Background
The Sustainable Development Goals (SDGs) encourage states not to ‘leave behind’ populations who have been forcibly displaced by war and other extreme hardships in development work. Such forced migrants include people who are internally displaced to areas within their own country where different languages, ethnic groups and customs may predominate, as well as refugees who have left their country and are seeking protection from another.
Finding effective ways to include forcibly-displaced populations in global elimination programmes is important as conflict-affected regions are often the places where disease is most intractable [
1‐
4]. It is particularly important for elimination of sleeping sickness (a fatal parasitic infection also known as human African trypanosomiasis or HAT) as outbreaks in the past have been associated with forced migrations [
5‐
8]. Populations migrating to avoid conflict or returning after displacement are particularly vulnerable to sleeping sickness through: exposure to tsetse flies, which carry the disease, when settling rural uninhabitated areas; famine and stress which may make infected carriers more likely to develop disease and transmit infection; as well as difficulty accessing health services to detect and treat the disease [
5,
9,
10]. Syndromic-based detection of sleeping sickness during routine care visits, which requires health staff to recognise symptoms variably affecting the mind and several body systems and producing different meanings in biomedical and customary health systems, may be particularly difficult in a cross-cultural context [
5,
11]. Humanitarian agencies such as Médecins Sans Frontières (MSF) who have been key providers of sleeping sickness services for conflict-affected populations in the past are disengaging from control as disease prevalence declines. It thus increasingly falls to national programmes and partnerships to secure displaced peoples’ inclusion in elimination activities. Here, we report on governance challenges experienced by Uganda’s sleeping sickness elimination programme to include South Sudanese refugees in facility-based medical surveillance.
Two recent promising, but relatively unstudied, global policy trends provide favourable conditions for ensuring refugees’ access to sleeping sickness surveillance in national elimination programmes: the development of rapid diagnostic tests (RDTs) for use in frontline facilities and the adoption of refugee policies which integrate health services for refugees in national systems.
With sleeping sickness infection recognised as both an outcome and driver of poverty, programme outcomes for this and other ‘neglected tropical diseases’ (NTDs) have been proposed as tracer indicators for a number of other SDG targets to monitor social equity [
12]. Similarly, global NTD plans stress the need to make disease control services universally accessible at the primary healthcare level [
13]. For sleeping sickness, this has become more realistic within the last few years with the development of RDTs which, unlike previous diagnostics, do not require electricity, refrigeration or specialised technical expertise to administer, although further parasitological confirmation is still required before treatment can be given. Because RDTs can be integrated into the routine activities of primary healthcare facilities, the need for external actors to support expensive, independent mobile teams who screen at-risk populations systematically is theoretically less important. This new technology therefore enables a shift in sleeping sickness control governance away from a largely vertical approach, often with multiple actors working in parallel on short-term objectives, towards a long-term, coordinated approach appropriate for elimination that is integrated into public health systems and strengthens them [
5,
14]. The use of RDTs to secure access to sleeping sickness services for forced migrants, however, may involve additional social and governance-related considerations given that forced migrants typically face numerous constraints to health, including the embodiment of social stresses related to their experience of exclusion or marginalisation [
15‐
17]. Refugee health and agency is particularly influenced by the policies of humanitarian and receiving country government systems.
The idea that refugees should be integrated into national development projects has been proposed as a policy solution to the negative effects of social marginalisation on refugee health since the 1980s [
18]. Governmental, humanitarian and development actors should arguably share responsibility for displacement because promoting long-term refugee well-being and independence from aid is also good for the host community. Host communities typically face the same regional development challenges as refugees, such as inadequate healthcare. Addressing refugee needs sustainably can therefore benefit everyone “like a rising tide lifts all boats” [
19]. While ‘interim integration’ of some services such as for healthcare is increasingly popular [
20], most African states typically oppose comprehensive social integration which uniformly grants refugees the same rights as host nationals, including to claim citizenship or permanent residence [
18]. Within the field of public health, there has been remarkably little reflection on the implications of different refugee healthcare governance models on long term goals such as disease elimination [
20].
The current humanitarian crisis in South Sudan has caused the displacement of more than 3.7 million people, including 1 million refugees to Uganda since December 2013 [
21], particularly to the north-western West Nile Region where the Ugandan government is operating a programme to eliminate sleeping sickness. This area was the first in Africa to integrate sleeping sickness RDTs into primary healthcare facilities on a large scale. Uganda has also pioneered a refugee policy which favours integration of primary healthcare services for refugee and host populations [
22], making the region well-placed to incorporate refugees into sleeping sickness surveillance activities. Nevertheless, one year into the response, an incongruous situation emerged in which sleeping sickness RDTs, a key surveillance tool and indicator of access to sleeping sickness care, had been removed from facilities serving high concentrations of refugees who were believed to be at risk for the disease. This exacerbated an already existing gap in equitable access to elimination initiatives between host and refugee populations in West Nile that persisted for at least 3 years. Through close examination of the politics and experiences of refugees and implementers, this study investigated issues with these tandem processes of integration of technologies and people into government systems to explain this inequitable outcome and understand key governance challenges sleeping sickness programmes may face to achieve SDG equity goals among populations of forced migrants.
Discussion
This study has demonstrated an important limitation to Uganda’s integrated refugee policy by observing the national sleeping sickness programme’s response to an influx of South Sudanese refugees from 2013 to 2016. We observed several entrenched norms and practices that worked against integration of refugees into the national sleeping sickness medical surveillance system, despite the availability of a promising technological innovation, an RDT, that could be deployed in the government-controlled spaces where refugees were being provided care.
Before the refugee influx, Adjumani District was assumed to have disease prevalence so low that there would be little need for surveillance. These assumptions were not contradicted by RDT-based monitoring data produced during the first six months of the programme, so the programme reduced surveillance intensity by removing RDTs from most facilities in the district. This had harmful, if unintended, consequences for surveillance equity in West Nile, as Adjumani was the place where most refugees were sent and sleeping sickness experts both within and outside the country believe refugees are at particular risk of disease.
This story has important implications for global sleeping sickness programmes seeking to uphold commitments to addressing disease in vulnerable populations as well as the legitimacy of their claims about elimination. Finding no cases from areas or populations that have been largely excluded from surveillance appears as a programmatic success but it may alternatively be explained as a failure of implementation. Ethnographic study of other global health programmes in Uganda [
45] and elsewhere [
46] suggest important incentives for coordinators not to “look beneath the surface” of successful outcomes data “to see how stated practice relate[s] to actual behaviour” because of the need to prove normative progress towards global goals or to justify subsequent rounds of funding [
45]. Indeed, the need to prove that elimination was happening is a likely explanation for so many refugees’ social exclusion from surveillance in West Nile.
Poor quality implementation was anticipated here because of historical expectations ISSEP coordinators had about the quality of their relationships to Adjumani and some other districts. Additionally, district and national supervisors felt reluctant to engage with refugee-specific issues such as communicating with government health staff recruited for the humanitarian surge response to familiarise them with RDTs, their expected role in the elimination programme, or address their diagnosis-related communication challenges with patients. Coordinators also, however, wanted to demonstrate to donors that they were spending resources efficiently. So, in practice, staff not using RDTs at sufficient levels to justify the cost of monitoring was at least as important a reason for withdrawing RDTs from facilities as was coordinators’ perception of low sleeping sickness prevalence in the district. The absence of international guidance on the quality of surveillance required to produce evidence of elimination before restricting resources in a sentinel surveillance strategy also presumably contributed.
Fully integrating refugees into vertical health programmes coordinated at the national level seems to be a common problem in Uganda. This may not have affected sleeping sickness control until recently, however, because medical humanitarian agencies have historically been such central actors involved in sleeping sickness interventions in this region [
14,
24]. Concerned with both the high mortality of this disease during epidemics and the affected populations who were displaced by conflict, there was substantial overlap in these agencies’ refugee health and sleeping sickness mandates and their responses typically occurred parallel to government structures. As disease has receded, however, humanitarians have disengaged with sleeping sickness control globally and endemic country governments keen to ‘accelerate’ progress towards elimination are now firmly in the driving seat in elimination programmes. It is perhaps not surprising then, that the ISSEP, which channels international funds through the Ministry of Health, has had difficulty integrating refugees. Refugees have not been the Ministry’s responsibility by long-standing tradition in sleeping sickness [
14].
Forced displacement, however, is a growing problem globally. Notably, all 36 countries at risk of sleeping sickness host forcibly displaced populations including refugees, internally displaced persons or recently returned displaced people, with nearly half (17 or 47.2%) supporting large displaced populations of at least 50,000 people, many of whom could be living in areas which support transmission (Additional file
2). New norms, incentives or structures thus urgently need to be established to ensure the needs of displaced people are not left behind by government sleeping sickness programmes in their enthusiasm to demonstrate elimination progress. The Global Fund to Fight AIDS, Tuberculosis and Malaria has successfully encouraged integration in some national malaria elimination programmes by prompting countries to include additional provisions for refugees in their applications [
4]. The SDGs commit countries to monitoring progress towards all targets according to characteristics of vulnerable populations, including migratory status [
47]. Elimination programmes can also be evaluated on their inclusiveness towards forced migrant populations such as whether health outcomes for both host and migrant populations are improved, something which was beyond the scope of our investigations [
20].
This study also highlighted important limitations of RDTs to produce quality and relevant data for elimination. Despite the technology’s appealing simplicity, RDTs are always controlled by human decision-making and behaviour. This is especially important for sleeping sickness programmes like the ISSEP which have abandoned more systematic approaches to case detection, and rely instead on social interactions between patients and providers to identify syndromic suspects for testing. Refugees in West Nile had to overcome substantial communication challenges, sometimes including discrimination, to leave a health consultation feeling satisfied. In other settings in Africa, people who have difficulty communicating their health problems to staff because of differences in literacy, class or ethnicity typically come away with fewer medicines [
48]. Crossing an international border also sharply affects power relations even when refugees and the host population belong to the same ethnic group [
49]. We should assume that it is more difficult for a national health worker to suspect that a refugee than a citizen patient is affected by sleeping sickness through conversations about symptoms and alternative diagnoses, given how exasperated with service delivery both parties felt. While health providers may not like refugees standing over them, demanding to know what tests and medicines they are prescribing, the sleeping sickness literature consistently concludes that patients need to be persistent seekers of healthcare in order to receive a correct diagnosis [
17,
50]. Indeed, patient-led detection is a key reason why internally displaced people have been detected so successfully elsewhere [
17]. Persistence in healthcare should thus not be dismissed as bad behaviour in the refugee context.
Recommendations
Despite their structural and historical basis, the problems highlighted above can be overcome in West Nile and avoided elsewhere. Donor conditionality that incentivises programmes to anticipate refugee needs and disaggregate reporting on vulnerable groups could improve programmes’ relationships with facilities that serve refugees. National trypanosomiasis coordinating bodies (such as COCTU in Uganda) can also promote the integration of migrant populations in national NTD programme policies, budgets and plans, per states’ commitments to the SDGs. How best to overcome systemic integration issues such as communication between government and humanitarian structures should be considered, to ensure that diagnostics are available and used in government and private facilities serving refugees who are at risk of disease. These spaces are key for the successful implementation of both elimination programmes and refugee integration policies. At RDT trainings for health staff, persistence in health-seeking as positive patient behaviour in sleeping sickness could be discussed and translators and community liaisons could also be invited to increase staff and patients’ awareness of available sleeping sickness diagnostics. International guidance on how long and at what intensity surveillance diagnostics should be kept in place is an outstanding issue [
26]. In the interim, supervisors and coordinators should be encouraged to investigate operational reasons for very low use of RDTs in individual facilities.
Acknowledgements
We would like to thank Uganda’s national sleeping sickness programme, the Foundation for Innovative New Diagnostics (FIND), the Office of the Prime Minister and the United Nations Refugee Agency (UNHCR) for providing us permission and opportunities for our observations. For logistics support we would particularly like to thank FIND’s Kampala office, the Danish Refugee Council and Lutheran World Federation field offices. Busitema University provided staff time and the following research assistants provided translation support: Mania Josephine, Apiku Richard Felix, Amour Elizabeth, Mathieng John, Iranya Dominic Eruaga and Anzoo Claire. Finally, thank-you to all the coordinators and supervisors of the ISSEP, Coordinating Office for Control Of Trypanosomiasis in Uganda (COCTU), Ministry of Health staff and displaced people who contributed time and data to provide us with a complete picture of the challenges they face delivering and accessing health care in this setting.