies, communities and individuals held beliefs about causation that hindered engagement with medical care and which perpetuated the stigma of affected individuals and their families. Consistent with other studies, experience and enactment of stigma was influenced by beliefs about the causes of lower limb lymphoedema, including ascribing the disease to religious retribution, genetics and transmissible infection [21,22]. As elsewhere, this affected interpersonal relationships [23] including divorce and exclusion from social events. Our study highlights the absence of sufficient psychosocial support for affected persons within current care provision. This supports the appropriateness of an intervention that integrates mental health and psychosocial care as well as addressing stigma and lack of knowledge among those living in endemic areas [24].

Some of the challenges noted by our study participants was the spatial distribution of affected persons across large geographical areas in potentially hard-to-reach rural areas, and a general neglect for people affected by lower limb disorders within the health care system. This supports integration of services for lower limb disorders into routine care, particularly within primary health facilities and outreach in the community by health extension workers.

Is it feasible?

Assessing feasibility of the program requires us to identify hurdles and how those can be overcome. Financial restraints, lack of material resources (including their supply chains and appropriate treatment spaces) and training of healthcare workers all need to be prioritised within current healthcare structures in order to facilitate the integration of services within routine care. Although there may be concerns that the package adds a burden of work, both integration of care across diseases and integration of holistic care (as opposed to vertical programmes) may improve the efficiency and cost-effectiveness of care, since resources–financial, material and human–are shared. Reassuringly, although the desire to provide a separate clinic to mitigate stigma would require extra space, this was not seen as a barrier of note by our participants. Collaboration between the health service, charities and community leaders was viewed constructively by our study participants, as urged by Engelman et al [25]; interestingly patient associations were seen as less helpful. However, some caution should be applied due to the limited inclusion of affected persons in this study; further exploration from their perspective on the role of patient associations would be helpful.

This study adds to the existing knowledge about barriers of care by highlighting the dearth of knowledge and experience amongst healthcare workers and care planners. Health care providers were inadequately prepared with either knowledge or resources to be able to manage people with lower limb disorders effectively. This is similar to a recent Rwandan study which demonstrated poor knowledge around podoconiosis as well as minimal clinical experience in treating affected persons [26]; this was compounded by shortages of supplies and drugs as in our study. Another recent study which explored the inclusion of podoconiosis in medical undergraduate studies across all endemic countries in Africa found there was insufficient specific teaching on podoconiosis [27], mirroring our respondents, along with earlier Ethiopia-specific reports [28] of high levels of misconceptions among health care professionals. Education interventions targeted to healthcare workers in endemic areas should be a component of any planned intervention to aid success. One advantage of implementing integrated programmes is that such interventions can be streamlined for training cohorts; within EnDPoINT, for example, health workers receive MMDP training for all three diseases combined as well as mental health training.

Vertical interventions requiring sufficient expertise to differentiate between the causative conditions may also be a barrier to care [29], and one of the benefits of our integrated care package is that it offers holistic treatment to people affected by lower limb disorders regardless of its cause.

Is it acceptable?

Barriers to accessing and ongoing engagement with care are most commonly ascribed to stigma and financial reasons [30,31], however there are other important obstacles. Our study showed that making the intervention acceptable may require a reframing of what counts as “treatment”. The materials as well as the clinical interaction suffer in affected persons’ regard due to their intrinsic simplicity, which results in a failure to engage with care. Non-engagement (with a specific focus on non-use of footwear by people living in endemic areas) has also been explored [32,33], but less is known about other reasons for poor adherence. Misconceptions of care have been touched on elsewhere, such as in Tsegay et al’s study in 2014 where disappointment with the simplicity of care on offer was related to discontinuing care [31]. Changing understanding of their importance and clinical value may facilitate better engagement with care by affected persons. Along with a focus on the “aetiology, preventability and treatability of the disease” [22], this could be included with the education and awareness initiatives. Our participants also felt that engagement would be improved by demonstrating success stories–actively through the use of expert patients [34,35]; and that failure to demonstrate a good response would have a negative effect, another “misconception of care” [30]. Setting realistic expectations for treatment response along with support by those who have been through the process themselves would be valuable in making the intervention acceptable to affected persons and their families.

In contrast to earlier reports of stigmatizing beliefs among healthcare workers [28], the attitudes of those interviewed in this study were much more positive about the conditions, recognised the stigma and were keen to work towards changing this. Participants were enthusiastic that an integrated program would provide tangible benefits for affected individuals and their families, while delivering better and more holistic care to a wider cohort of people living with lower limb disorders, done in a more efficient and cost-effective way.

However, there were also some concerns mentioned by our study participants, particularly in regard to integration of care for leprosy, since: leprosy programmes have traditionally been linked with those for tuberculosis in Ethiopia; leprosy can manifest itself in ways other than lower limb lymphoedema; and misinformation about the transmissibility of leprosy possibly limits acceptability of integrated services by people with podoconiosis and LF. Disentangling leprosy care from its current place within the TB programme may have negative sequelae, such as loss of expertise, and a requirement for extra resources, unless carefully managed. These kinds of challenges and concerns need to be addressed when planning and implementing integrated programmes. Interestingly, since completion of the EnDPoINT project, the Ethiopian Federal Ministry of Health has included leprosy into the priority NTD list /group in the 3rd National NTD Master Plan.

Strengths and limitations of study

The inclusion of a variety of stakeholders from an endemic region, with the combined use of focus groups to stimulate discussion and in-depth interviews to probe in depth were strengths of this study. This study is part of a bigger project underpinned with implementation research concepts which will allow a wide range of mixed method approaches to explore fully the processes and outcomes involved in integrated care. Furthermore the inclusion of implementers and policy makers alongside other stakeholders as part of the consortium means that they have been kept abreast of research findings throughout the study and any concerns raised have been fed into later stages of the project.

There are limitations to this study. Firstly, this study was conducted in one area of northwest Ethiopia and the results may not be generalizable to other settings. Secondly, the data comes from self-reporting rather than observed or witnessed behaviours. Thirdly, due to language limitations, one of the two researchers analysing data could not listen to the audio recordings; however, translations were done by the second researcher, who also conducted the interviews and contributed to data analysis; and we went back to the original data to settle queries about nuances. Due to the phased nature of the overall project, data collection in this study was limited by time constraints rather than primarily by data saturation.

An additional consideration is the limited inclusion of affected persons as participants. This was compounded by the merged nature of focus group 1, which held both affected persons and community members. While this could stimulate discussion, there is the risk that, despite the ground rules and the facilitators, affected persons may have felt restricted from fully sharing their experiences due to the presence of those by whom they may be stigmatized. In retrospect, including affected persons within the key informant interviews could have offset this, as well as recruiting more affected persons as participants.

The EnDPoINT care package was restricted to lower limb disorders and co-morbid mental health problems caused by podoconiosis, LF and leprosy. It may be interesting for future research to explore the feasibility and effectiveness of broadening the provision of integrated care to limb disorders caused by other non-communicable diseases, for example diabetic foot.

However, despite these limitations, the results provide support for the roll out of an integrated care package in the endemic areas of Ethiopia where the study was conducted.