A Qualitative Study Exploring Barriers, Facilitators and Solutions to Equitable Coverage of Preventive Chemotherapy Towards the Control and Elimination of Neglected Tropical Diseases in South Omo Zone, Nomadic Setting of Southern Ethiopia: Stakeholders’ Perspectives and Experiences

Despite substantial progress made towards the control and elimination neglected tropical diseases (NTDs) in Ethiopia using preventive chemotherapy (PC) for lymphatic lariasis, onchocerciasis, schistosomiasis, trachoma, and soil-transmitted helminths, its implementation is facing challenges to achieve equitable coverage especially under mobile populations. This study, therefore, aimed at exploring multiple stakeholders’ perspectives and experiences on barriers, facilitators, and potential solutions to equitable coverage of PC in a mobile nomadic setting of Southern Ethiopia. Methods This study was conducted Omo 2019. Qualitative data collection methods involving key informant interviews (KIIs) and focus group discussions (FGDs) were used to collect data from a total of 27 KIIs and 16 FGDs in a nomadic community of Southern Ethiopia using a semi-structured tool. Data were analyzed using a thematic approach, supported by NVivo 11 software.


Introduction
Neglected Tropical Diseases (NTDs) are a group of communicable diseases, disproportionately affect 1.5 billion people globally in 149 countries and territories [1,2,3]. About 90% of the total NTDs burden is contributed by ve diseases ─ lymphatic lariasis, onchocerciasis, schistosomiasis, soil-transmitted helminthiases, and blinding trachoma [4]. These diseases impose substantial health and socio-economic burdens including physical and intellectual impairments, school non-attendance among children, and reducing economic productivity [4,5]. The African continent bears 39% of the total global burden of NTDs [3], and Ethiopia has a high burden of NTDs in sub-Saharan Africa (SSA) [6].
The World Health Organization (WHO) recommends periodic, inexpensive, and safe preventive chemotherapy (PC) as the main public health intervention for the control and elimination of NTDs [7]. Out of the 1.5 billion populations affected by NTDs, 90% is accounted for by PC-NTDs. And up to 44 countries in the African region are endemic for at least 1 PC-NTD, 42 for at least 2 PC-NTDs, and 17 for all the 5 PC-NTDs [3]. An estimated 5.9 million years of healthy life were lost in SSA due to schistosomiasis (SCH), onchocerciasis (OV), lymphatic lariasis (LF), and soil-transmitted helminthiasis (STH) in 2013 [5].
Epidemiological studies revealed wide geographic overlap among these diseases, particularly in disadvantaged populations with limited access to health services and sanitation [8].
Ensuring equity in the prevention of NTDs is crucial to reach NTDs elimination goals, which is greatly linked with sustainable development goals to ensure equity of health service access in Goal 3 (target 3. 3) as well as to inform universal health coverage (UHC) [9,10]. It is often expected that NTD interventions are equitable and meet the needs of all groups in the population because efforts are already targeted at populations that are largely poor and rural. However, segments of the population remain not reached and there are areas of persistent disease transmission because of inequity of PC coverage [11][12][13]. While focusing on PC, global NTD initiatives have largely ignored other manifestations of neglect, such as equity and social determinants of health [6]. PC coverage could be affected by a range of individual, household, social, cultural, economic, and health system factors that could challenge the implementation of PC programs [4].
Ethiopia has been implementing PC as the main strategy to tackle PC-NTDs, and recent reports of drug treatment coverage showed remarkable progress ─ reaching a signi cant proportion of people in need [13,14]. However, routine therapeutic coverage reports of MDA showed low coverage of PC at South Omo district, a nomadic setting of the South Omo, as reported by the Ethiopian Federal Ministry of Health (FMoH) in 2018 [15]. This low coverage could be attributed to various challenges invisible to implementers. However, little evidence exist on these potential barriers to equitable access and use of PC at South Omo in Ethiopia particularly in nomadic and hard-to-reach areas. This study, therefore, aimed to explore stakeholders' perspectives and experiences on these gaps and participatory proposals on facilitators and solutions suggested by these stakeholders. We hypothesize that communities affected by PC-NTDs in the South Omo zone have inequitable access and use of PC services due to barriers ─ at individual, community, household, and health system levels.

Study design and period
A qualitative study was conducted August 2019 that involved both KIIs and FGDs to gain an in-depth understanding of stakeholders' perspectives and experiences on barriers, facilitators, and solutions to equitable access and use of PC. KIIs and FGD sessions are reported according to the consolidated criteria for reporting qualitative research framework [16].

Study setting
The study was done in the South Omo zone, Southern Ethiopia. The population in the zone is estimated to be 577, 673 (288, 638 males and 289, 035 females), according to the central statistical authority (CSA) in 2007 [17]. The zone has nine districts (which consists of both agrarian and nomadic communities). All PC-NTDs are known to be endemic in the zone and preventive chemotherapy has been put in place in the health system to tackle PC-NTDs [13]. The town, Jinka, is located at 770 km to the southwest of Addis Ababa or 540 km from Hawassa, the capital of Southern Nations, Nationalities and Peoples' Region [18].

Study populations
The study populations include community members who are involved in at least one round of MDA, community drug distributors (health extension workers and community members), and teachers in focus group discussions. On the other hand, all potential key informants from the community, health system, and other sectors were involved in the key informant interview. The selected key informants were NTDs focal points, NTDs program managers, health facility head, primary health care unit-in charge, the village chief, school director, and head of women and social affairs.

Inclusion and exclusion criteria
Community members who have received at least one round of MDA but preferably 2 treatment rounds of PC treatment and community drug distributors (teachers, health extension workers, and teachers) participated in the focus group discussion. Any individuals who were a potential source of information regarding MDA participated in the key informant interview. On the contrary, individuals with a serious illness during the survey were excluded.

Sample size and sampling technique
Participants were recruited from seven districts. First, lists of districts with low coverage of PC and or with a history of MDA interrupted in 2019 were identi ed from our earlier baseline study as well as from record data of the South Omo zone health department. Then, by taking information saturation and homogeneity and or heterogeneity and locations (urban/semi-urban/remoteness) into consideration, sampling was performed. A total of sixteen FGDs and twenty-seven KIIs were conducted. Averaging at least 8 participants per session in FGD were involved. Due to differences in coverage of PC between males and females, gender norms were considered as criteria for the composition of groups. Moreover, the number of participants in FGDs and KIIs was determined based on their relevance to the study.

Study variables
Variables included in this study were barriers, facilitators, solutions to equitable access and use to PC for PC-NTDs, and availability, accessibility, acceptability, and effective coverage were taken into consideration to collect data on equity of PC.

Data collection
Key informant interviews and focus group discussions were conducted using semi-structured interview guides. A team of experienced qualitative research experts with social and public health professionals collected the data. All materials used for the study were developed in the English language, and then translated into the local language and then back to English. The FGDs and KIIs were conducted at places where convenient for discussion and interview after working hours. The FGDs were led by a moderator and a note-taker using an FGD guide and a voice recorder was conducted at o ces or places where the suitable place for an interview. Also, an interview guide was used for KIIs.
The FGDs were held after the completion of any key informant interviews; and consideration was given to keep homogeneity of participants. This is due to our assumption of certain issues arising during KIIs that could lead to modi cation of the composition of the groups and types of issues to be addressed Data management and analysis All KIIs and FGDs were audio-recorded, transcribed, and translated verbatim into English by the data collectors who had conducted the KIIs or FGD immediately after data collection. Thematic approach, a method for identifying themes in the transcripts were used for data analysis, using NVivo 11.0 software [19]. After an initial reading of the transcripts, all interviews were coded initially for emerging core descriptive content; these themes are discussed in greater detail in the results section.

Operational de nitions
Equity in health indicates that ideally everyone should have a fair opportunity to attain their full health potential, and no one should be disadvantaged from attaining this potential [22]. Hence, in this study, equity is de ned as anyone eligible for PC should not be left behind from accessing and use of PC.
Effective PC coverage is de ned as people who need health services get them in a timely fashion at a level of quality necessary to obtain the desired effect [20].

Data quality
Data quality was kept throughout the research process through pre-testing and standardizing tools; training of data collectors and supervisors; and daily checking of consistency and accuracy of data were made by supevisors.

Community engagement in the research process
The research process involved community representatives during gaps identi cation and data collection. Health professionals and social workers from the community were involved in the data collection.

Socio-demographic characteristics of participants
A total of 155 participants (27 KIIs and 16 FGDs) were involved from 7 districts, mainly from nomadic and remote settings. Of the 155 individuals, 128 participated in 16 FGD sessions, averaging at least 8 individuals per session. Participants in KIIs were primary health care unit-in charge, district NTDs focal points, zonal and regional NTD program managers, health center heads, village chief, school directors, and women and child issues department heads (Table 1). FGDs participants were community members (anyone who received at least one round of MDA including religious leaders and school-aged children), while drug distributors (CDDs) were health extension workers, health development army, and teachers ( Table 2). the perspectives and experiences of stakeholders who participated in the MDA process. In cases where views are speci c to particular participants, these are noted within the manuscript.
The emerged barriers to equitable access and use of PC were lack of resources ( nancial, human, communication and vehicle), low community drug distributor's motivation; inadequate information, mobilization and community engagement, lack of awareness, inappropriate scheduling of MDA, mobile nature of nomadic communities, security challenge, low school attendance, drugs distribution at central point, scatter pattern of households, misconception, misbelief, distrust of government, lack of PC program ownership; rumor and drugs' side-effects; and weak planning, leadership, supportive system, and coordination (Table 3).
On the other hand, the emerged facilitators to equitable access and use of PC were the presence of community structure (health development army), provision of drugs free of charge, availability of community volunteers, the presence of community health workers, partners support to train drug distributor, the presence of the religious and cultural organization, and decentralization of government structure to smallest point (kebele) ( Table 3)  "…People thought as drug drugs are provided for political purpose, and they did not take drugs due to distrust of the health system and government…" [Community member, SAC] Low attention is given to NTDs and PC "…PC intervention has been implemented for many years, however people are saying other interventions, such as WASH and vector control should be implemented. For example, 50% of the community did not have clean water; many people are taking water from river…" Planning, coordination, and supportive system Study participants mentioned that access and uptake of PCT were affected by planning, coordination and supportive system. Lack of motivation "…The per-diem we have been paid was inadequate ─ we are working in hardship setting, even we did not have money to move from pace-to-place. We only received a small per-diem during training. So, this does not motivate us to reach all segments of our community…" [CDD, community member] "…CDDs leave out from the MDA role mainly due to lack of per-diem, which resulted in CDDs not get motivated…" [Key informant, health center head] "…In the school-based deworming, teachers were not collaborative as they seek incentive…" [Drug distributor, HEW] "…We are not motivated by leadership of the health system; the per-diem payment was unfair, not paid equally for equal work. We worked hard but those who were not on work more paid than us..." [CDDs,

HEW]
"…CDDs usually leave the program due to lack of incentive…" [Key informant, NTDs focal point].

Facilitators to equitable access and use to PC
Availability of drugs at no-cost in most endemic districts, presence of community structure, partners support to train CDDs, experience on MDA, and availability of community volunteers were important facilitators to equitable access and use of PCT. The following anecdotes vividly illustrate this. Our study revealed that lack of resources ( nancial, human, communication, and vehicle) affect the equity of PC coverage in terms of drug availability and CDDs' motivation. For instance, due to lack of nance, CDDs pay out from their pocket for transport and phone expenses ─and this affects their motivation to reach around their communities. The health system should support CDDs to keep their motivation through incentives and supervision. This agrees with the ndings of systematic reviews and studies done in Kenya and Bangladesh [21][22][23][24].
It is known that the coverage of PC could be limited by the number of people who can access it. This study revealed that inadequate information, mobilization, and community engagement adversely affect equitable access and use of PC. This is possibly happened because limited access and use of information and community engagement hinder people to get informed and know about the purpose of drugs. This nding is consistent with the nding of a study conducted in Bangladesh, which stated that "communication and information gaps between CDDs and community people were found as a barrier to the utilization and coverage of MDA" [23]. Also, a study conducted in India corroborates the ndings of our study [21]. Moreover, studies conducted in Kenya and Zambia revealed that contextually-appropriate community engagement strategies were vital for the uptake of drugs by encouraging a sense of program ownership, trust, and demand [22,25]. Further, scheduling MDA also affects the access and use of PC due to lack of treatment round schedule that meet busy working hours of the community. This is in line with the study conducted in Coastal Kenya [26,27]; and evidence explored by researchers stated that "schedule of MDA that do not match communities' work time schedule affect MDA" [28].
On the other hand, though resources are available and accessible, drugs may not be used if the population does not accept them. Hence, acceptability including non-nancial factors, such as culture, beliefs, religion, and gender aspects that relate to people's perceptions of the worth of health services impact the equity of PC [29]. Our study identi ed that perceiving deworming drugs as safe was important for MDA uptake. On the contrary, fearful perceptions and beliefs about PC were fueled by rumors disseminated in the community. This nding is supported by the nding of studies conducted in Kenya and Bangladesh, where perceptions and experiences related to side-effects of drugs affect the uptake of drugs in the community [22,23]. Moreover, the ndings of our study suggest that misconceptions and misbelief adversely affect equitable access and use of drugs. This might be due to people perceived as drugs are provided for family planning and it may cause infertility and provide for an evil purpose.
Finding our study is consistent with studies conducted in Coastal and Western Kenya [26,27,30,31].
Also, in this study, participants revealed that lack of knowledge on NTDs and PC made it di cult for them to accept the medicines [26]. Additionally, our study found that poor planning in terms of adequate time and schedule affect the MDA coverage since CDDs could not get people at home. This result is in line with the nding of other studies [26,28]. Furthermore, this study explored that uptake of PC can be affected by mistrust of government related to the drugs. This nding is consistent with the nding of a study conducted in Western Kenya [32].
Likewise, our study explored that security challenges hinder accessibility to reach communities affected by NTD. This is because the capacity of health service is limited by the number of people who can reach and use it. This nding is supported by a study conducted in Western Kenya [27]. Furthermore, our study also revealed that the movement of people from place-to-place, scatter pattern of HHs, and low school attendance of children as potential barriers for equitable access of PC in the nomadic setting. This nding is mainly because nomadic communities have a habit of moving place-to-place and a culture of allowing children to keep cattle at le while CDDs visit their home. Also, the distribution of drugs at a central point in these settings could not be a good platform to reach those who require drugs. This nding is supported by a study conducted in Bangladesh [23].
Similarly, our study revealed that effective coverage of drugs was affected by weak planning, leadership, and support system and coordination. This nding is consistent with the nding of a systematic review and meta-analysis and a study conducted in Tigray [31,33].
On the other hand, ndings of our study suggest the presence of community structure (health development army), presence of community health workers, awareness creation, religious and cultural organization, and availability of partners to train drug distributors and deliver drugs for free acted as drivers to equitable access and use of PC. This nding is consistent with the nding of systematic reviews [27,32,33]. Most importantly, awareness creation through community participation would be impactful to improve uptake of PC. Our study also revealed decentralization of government structure to the smallest point (kebele) and support system facilitate uptake of PC. The ndings of our study are consistent with evidence explored from reviews of studies conducted in different countries and in Kenya [24,25,33].
Concerning limitations of the study, rst the purposive selection of study participants might resulted in selection bias. Secondly, social desirability bias may perhaps affected the response of the participants. On the other hand, the study could have been strengthened by conducting both focus groups and key interviews to participate all types of stakeholders.

Conclusions
This study explored potential barriers that could affect efforts in equitable implementation of MDA. Also, facilitators and solutions were revealed to ensure no one left behind from PC in nomadic and remote settings of Southern Ethiopia to inform PC program. Given NTDs are targeted for elimination by 2030 as part of the sustainable development goals; implementation of PC should consider the setting of the interventions particularly in nomadic and remote areas to reach all segments of the populations. Most importantly, information, mobilization, and community engagement; the mobile nature of the nomadic community, and the high level of low school attendance should take special consideration when MDA is planned in these settings. Moreover, the impact of combined MDA models (school-based and communitywide) with tailored community mobilization and awareness campaign on coverage of praziquantel and mebendazole or albendazole in nomadic settings need to be studied. The South Omo zone health department should work in collaboration with partners and the community to address the barriers explored. Further, our study ndings may be useful in improving implementation of MDA in other similar settings.

Declarations
Ethics approval and consent to participate Ethical approval was granted from the Institutional Research Ethics Review Board of Arba Minch University, College of Medicine and Health Sciences, Ethiopia (reference number: CMHS-12032750/111). Participation in the survey was voluntary, and oral and written consents were obtained from study participants before the data collection. Upon arrival in the district, there was a meeting with the district's health o ce head and village chief, where the purpose of the survey was explained and verbal permission was obtained to perform the survey. Participants in KIIs and FGD sessions were given brief health education on the importance of regular periodic deworming, environmental sanitation, and personal hygiene to prevent NTDs in their community at the end of data collection. However, there was no incentive reward for being part of the survey, but participants in the FGDs received a bottle of soft drink, which could worth $12 (412.36 ETB).