An exploratory qualitative study was carried out in Kanungu district, south western Uganda, in five Batwa settlements in the district. Qualitative research gathers participants’ experiences, perceptions, and behavior. It answers the hows and whys instead of how many or how much [33] . Five Focus Group Discussions (FGDs) with an average of 5 participants each were organized in five purposively selected Batwa settlements to elicit malaria lived experiences of Batwa people, in particular, the symptoms, prevention, treatment and barriers to accessing formal health care services. On average each FGD took one hour. The criteria for selection of Batwa settlements to conduct FGDs were: 1) high malaria prevalence in previous surveys [5] [34] [35] [36] (i.e., Kebiremu, Kitariro & Byumba), 2) proximity to health facilities (i.e., those very near like Byumba & Kitariro), 3) diverse geography or place factors, especially high altitude (Kitahurira), low altitude (Kebiremu), location near wetlands, forest and river basin (Kihembe & Kitariro).

In addition, 13 in-depth key informant interviews (KIIs) (n = 13) were conducted to bring out the lived experiences of Batwa of key malaria risk factors and enabling factors for access to the formal health system. The KIIs included 7 Batwa resource persons and six non-Batwa health workers. On average a KII took 45 minutes. Responses from Batwa FGDs were triangulated with responses from KIIs and vice-versa for validity.

Kanungu district in Southwestern Uganda is located 29˚50'E and 0˚45'S of the Equator, covers an area of 1297.5 sq. km. and it borders Rukungiri district to the east, the Democratic Republic of Congo (DRC) in the west, Kabale and Kisoro districts on the south. The district is partly covered by impenetrable forests that stretch from Rwanda and DRC inhabited by mountain gorillas. Like much of Uganda, Kanungu typically experiences two rainy seasons—March to May and August to October—with an annual rainfall of 1000 - 1500 mm. The district is largely a rural area of rolling hills located at an average elevation of 1310 m above sea level [37] . The study was conducted in 5 of the 10 Batwa settlements in the district (Figure 1 adopted and modified from [38] ).

The Batwa population in Kanungu District was resettled in 10 settlements. Originally living in the Bwindi Impenetrable Forest as hunter-gatherers, in 1991 the Batwa were removed by government to establish national park to protect mountain gorillas. The communities are small with 25 - 200 people. The total Batwa population Kanungu district ranges between 500 and 700 [39] . The Batwa now engage in small subsistence cultivation, tourism activities and offering paid labour [24] .

The study obtained ethics approvals from Institutional Review Board (IRBs) of Mbarara University of Science and Technology, (ref. 29/04-18) and by the Uganda National Council for Science and Technology (ref. HS2460). Administrative clearance was sought and given from Batwa Development Programme (BDP) and Kanungu District Local Government. Prior to all FGDs and KIIs the principal investigator introduced the research team, and explained the purpose of the study. Verbal consent was then sought, since majority of Batwa do not read or write. Interviews and discussions would only proceed upon receipt of consent from all participants.

The majority of participants in the FGDs were female (n = 19) representing 73.1%, while males were fewer (n = 7) representing 26.9%. The females were aged between 24 and 80 years, while the males were aged between 22 and 50 years (Table 1). Females are more involved in caring for the sick and this would enrich the lived experiences. Regarding KIIs; out of the 13 interviewees 7 were Batwa resource persons in the community and 6 were non-Batwa health workers at health facilities where Batwa access health services. The inclusion of non-Batwa among KIIs in the study aimed at obtaining views from the Batwa as well as non-Batwa. The non-Batwa interviews provided context in terms of the health system and were considered separately, alongside Batwa perspectives [40] [41] [42] .

In terms of education attainment, majority participants in the FGDs had primary and secondary education. Overall, 13 KIIs were conducted i.e., a total of n = 7 KIIs across the 5 Batwa communities (Byumba n = 1, Kitahurira n = 1, Kihembe n = 2, Kitariro n = 2, and Kebiremu n = 1). Also 2 non-Batwa (Bakiga) community members (in Kitariro and Kihembe), and non-Batwa representatives from BCH (n = 2), Byumba HCII (n = 1), and Kanyashogye HCII (n = 1). The community KIs; were community leaders while the representatives BCH, Byumba HCII and Kanyashogye HCII comprised public health officers and nurses. For the KIIs more than 50% (n = 9) had secondary and tertiary education with average age of 34.5 years.

The respondents had divergent views on what causes malaria in the Batwa community. The majority of the respondents correctly associated malaria with mosquitoes based on what they had experienced and known about malaria. However, others believed it to be caused by other factors:

Going in dirty water, as in walking barefoot in water considered dirty: swamps, mud. FGD1 Kitariro.

Being bitten by mosquitoes, poor diet, and coldness. FGD 3 Kitahurira & FGD 4 Byumba

Not sleeping under mosquito nets, living near water bodies, such as swamps. FGD 2 Kihembe.

The majority of participants’ knowledge on the ways of preventing and controlling malaria at the community level concentrated on the use of mosquito nets;

If you sleep under the mosquito net malaria will not catch you. Mosquito nets were distributed to us, but are inadequate in numbers not covering the total number of people in the household leaving others without. FGD 3 Kitahurira.

Mosquito nets were distributed to us; we all have them. I got 4 nets; every child has his/her net up. The rats damaged them all and they are all dysfunctional except the one on my bed. FGD 4 Byumba.

Meanwhile other participants knew environmental malaria control such as slashing, early closure of the house entrances in the evening.

Slash bushes around the home, removal of rubbish, closing windows and doors in the evening can prevent malaria. FGD 2 Kihembe & FGD 3 Kitahurira

Other participants mentioned government programme that was implemented in the region to control the mosquitos and malaria.

Village Health Teams (VHTs) and health workers from Bwindi Community Hospital visit our households to sensitize us about malaria prevention. KII-06.

Yes, we access preventive measures like mosquito nets…but we are still careless on other prevention measures. The lower part of Kebiremu is so badly off, with bushes all over the place. KII-07.

The majority of the participants interviewed had beliefs of using other remedies such as herbs prior to accessing health facilities.

I used herbs to treat the child, but these days we go to the health facility immediately because of closeness to the health centre. FGD1 Kitariro.

I boil water for them to bathe in order to reduce the temperature, the problem is coldness because this area is colder. Then I apply some herbs, if there is no improvement, I go to the health facility. I use herbs such as Ebizinya muriro, applied with warm water. FGD3 Kitahurira.

Some of the participants reiterated that the use of the herbs among Batwa has reduced to some extent compared to when they had just come out of the forest over 20 years ago. It is now common for Batwa people to seek western medicine at the health facilities in the area.

In the past we used herbs a lot, but nowadays we do not use them a lot as we go to the hospitals especially now that even some of the herbal resources, we used to rely on have dwindled. FGD2 Kihembe.

I do self-treatment using herbs, then going to the hospital later. FGD 3 Kitahurira.

For one participant in Kitahurira, the first course of action on suspecting malaria was purchasing drugs from a nearby drug shop.

“I bought some drugs at the shop…after taking those drugs, I did not feel well. I took the drugs in the morning. By evening, I was worse. I spent two days at home after that before eventually going to Kanyanshogye health facility, where I was tested and found to have severe malaria and was given coartem, Panadol, and an injection. I took the drugs as I was directed and I felt better after two days of taking them”. FGD3 Kitahurira.

Other participants also mentioned the role played by VHTs in the treatment of malaria in the community.

VHTs have helped us in quick treatment of malaria. Here one of us Batwa was trained as VHT and she helps us to advise us and to give us some tablets before we can go to the health facility. FGD 2 Kihembe.

Sharing of malaria medicines was also noted by some participants. Participants agreed that this practice is common in the area.

I developed high temperature and headache. I never went to the health facility, instead I took leftover tablets from my mother-in-law, but I still felt unwell. FGD 4 Byumba.

Many of the participants talked about what they knew about malaria disease, the signs and symptoms, their experiences as patients who suffered what they defined as “the malaria disease”. From the FGDs and KIIs, the definition of the signs and symptoms of malaria were based on individual experiences and the local context as illustrated below:

Malaria breaks up the back, whole body, eyes water, legs weaken, causes terrible pain in the head, and you become weak, have no peace. FGD1 Kitariro.

I determined malaria by experiencing shivering, high temperatures and even if they gave me the food, I like most, I would fail to eat it. KII-02.

Malaria brings sourness in the mouth, making the taste of water or food in the mouth sour, leading to vomiting. Another thing, I also get diarrhea, increase in temperature and body weakness. FGD1 Kitariro.

On the one hand others described malaria signs and symptoms that come in form of fever causing goose pimples, shivering, body pain and sometimes high temperatures.

It comes with goose pimples and shivering; I crave to go under the sun or near the fireplace and body joints weaken. Eyes water due to the extreme headache. Then I know its malaria and I organize to go to the health facility for attention. FGD1 Kitariro.

I become weak and fail to work, come down with a high temperature. However, I wait for one or two days, then on the third day of the onset of the above symptoms, that’s when I often go to the health facility. If I go immediately upon the onset of the symptoms, the test will come out negative. FGD1 Kitariro.

On the other hand, some of the participants were able to define malaria as causing change in temperature, headache, backache, diarrhea and body weakness.

I was pregnant around six months when I got a temperature and developed severe headache in the morning. By the time I reached Bwindi Community Hospital, it was 11:00 pm in the night after paying 50,000/ = for boda boda. When I arrived, I was told that it was already too late. The doctors did everything to save me. I was injected many times, on the hands and on the leg and I was put on drip. I sweated so much during which time I was told that my babies were coming out. I had a miscarriage and I lost both babies. They were twins. I spent four days in the hospital after that. The pregnancy that I lost was the fifth pregnancy. FGD 2 Kihembe.

I know I have got malaria when I experience backache and high temperature. In addition, I develop diarrhea, and then I become too weak. FGD2 Kihembe.

Other participants were able to contextualize malaria as causing a queasy feeling in their stomach accompanied by vomiting and passing yellow stool, body pain, body weakness especially the joints, dizziness, change in temperatures and fever.

Malaria gives a queasy feeling in the stomach which may be accompanied by passing lose yellow stool and vomiting, then it goes to the head causing extreme shattering pain, weakness in the joints of the hands and feet, shivering and high temperature. FGD 4 Byumba.

The barriers to accessing treatment at health facilities that most of the respondents shared were long distances, financial and difficult geography.

Distances to health facilities are too far, treacherous across the forest and hilly terrain especially in severe cases that required referral to Bwindi Community Hospital. FGD 1 Kitariro.

Movement from home to the health centre is the biggest challenge due to transport costs. After the onset of malaria, the body is weak and to reach the nearest road to access road transport is also difficult because of hilly and steep terrain. This is difficult for a person already sick from malaria. KII-05.

Some of the participants identified marginalization of the Batwa, favouritism of non-Batwa and poor inaccessible transport network as major barriers to accessing health care services.

We are marginalized by health workers who favor non-Batwa who are given priority over us when at the Bwindi Community Hospital. FGD 5 Kebiremu.

In addition, the drug stock-outs at the health facilities and lack of regular outreach to the community were barriers to accessing health services.

Mobile Clinics stopped, or occur each month and there’s nothing in between. So, if malaria strikes, it is a challenge to access treatment quickly. FGD 3 Kitahurira.

Others expressed how they were not able to move to the distant health facilities where it required them to stop over and buy food.

Lack of funds for food while on the way or at the health facility—the fear of moving away from the place of safety where food can be got to one of biting hunger, that’s on the way to and from hospital and in case of inpatient stay, during hospital stay is a big challenge. FGD5 Kebiremu.

Some health workers at the health facilities where Batwa access services indicated that Batwa have overtime developed a mentality of entitlement.

Batwa people have an attitude of entitlement. They have a mentality that organizations like Batwa Development Programme and the government should provide them free health services. The Batwa should be empowered to establish sustainable livelihoods like agriculture for food production and crafts to earn income. KII-01.

We found that majority of respondents, both Batwa, and non-Batwa, had experienced malaria in their households and believed in using remedies such as herbs prior to accessing malaria treatment from health facilities. This is similar to Indigenous communities in India who knew about malaria from having suffered from it previously, or through discussions with friends, family, or neighbours who had been infected [22] . In Burkina Faso malaria treatment was often reported to be a combination of both scientific and traditional methods. Depending on the type of malaria and its severity, people usually started with some traditional therapy, followed by scientific treatment in case of failure [44] . In Ghana, visiting health facilities after the onset of fever was the second-best option for participants; and it was significantly reported among civil servants and traders, probably because of the financial implications vis-a-vis the relatively higher out-of-pocket expenses compared to registered medicine stores [45] [46] [47] . The choices people make in seeking appropriate care when a family member becomes ill are influenced by the broader social context within which they live. Decisions about where to seek treatment are also influenced by interpretations of severity and cause according to people’s experiences [47] .

We found out that many Batwa are dependent on traditional medicine as first-line treatment when ill. This may be because they are renowned for their knowledge of herbal and other traditional treatments for illness, or they lack money to pay for services and they may also experience discrimination [15] [26] [34] [38] . However, use of traditional remedies and inaction to observe the symptoms has been attributed to delays in proper management of malaria leading to complicated, severe and life-threatening episodes especially in children and pregnant women [45] [48] [49] . Malaria control requires an integrated approach including prevention and prompt treatment with effective antimalaria agents [50] [51] . Artemisinin-based combination therapies (ACTs) are the first line treatment for uncomplicated malaria in Uganda [52] [53] but specific interventions may be needed to reach remote at-risk communities and to ensure that they are used appropriately [50] [54] . Participants in this study reported using ACTs like coartem either from health facilities, VHTs or from drug shops in the community [22] [44] [54] . However, a major concern from this study was the sharing of malaria medicines by participants from their family members. This practice points to poor adherence to malaria medication and may lead to ineffective dosage, developing severe malaria and drug resistance [22] . It is, therefore, very important that health education in rural settings targets the importance of seeking help promptly and from the right place when ill with symptoms of malaria thus creating an enabling environment for malaria management (see Figure 2). Going for a malaria test when there is any febrile illness, to rule out other causes of fever and use medication appropriately is crucial [55] .

Our study also showed the grassroot role played by VHTs in creating awareness about malaria control and its management. VHTs are provided with pre-packed ACTs and can administer simple treatment for malaria [56] . Evidence from India with Community Health Workers—CHWs [23] and Cambodia with Village Malaria Workers—VMWs [54] indicated that provision of free diagnosis and treatment through these community-based structures is an effective means of increasing access and coverage especially for marginalized populations.

Experience of barriers to malaria treatment are an integral part of issues relating to both the social and cultural determinants of health and how they may hamper Indigenous patients’, their families’ and communities’ from accessing care [12] . Our research revealed multiple barriers faced by Batwa Indigenous people in accessing malaria treatment that included: 1) long distances to health facilities, 2) geographically difficult terrain, 3) economic constraints, 4) lack of regular health outreaches and 5) stockouts of malaria medicines at health facilities.

Many participants in the FGDs and KIIs indicated long distances to health facilities as a major impediment to accessing malaria treatment. This finding on long distances is consistent with an earlier study [32] where participants from most Batwa communities described distance to the health centres as a major determinant and prohibitive to antenatal care attendance (ANC).

Furthermore, geographically difficult terrain comprising of steep hills, wide valleys some occupied by wetlands, swamps and rivers were reported by some respondents to hinder easy movements to and from health facilities for treatment. This was more pronounced during the rainy seasons when the unpaved roads become very difficult to use and some roads and village paths are flooded. Similar evidence of geographical barriers to accessing malaria treatment have been reported among Indigenous communities in India [22] and among Guna Indigenous peoples in Panama [23] . In India, Indigenous communities often live in remote, often heavily forested and rural areas while in Panama geographic isolation, limited access to health services, while poverty and illiteracy are also noted as main problems of accessing health care [22] [23] .

Economic constraints comprised another barrier to accessing malaria treatment by Batwa. We found these ranged from inadequate money for transport to and from the health facility, payment for services and food while traveling or when admitted at the health facility. Other studies have noted that costs of transportation and hospital expenses were a barrier [12] [23] [57] [58] . However, where health microinsurance schemes exist covering costs of care at health facilities, hospital expenses were no longer a barrier, leaving transportation as the only concern [57] . Many Ugandans spend substantial out-of-pocket expenses in seeking health care, especially for recurrent diseases like malaria [59] ; thus, the poor and those without adequate finances (like the Batwa) are forced to seek cheaper alternative treatments, such as herbal remedies [60] . Similar to Batwa IPs, in Nigeria households in rural areas suffer a series of deprivations with respect to access to the basic necessities of life including health care [61] [62] [63] . This finding has implications for policy, especially regarding establishing universal health coverage through community or national health insurance schemes to reduce financial vulnerability of households and improve health seeking behaviour—including the use of preventive measures for malaria [64] [65] .

We found out that Batwa were used to tailored mobile clinics and outreaches from BCH to access services for malaria (Figure 2). However, mobile clinics stopped and outreaches were now irregular—only once a month. Yet, community outreach in malaria hotspots with screening and treatment to supplement health facilities drastically reduces cases like it has been reported in Kisoro district in Uganda [52] . We therefore recommend consistent regular outreaches, especially in remote isolated areas where access to healthcare is limited. Alternatively, the government and partners in malaria control should ensure that VHTs have regular supplies to provide the first line of malaria treatment and other common diseases (Figure 2).

Another group of participants mentioned stockouts of malaria medicines as an impediment to treatment. The key causes of medical stockouts in Uganda are poor planning, prioritization and weak tracking system. Non-availability of certain types of supplies (e.g., malaria ACTs and diagnostics) at a particular time may directly translate into loss of lives and undermine the health system [65] . To this end, the government and development partners should ensure proper quantification and consistent supply of malaria medicines and diagnostics at all levels (Figure 2).

Need factors are the most immediate cause of health service use and reflect the perceived health status of the individual [27] . Our study proved that Batwa can clearly identify the symptoms of malaria from the onset of an episode. They ably described the symptoms of malaria based on their experiences of the illness and local context. This is consistent with other studies [43] [44] [50] . We found out that the knowledge of symptoms, and the severity of malaria are enablers for seeking treatment. Contrary to this finding, some Indigenous communities in India lacked knowledge regarding malaria symptoms and transmission [22] . Other published previous studies; found that 42.2% of the study population and up to 18.2% who attended university did not know a symptom of malaria [66] . Yet, knowing and being able to characterize the symptoms of malaria is a very important step since it can induce other actions related to management of the illness. For instance, fever is usually a common malaria symptom that triggers the decision to seek care compared to other symptoms [67] . Ability of the ill person to recognize symptoms and communicate them to the professionals is a crucial first step [68] . Caution should however be taken that the symptoms of malaria are non-specific. Malaria is suspected clinically primarily on the basis of fever or a history of fever. There is no combination of signs and symptoms that reliably distinguishes malaria from other causes of fever (e.g., febrile illness); diagnosis based only on clinical features has very low specificity and results in overtreatment. Therefore, malaria diagnosis should aim to identify patients who truly have malaria (using malaria rapid diagnostic tests mRDTs and microscopy) to guide rational use of antimalaria medicines [50] [55] .

People in different societies hold a variety of beliefs about the cause and transmission of malaria that vary according to cultural, educational, and economic factors, and have direct consequences for both preventive and treatment-seeking behaviour and activities to control malaria [69] . Following on this, we found divergent views on what causes malaria among Batwa, but the understandings were largely consistent with scientific perspectives, therefore highlighting Batwa’s knowledge about malaria. These findings are consistent with other studies that discovered diversity in explanation as to whether it is simply the presence of mosquitoes or their bite specifically that transmits infection among rural farmers in Burkina Faso, Indigenous communities in India and Batwa in Kanungu district respectively [22] [31] [44] . Similarly, regarding malaria prevention, a variety of prevention methods were mentioned by participants. This finding is in conformity with studies where IPs were familiar with scientific approaches to disease prevention [13] [22] . Therefore, this research provides a good baseline for government and stakeholders involved in malaria control especially in rural settings to strengthen health promotion towards malaria prevention and control.

A major concern from this study relates to availability and use of mosquito nets. Nets were inadequate to cover all members of the households. Other responses indicated nets had been damaged and no longer served their purpose. Some studies have shown a significant relationship between ownership of LLINs and malaria prevalence, with malaria rates being lowest shortly after targeted distribution of LLINs [30] . Still, the poorest and most marginalized populations, notably IPs, may still be disadvantaged from ownership and use of LLINs compared to general Uganda population [70] . This suggests that high malaria prevalence among Batwa may be partly a result of poor LLINs coverage. Based on this result we recommend that government and partners should develop affirmative action in malaria interventions for IPs.

Limitations of this study may include the following: Only 5 of the 10 Batwa IPs settlements were purposively selected for this study leaving out another 5 which could have left out some malaria lived experiences from these communities. However, the inclusion criteria based on those experiencing high malaria burden, distance to health facilities, and physical factors like elevation and natural vegetation cover were considered to bring out a rich mix of experiences. It is also noteworthy that Batwa IPs are a relatively homogeneous community, meaning that the 05 settlements are very representative. Nonetheless, further studies covering all settlements could be undertaken in future to capture more insights on malaria. An additional limitation could be potential for social desirability bias [71] whereby positive responses would be related to participants desire to please the researchers and with some expectations. However, all participants were asked to share their experiences truthfully and without bias and expectations.