Introduction

Trachoma is an ocular infection caused by Chlamydia trachomatis and is a neglected tropical disease found predominantly in developing countries [1]. As of 2020, the World Health Organisation (WHO) estimates that 137 million people worldwide are infected with trachoma, making it the leading cause of infectious blindness globally and a public health priority [2]. The SAFE strategy, which stands for Surgery (S), Antibiotics (A), Facial cleanliness (F) and Environmental change (E), was devised with the goal of eliminating trachoma by 2020[3]. Despite this, the disease is currently still endemic in 44 countries [2].

Surgery has been repeatedly demonstrated to be beneficial in preventing the worsening of trichiasis and preserves visual acuity, as well as reducing pain and associated symptoms [4,5]. Antibiotics in the form of mass drug administration (MDA), a common method of drug delivery for neglected tropical diseases, is effective in reducing the prevalence of active trachoma and therefore the transmission of the disease [6–9]. The F and E components of the strategy focus more on the primary prevention of the disease and relies on broader community hygiene interventions and behavioural change [10]. There is evidence for the benefit of improving water supply and sanitation in reducing the risk of trachoma [11]. A few studies have found that having health education on improving hygiene practices are correlated with a lower trachoma prevalence [12,13]. In various countries where the full SAFE strategy has been implemented effectively, trachoma has often successfully been eliminated, with the most recent success being in Ghana as of 2019[14]. However, one of the caveats to the effectiveness of the SAFE strategy is that all four components need to be implemented simultaneously in order to successfully eliminate trachoma.

Aspects of the SAFE strategy have been successful in some districts in Tanzania [15]. However, it has been recognised that some districts with marginalised communities, mostly consisting of the Maasai populations, have not seen the same success [16]. In some of these predominantly Maasai districts, the prevalence of active infection, including the presence of follicles and papillary inflammation, has been found to be 33.6% and 31.6% respectively in children aged 6 to 10 years old [17], which is far greater than the WHO elimination threshold of <5% [2]. The Maasai ethnic group occupies the northern regions of Tanzania and the southern regions of Kenya [18]. Many of the Maasai people still live by their traditional pastoralist lifestyles, rearing and living in close proximity to their cattle, as well as following a semi-nomadic way of life.

Trachoma control programmes in Tanzania are implemented by the Tanzania National Neglected Tropical Diseases (NTD) Control Programme in collaboration with non-governmental organisations (NGOs). The organisations might only support one of the components of SAFE, meaning that the Tanzania National NTD Control Programme coordinates activities to ensure endemic districts are covered with applicable components of SAFE [19].

A MEDLINE literature review has shown that there is a gap in the literature, as there are few previous studies investigating the perspectives of NGOs on the implementation of trachoma interventions. One study looked at the barriers to implementation of SAFE in Australia in Aboriginal communities from the perspective of healthcare staff, however, owing to the completely different context and environment, it is difficult to say whether the findings of the study are generalisable to the Tanzania context [20]. Therefore, this study aims to understand programme implementation by NGOs on a broader level in these communities in Tanzania. The study explores the facilitators and barriers of implementing trachoma interventions in the context of marginalised communities in Tanzania, which are predominantly inhabited by the Maasai, the effect of having multiple actors involved in delivering trachoma interventions and the factors that influence decision-making relevant to the programme.

Methods

Results

The participant demographic that this study aimed to include were primarily those running the NGOs, such as programme directors, managers, or regional technical advisors. A summary of the participant demographics is provided in S1 Table. Eighteen organisations were identified to potentially suit the study criteria and seven organisations participated in the study. The remaining organisations did not participate in the study for the following reasons: seven did not respond; two refused to participate and two were not able to be interviewed due to time or technical difficulties. No participants were excluded due to language skills, as they all had working levels of English. However, on snowball sampling, the majority of the participants reported that there were only a few organisations implementing trachoma interventions in Tanzania at the time. The number of interviews was therefore deemed representative of the current key stakeholders, especially since data saturation was reached. Although this study set out to understand participants’ experiences with marginalised communities in general, the responses given by participants were focused heavily on Maasai communities. This is likely due to the fact that the endemic areas are populated mainly with Maasai populations, and thus NGOs targeting these areas are currently facing problems adjusting to their unique livelihood and cultural background. This explains why the participants yielded responses primarily focused on the Maasai, despite the study looking at marginalised groups as a whole.

The study found four key themes affecting trachoma intervention implementation: 1) contextual factors; 2) trachoma-specific factors; 3) decision-making factors and 4) the effect of multiple actors.

Discussion

Many of the NGOs felt that the understanding and treatment of the disease was often poor, which acted as a barrier to implementation. NGOs felt that this poor understanding contributed to the fear of surgery within the Maasai. However, it is important to understand the basis of that fear, rather than using the general concept of ‘fear’ as a blanket statement to explain refusals in a simplistic manner, in order to be able to help the community overcome their fear of the treatment. Within Maasai communities, the fear was often not of the surgery itself, but rather a fear of the consequences associated with surgery, such as fear of long recovery times, fear of pain after surgery or fear of blindness after surgery [22]. Additionally, the poor understanding of trachoma as a disease, where active trachoma and trichiasis were not associated with each other owing to the lengthy time-lag contributed further to difficulties with trachoma interventions. Within the Maasai culture, healthcare decisions are often made by a family unit, rather than solely dependent on the wishes of the individual [23]. This therefore highlights the need for improved counselling services to explain the practical aspects of surgery or any concerns and should be done with family members that are of age. This is particularly relevant if the person affected with trichiasis is female, since males are the predominant decision makers in a domestic unit [24]. The majority of counselling should done by the surgical outreach team early on and there should be at least one person on the team dedicated to counselling to render the service more focused, to reduce the likelihood of counselling being evaded, and to instill a sense of responsibility in the counsellor to ensure that potential patients and their families are well-informed. Ideally, this role should be undertaken by someone to whom the other members of the community can relate and who speaks the local language proficiently. Asking members of the community who previously have had positive experiences with surgery to become ambassadors may be an effective way of advocating surgery.

One study looking at the Maasai perspective of trachoma found that the Maasai recognised that western medicine was effective [25]. This was contrasted by the findings of this study, where NGOs perceived that the Maasai were skeptical of, and therefore reluctant to receive western medicine. This may be due to miscommunication on both fronts and the community’s poor understanding of NGOs’ goals [25]. Because many of the workers that implement trachoma interventions are not necessarily Maasai, implicit preconceptions of the marginalised groups may have been made, not out of malicious intent, but derived unconsciously. While pragmatic issues during interventions certainly exist, such as language barriers or travelling to remote areas, these preconceptions may further negatively impact the intervention provided, by affecting judgement and altering non-verbal behaviour subconsciously [26], and impact the receptivity and acceptability of the interventions to marginalised populations.

A systematic review demonstrated that the levels of implicit bias exhibited by healthcare professionals are similar to that of the general population [26], which may be mirrored when implementing trachoma interventions. Seeing as the Maasai are relatively disadvantaged as an ethnic group in terms of health compared to other surrounding ethnic minority groups [27], it is important to minimise “corrosive disadvantage”. This is where an already vulnerable group is further disadvantaged as a result of current disadvantages [28]. A study in Tanzania, Kenya and Namibia assessing stigma exhibited by tuberculosis health care workers on patients with HIV established that teams that received regular supervision as well as training to better understand patient context and ways to recognise implicit bias improved care [29]. Having regular training sessions may therefore help health care or educational personnel to recognise and reflect upon any pre-existing ideas or misconceptions and improve the dynamic between the implementing teams and the community.

NGOs described that simply providing behavioural change interventions is not sufficient, in both school-based and community-based interventions, if there are no facilities to actually carry out the change. Partners recounted various difficulties in funding, logistics, time and manpower as the main obstacles for erecting sufficient infrastructure, such as latrines or boreholes, for good hygiene practices. However, there is a certain degree of cognitive dissonance with this mode of reasoning, as it fails to take into account the biosocial needs of the community. In practice, even in the instances where such infrastructure has been built, explanations for why or how this equipment should be used is not always provided, giving little rationale for members of the local community to utilise them. Additionally, focusing too much on an immobile option to improve sanitation overlooks the pastoralist nature of the Maasai. These issues have been reflected in the literature, as there have been instances where simply increasing the number of latrines or other facilities does not improve hygiene outcomes, even when provided synergistically to a behavioural change water and sanitation programme (WASH)[30]. A review by Kamara et Al. showed that when behavioural changes do occur, it has been associated with improved hygiene outcomes, however, actually sustaining these behavioural changes in the long-term has repeatedly proven a challenge [31]. The pinnacle is that it does not appear that the lack of infrastructure is the main barrier to partners for low levels of sanitation in the communities. Ultimately, it is more important to find other options to better suit the culture of the Maasai, as well as determine superior ways of instilling and maintaining behavioural change for good hygiene practices through education, and further research would be needed to identify the best method to do so.

In Tanzania, the current approach for behaviour change regarding WASH uses community-led total sanitation (CTLS) and is implanted by the governmental sanitation programme. This approach focuses on evoking strong emotions to trigger the community to take responsibility for the cleanliness of their environment [32]. Although CLTS has been globally recognised as a relatively inexpensive form of sanitation programme and has been found to reduce the prevalence of open defecation in some communities, the widespread benefits that have been reported are often overly optimistic [33]. However, the very nature of the programme, in using emotions such as shame or disgust, may result in the community conforming to standards that may not necessarily be compatible with local culture or the socio-economic and political aspects of the local community [32–34]. There has also been discussion as to whether this type of programme propagates further stigmatisation of marginalised groups [35]. Such approaches therefore may not be the most appropriate initiative for marginalised communities, particularly in groups like the Maasai, who hold their cultural and traditional beliefs very strongly.

NGOs and the government should collaborate to gain further insight in the needs and the dynamics of the local communities, either through empirical experience or further research, in order to improve the acceptability of the national sanitation programmes in these marginalised communities. There should either be a change in the approach that the current CLTS programme uses, where the local context can often be neglected due to the lack of tailoring in the programme, or a separate guideline conscious of the context in these communities should be developed. Alternatively, focusing on other types of sanitation programmes, such as school-based interventions, may also contribute to more successful outcomes. School-based interventions have been described to be somewhat effective and children that attend school have a greater awareness of general sanitation [36]. One study found that in rural Tanzania, with a predominant Maasai community, mothers with a higher level of education are significantly more likely to have better hygiene practices and their children were less likely to suffer from diarrhoeal illness [37]. It has also been repeatedly found that households with better-educated caregivers maintain a higher standard of sanitation in Sub-Saharan Africa [38]. This suggests that the effect of school-based interventions may instil better hygiene habits in an intergenerational manner.

Multitudes of healthcare models and determinants of health have been developed in order to offer guidance in the programme design, but in reality, only a minority use a model as a basis for their intervention, especially in resource-limited settings [39,40]. Considering the strong cultural beliefs and the nature of the Maasai, the design and implementation of a behavioural change intervention would be most effective if underpinned by a behavioural model that considers the contextual, psychosocial, economic and political context. Having sufficient community engagement in both the planning and implementing phase should also be regarded as a priority, in order to tailor programmes and minimise issues related to the perception of power imbalance and control between the outside “implementers” and the “community”[41]. This empowers the members of the community to advocate for their own hygiene and sanitation. By extension, this would also reduce any unconscious preconceptions and implicit bias from an external implementor.

Within the community, the majority of patient recruitment for trachoma interventions are done by community health volunteers. Although there often is an expectation for individuals to take responsibility for the health of their community, and there is encouragement to recognise that “these are their issues and they need to fix it themselves”, the lack of financial remuneration may result in lower quality recruitment and other aspects of intervention [42]. In the Morogoro region, there was a high spirit to volunteer and improve community health, but the inability to generate enough income to survive, owing to the volunteer nature of the community health worker as well as the opportunity cost of time, served as a deterrent and resulted to less time devoted to community health activities [25,43]. As a result, not only are there difficulties on the part of those receiving trachoma interventions as a result of economic context, but there are also difficulties in terms of members of the community advocating trachoma intervention as well. NGOs should consider providing transparent and fair compensation for the time provided by community volunteers [25]. It is understandable, however, that from a practical perspective, NGOs may not have sufficient funds for this and therefore would require additional input from the government, alongside a strong political will to carry out significant changes in the current system.

The economic context faced by NGOs is also relevant in implementing their programmes. One study in Australia described an increase in prevalence in trachoma when funding, and therefore trachoma intervention, was delayed [44]. A systematic review determined that, in low- and middle-income countries, financial limitation was the main barrier to exchange of health information, which is relevant to the F and E components of the SAFE strategy [45]. This study therefore builds upon the concept of the lack of funding, as multiple participants have described a disparity between the funding towards F and E programmes compared to S and A. For Tanzania, this disparity may not be because of the lack of available funds per se, but because of the poor perception of the effectiveness of F and E programmes by donors, which drives financial constraints. Therefore, changing the attitudes of donors regarding the SAFE strategy may be an essential part in the elimination of trachoma in marginalised communities.

NGOs described considerations of overall programme quality and needs of the specific community when making programmatic decisions. On an organisational level, both quality checks and research inform decisions within the organisation by providing feedback on the programme and guides supervision. The benefit of receiving and incorporating feedback into healthcare programmes has also been described in other studies on healthcare improvement in lower income countries [39–41]. Increased supervision of field teams can be used as a form of quality check by NGOs that should be done regularly but randomly, so as to ascertain a more accurate picture of the effectiveness of interventions. For MDA use in NTDs, several studies have highlighted a discrepancy between actual drug intake and the amount of drug provided [46–48]. Often, it has been found to be a result of poor compliance and a lack of understanding of the treatment, which are consistent with the barriers depicted by NGOs in this study. However, none of the NGOs discussed the possible inaccuracies associated with the reporting of MDA figures as a reason for the persistence of the disease. The current method of reporting MDA coverage is based on how many pills are given out, which fails to consider whether MDA is actually received or consumed by the target population [25,46–48]. As a result, the coverage figures may not mirror reality. There is no doubt that MDA is useful in the elimination of trachoma, but in order to fully eliminate the disease, particularly within hard to reach communities in Tanzania, reporting and feedback must be accurate to inform and improve the programme. Hence, there is a need for sufficient quality checks in the implementation of programmes as well as an increased stringency on the methods used to report MDA coverage.

In order to improve programme quality, by both improved intervention provision, improved communication or assessing inefficiencies, NGOs can consider incorporating the use of technology in interventions, which has been explored in other Sub-Saharan countries. Despite the fact that the Maasai live traditionally, there has been a recent increase in the use of modern technology such as mobile phones in communities [49]. Many of the Maasai communities are already using technology for business as well as getting in touch with members of other bomas [50].

In Chad, mobile technology has been used as a low-cost method in order to better understand the behaviour of pastoralists, including migration patterns [51]. Surveys were also being done with the use of technology and were found to be feasible [51]. Considering that NGOs describe that the remote nature and the mobile lifestyle adopted by the Maasai were barriers to trachoma intervention, it may be worthwhile exploring whether this technology in the future could be used to facilitate locating groups in remote areas. It may also provide opportunities for community members to provide additional feedback to improve programmes. There have already been instances where successful mobile health interventions in Ghana and Nigeria have been developed with the backing of the national government, though there is no doubt that a strong political drive has played an important role in rolling out these interventions [52]. However, as the current literature on incorporating such technologies in a resource-limited setting with pastoralist communities are limited, further research on understanding facilitators and barriers of using newer technology, as well as studies to determine the most effective type of mobile intervention would provide valuable information for the development of a reformed programme.

The presence of multiple actors implementing trachoma interventions was a facilitator in reaching the goal of trachoma elimination. The Global Trachoma Mapping Project described that collaboration between different partners and the government is essential in order to carry out trachoma elimination [53]. The findings of this study emphasised the fact that the current system between partners in Tanzania appears to work well and is supported by epidemiological studies demonstrating a significant drop in trachoma prevalence [53]. However, collaboration between stakeholders, either in the field or through joint training programmes, should be strengthened even more to enhance the effectiveness of interventions in marginalised communities and to reduce intervention variability between districts. Quality checks between organisations should be standardised to better measure and improve any variability.

Although the sample size of the study was relatively small, there is only a limited number of NGOs implementing trachoma intervention programs in Tanzania. As a result, the majority of the key stakeholders were included in the study and data saturation was reached. This in turn suggests that the data is somewhat generalisable to the implementation of trachoma programmes in Tanzania.

One of the limitations of this study was the risk of social acceptability bias. The nature of the questions asked in the study may have resulted in participants focusing more on the issues that were external to their organizations, rather than factors within their own organizations that were barriers to programme implementation. As a result, much of the context provided by these organizations may not be a wholly accurate representation of the beliefs and behaviours of Maasai or other marginalised communities in Tanzania. Despite this, understanding the perspectives of these organisations is still valuable to improving trachoma interventions.

Conclusion

This is the first study looking at the perspectives of NGOs on implementing trachoma programmes in marginalised communities in Tanzania and specifically Maasai communities. Overall, difficulties experienced by partners in providing health education and sustaining behavioural change has limited the success in reducing the prevalence of trachoma. These changes need to be finely tailored to the needs and culture of the target community to ensure that interventions are acceptable and sustainable. Without significant improvements in community education, in counselling, understanding of the disease as well as basic hygiene measures, programmes may continue to face poor uptake. Organisations should also ensure that their programmes have ways to incentivise high-quality work as well as have appropriate procedures to monitor the programme quality and ensure the accountability of staff. Incorporating the use of mobile technology in the future may have some potential to improve programme quality and reduce some of the barriers faced by NGOs, though there is a need for further research to determine the feasibility and logistics. If there is a strong political will from both the district and national government to make these changes, eliminating trachoma in these marginalised communities may be realised sooner rather than later. Future research should focus on improving understanding of perceptions of marginalized communities and their behaviours in relation to trachoma, as it would provide insight on how NGOs can better tailor their future programmes to suit these communities.

Supporting information

Acknowledgments

We would like to thank Dr Victor Hu (Clinical Assistant Professor at the International Centre for Eye Health at the London School of Hygiene and Tropical Medicine) for his assistance in the realisation of this project.