Impact of Intervention Practices on Malaria Treatment Outcomes Among Patients in Bushenyi District, Uganda


 BackgroundMalaria remains a major vector borne disease causing both mortalities and morbidities in the world. Uganda as a country has currently scaled out major campaigns to reduce and eliminate the disease using different interventions. However, there is no clear data on the impact of such interventions on malaria treatment outcomes. Therefore, this study was aimed at assessing the impact of malaria intervention practices on Artemether- Lumefantrine (AL) treatment outcomes among the residents of Bushenyi district, Uganda, a high intensity malaria transmission area.MethodsThis was a descriptive cross-sectional study carried out among 184 study participants for a period of one year (August 2017 to August 2018) in four selected health centers in Bushenyi district, Uganda. The investigative methods used included a researcher administered questionnaire, laboratory and clinical evaluations of participants. Data analysis was done by using statistical package for social sciences (SPSS version 10 windows) for descriptive statistics. ResultsStatistically significant factors for treatment outcome at p ≤0.05 were; practicing indoor residual spraying (IRS) at home (𝑃 = 0.001; CI), source of prescription (𝑃 = 0.018; CI), finishing dosage (𝑃 = 0.006; CI), frequency of malaria infection (𝑃 = 0.028; CI), Frequency of antimalaria usage (𝑃 = 0.042; CI) and sleeping under insecticide treated nets (ITNs) (𝑃 = 0.039; CI) respectively. ConclusionsIRS and ITNs were found to be the major intervention practice of malaria reduction after treatment with ACTs.


Abstract Background
Malaria remains a major vector borne disease causing both mortalities and morbidities in the world. Uganda as a country has currently scaled out major campaigns to reduce and eliminate the disease using different interventions. However, there is no clear data on the impact of such interventions on malaria treatment outcomes. Therefore, this study was aimed at assessing the impact of malaria intervention practices on Artemether-Lumefantrine (AL) treatment outcomes among the residents of Bushenyi district, Uganda, a high intensity malaria transmission area.

Methods
This was a descriptive cross-sectional study carried out among 184 study participants for a period of one year (August 2017 to August 2018) in four selected health centers in Bushenyi district, Uganda. The investigative methods used included a researcher administered questionnaire, laboratory and clinical evaluations of participants. Data analysis was done by using statistical package for social sciences (SPSS version 10 windows) for descriptive statistics. Conclusions IRS and ITNs were found to be the major intervention practice of malaria reduction after treatment with ACTs.

Background
Malaria infection remains the most prevalent vector borne disease [1] and is transmitted by the female anopheles mosquito. The major species involved in transmission in the sub Saharan Africa include Anopheles gambiae (complex), Anopheles funestus and Anopheles maculipalpis [2]. Malaria occurs in over 90 countries worldwide with the majority of cases reported in Sub Saharan Africa. Each year an estimated 300 to 500 million clinical cases of malaria occur, making it one of the most common infectious diseases worldwide [3]. Over 90 % of the Ugandan populations are at risk of the infection [4]. Malaria management programs have positively reduced malaria. It has been recently documented that since 2000, 37 countries in continents other than Africa have experienced at least a 75% reduction in newly reported malaria cases [5].
Malarial intervention methods used to reduce malaria burden include: appropriate diagnostic testing, Insecticide treated nets (ITNs), insecticide residual spray (IRS), intermittent preventive therapy (IPT) in pregnant women and infants. In 2015, World Health Organization (WHO) recommended universal coverage for ITNs use by all persons in all households [6]. WHO also recommends the provision for adequate laboratory testing of all suspected malaria episode. In relation to the Roll Back Malaria Initiative developed from the United National Millennium Development Goals, extensive interventions has greatly reduced malaria; by 40% in sub-Saharan Africa, However, these gains are being compromised by the evolution and spread of Plasmodium [7]. According to the Uganda Demographic Health Survey in 2018/2019, 83% of the total population was in possession of at least one ITNs per a house hold. However, it was noted that ownership was directly proportional to the wealth index. West Nile region had the highest possession of ITNs at 92%, with Karamoja region having the lowest ownership of the ITNs at 58%. Ankole region where the current study area is located had the ownership of ITNs at 86%. [8] In 2012, 88 countries in Africa were implementing IRS, and the number of people at risk protected by IRS increased from 5% in 2005 to 11% in 2010. However, it has been shown that resistance to insecticides has emerged as a problem in Africa [9]. Intermittent Preventive Treatment in Pregnancy (IPTp) is one of the methods that have been employed to control malaria in pregnant women; however, its coverage is still limited. Pregnant women presents one of the major risk group for malaria infection and malaria in pregnant women may not always manifest as febrile illness, but it leads to severe anemia, low birth weight and other fetal abnormalities [10]. It was documented in 2019 that in 2018 worldwide, only 60% of pregnant women received one dose of IPTp, while 49% received two doses and 31% received three doses [11]; generally, coverage of the required three doses of IPTp was inadequate. For instance, a previous retrospective study conducted in 2019 in Uganda revealed that uptake of three or more doses of IPTp was 18%. This was a follow up analysis of data from the Uganda demographic and Health survey, 2016 [12]. In contrast, a recent study conducted in Uganda, reported that 41% of mothers who had given birth two years before the survey received three or more doses of IPTp during their immediate previous pregnancy [13].Therefore, malaria intervention methods have been impressive in the world though it remains unclear what overall reduction in transmission is obtainable by using currently available interventions. Thus, the current study was aimed at assessing the impact of malaria intervention practices on malaria treatment outcomes after treatment with the currently recommended drugs of ACTs in Bushenyi district, Uganda, a high intensity malaria transmission area.

Study area
This study was carried out in Bushenyi District, Uganda ( Figure 1). The district has eleven (11) Sub-counties; 76 parishes and 529 villages. The district is located approximately 380 kilometers from the capital city, Kampala. It lies on latitude: 0° 29' 27.6" (0.491°) South, longitude: 30° 10' 58.8" (30.183°) East. The district has a total land of 3,949 square kilometers and it lies between 910 and 2,500 meters above the sea level. It is hilly with forest and swampy vegetation and is characterized by seasonal water bodies, net-works of streams, stagnant pools, over lled blocked gutters and drainages. The region experiences three wet seasons (April-May, August-September, and November-December) and one dry season (January and February); the temperatures are about 25°C-37°C no wonder the major economic activity of this community is agriculture. The communities reside mainly in semi-permanent and mud-thatched houses. It has over grown bushes and elds of bananas around residential homes which favors the survival and multiplication of mosquitoes. The land under cultivation consists of 2,215 square kilometers and wetlands covers 183 square kilometers.
This study was conducted at the four selected health centers, namely; Kampala International University teaching hospital, Ishaka Adventist hospital, Kyamuhunga health centre and Kyeizooba health Centre. The study area has an entomological inoculation rate (EIR) of 100 per annum [14] and Anopheles gambiae and Anopheles funestus are the main malaria vectors in the region. The control methods which have been scaled up by the government in this region to prevent malaria infection include proper case management with antimalaria drugs such as ACTs, intermittent prophylaxis during pregnancy (IPTp), use of mosquito nets and indoor residual spraying (IRS) [15].

Study Design and study population
This was a descriptive cross-sectional study carried out to assess the in uence of malaria intervention practices on in-vivo Artemether-Lumefantrine (AL) treatment outcomes in Bushenyi district, Uganda. The investigative methods included questionnaire, laboratory and clinical evaluations. The study participants were recruited from the health facility regardless of the age speci cations.

Inclusion criteria
All participants were out-patients who were presenting with signs of uncomplicated malaria to the clinician in-charge; those who had resided in Bushenyi district for at least the last six months prior to the commencement of the study, those who con rmed that they were able to comply with the stipulated followup visits, and those without severe malnutrition as per the 2015 WHO guidelines.

Exclusion criteria
Those who declined to consent to the study, nomadic patients, intolerant to oral treatment according to modi ed WHO 2010 criteria, hypersensitivity or allergy to ACTs, Febrile conditions due to other malaria, and those under medications which may have interfered with pharmacokinetics of anti-malarial drugs and those with severe malaria as per the WHO guidelines were excluded.

Study setting
All adult patients signed the informed consent form and parents or guardians signed the informed consent on behalf of the patients below 18 years. Out of the 283 participants recruited to this study, 194 completed the follow up schedules while 89 were withdrawn from the study because they did not adhere to the follow up visits in the rst three days.

Questionnaires
The study was conducted by using pre-tested administered questionnaires which were used to determine the demographic patterns and effects of malaria intervention programs used by the participants to prevent malaria infection. The questionnaires were developed from the demographic health survey (DHS). This was adapted to re ect the population and health issues relevant to malaria infection in Uganda [17]. The questionnaire was pretested to verify for clarity and administerbility. Consequently, questionnaires were translated to the two most common local languages of Luganda and Runyankole. Information on the following independent variables was collected; socio-demographic parameters (age, sex, number of members in each household, educational level of patient or household head, protective facilities from mosquito bite, symptoms of malaria experienced, frequency of taking anti-malarial drugs, misuse of ant malarial agents, prescription of drugs, hospital visit for fever management, pattern of febrile illness, dwelling unit long-lasting insecticidal nets (LLIN) use, indoor residual spraying (IRS) in the household and intermittent preventive treatment of pregnant women (IPTp) intake were determined [18]. Every participant was allocated a unique recognition code at the time of the questionnaire administration. The names of respondents and households were kept as total con dential information and were not to be used in the presentation of results or available to anybody.

Determination of malaria treatment outcomes
The Plasmodium falciparum positive participants were treated with Artemether-Lumefantrine (AL) drug and monitored at day 0, 1, 2 3,7,14 and day 28. The e cacy of the Artemether-Lumefantrine (AL) treatment was assessed as previously described [19] by evaluating clinical and parasitological outcomes as per the WHO in-vivo clinical and parasitological classi cation criteria for areas of intense malaria transmission [20] which classi es the response as adequate clinical and parasitological response (ACPR) if there is no treatment failure, late parasitological failure (LPF) if P. falciparum parasitemia occurs between 4 and 28 days without fever, late clinical failure (LCF) if P. falciparum parasitemia occurs between 4 and 28 days with fever and nally as early treatment failure (ETF) if there is development of severe symptoms, or insu cient parasitological response by day three.

Data analysis
Data was entered in excel spread sheet 2007, by considering different parameters against the study participants. Data retrieved was analyzed using statistical package for social sciences (SPSS version 10 windows). Descriptive statistics was done using tables showing frequencies and percentage distributions.
Factors that were statistically signi cant (p ≤0.05) at bivariate analysis were included in the multivariate analysis. The model was checked for best t of data and then used to compute adjusted odds ratios of factors that were associated with malaria treatment outcomes. Statistical signi cance was considered at 95% level of con dence. Interaction and confounding factors were assessed, and values of p ≤ 0.05 were regarded as statistically signi cant relationships.

Results
Baseline characteristics of the participants The study recruited 283 respondents, 194 (68.6%) participants completed the follow up schedules while 89 (31.4%) were withdrawn from the study due to violation of the study protocol. The participants were grouped in relation to their social demographic patterns such as age, gender and weight as presented in Table 1. By considering the age, participants with 0-5 years represented the least number of 3 (1.5%) in relation to those having 18 years and above who were the majority with 115 (59.3% (0%), LCF of 1(33.3%), LPF of 0 (0%) and ETF of 2 (66.7%) respectively for the participants who were treated by using Fansidar ( Table 2).
Majority of the participants (n=100; 51.5%) received the recommendations on the type of drug to use by a medical doctor while 57 (29.4%) received recommendation from the nurse. Self-medication was practiced by less percentage of participants at 37 (19.1%). In relation to treatment responses, those who were recommended by a doctor experienced more ACPR at 32 (32%) while those whose recommendation was from a nurse the ACPR was 14 (24.6%) and second most ETF at 9(24.3%). Moreover, those who practiced self-medication experienced the least ACPR at 7 (18.9%). The study indicated that those who received the drugs as a result of the doctor's recommendation had the highest ETF at 42 (42%) and those who practiced self-medication had the lowest ETF at 9 (24.5%)( Table 2).
Majority of the participants purchased the drugs from a health care facility at 119 (61.3%) followed by chemist (drug-shop) at 39 (20.1%) and pharmacy shop at 35 (18%). Those who obtained the drugs from friends were the least at 1 (0.5%). The study found that those who purchased the drugs from the hospital had the highest percentage of ACPR at (28.6%; n=34) compared with those who purchased from pharmacy at 9 (25.7%) and chemist at 9 (23.1%) ( Table 2).
The study reported that 130 (67%) of the participants were able to nish the recommended dose and those who were unable to nish the dose were 64 (33.0% infections twice a year developed ACPR compared to ACPR of 1 (33.3%) for those who experienced malaria infections once in a month ( Table 2).
The study revealed that the highest percentage (51%, n=99) used anti-malarial drugs as recommended in the health care facility, 4 (2.1%) used anti-malaria drugs once a in a while and 91 (46.9%) of the population used anti-malaria drugs any time they felt feverish. It was found out that ACPR was highest (n= 30, 30.3%) among those who used the drugs as recommended in the health care facility compared to 24.2% (n=22) of those who used anti-malaria drugs anytime they felt feverish and 1 (25%) among those who used the drugs once in a while (Table 2).
Malaria control practices associated with treatment outcomes  (Table  3).

Multivariate regression analysis of predisposing factors associated with treatment outcomes
Multinomial regression model was used in bivariate analysis to obtain crude odds for the strength of association between in uencing factors (malaria treatment practices and malaria control interventions) on malaria treatment outcomes. Factors that were statistically signi cant (p ≤0.05) at bivariate analysis, were included in the multivariate analysis. The model was checked for best t of data and then used to compute adjusted odds ratios of factors that were associated with malaria treatment outcomes. Statistical signi cance was considered at 95% level of con dence (Table 4).

Discussion
Malaria treatment and control practices have been generally implicated in the overall reduction of clinical episodes across the world though the exact reduction rate is unknown. The Ugandan government has currently scaled out different intervention programs to combat malaria infection. Despite the efforts being embraced, there still remains huge percentage of mortalities and morbidities caused by the disease. Many studies have been carried across the world and Uganda to determine the usage of different malaria intervention practices on the occurrence of malaria [21; 22; 23; 24]. This study assessed the impact of malaria intervention practices on malaria treatment outcomes on patients treated with ACTs, the rst study of its kind to be conducted in Bushenyi District in Western Uganda. The development and spread of resistance to previously used anti-malarial drugs (sulfadoxine-pyrimethamine,and chloroquine) has prompted Uganda to adopt a new malaria control policy based on three major interventions: vector control with long lasting insecticide-treated nets (LLINs), the use of artemisinin-based combination therapy (ACT) for treating uncomplicated malaria; and chemoprophylaxis for pregnant women [25].
Among all control measures that are used against the malaria spread, chemotherapeutic agents seem to contribute largely to the strategy of combating malaria mortality in the world. Quinoline derivatives and antifolates were the rst mainstream malaria agents in the world; however, resistance has been reported in some parts of the world [26; 27; 28]. Currently they are recommended for use as malaria prophylactic agents in infants and pregnant mothers [29]. Presently the World Health Organization (WHO) recommends artemisinin-based combination therapies (ACTs) for the treatment of uncomplicated malaria in most countries particularly where resistance has developed towards other drugs. Uganda adopted the use of artemether-lumefantrine (AL) as the drug of choice for treatment of uncomplicated malaria, however the actual implementation started in 2006 [30]. The current study reported the highest participants adopting ACTs for malaria treatment at 84.5% which is in agreement with other previous studies done in Africa [31;32;33]. This could be attributed to the fact that ACTs are provided free of charge by the governments in the study sites. The high prevalence for ACTs usage might have been because it is the rst line therapy for uncomplicated malaria in Uganda. Moreover, it is mostly available in public health facilities. The low level use of fansidar observed in this study could be due to the fact that it is recommended by Ministry of Health for prophylaxis majorly to pregnant women and at the same time it is not available in most health facilities.
The present study recorded the highest ACPR (30.5%) among participants who used ACTs. This is not surprising since ACTs are currently recommended by WHO and adopted by the Ugandan government as the rst line treatment regimen for uncomplicated malaria. This nding is in agreement with a previous study conducted in Gulu district of Northern Uganda [34] which recorded ACPR of 95.2% with all participants being negative for parasites at day 2 of the follow up after treatment with ACTs. The study conducted in Tororo district of Eastern Uganda reported an ACPR prevalence of 40% among the participants [35]. Another study conducted at Tororo district recorded an ACPR prevalence of 50% [36] while similar study conducted at Gulu district of Northern Uganda recorded an ACPR prevalence of 42% among the participants [37]. Our study recorded high ACPR among participants who obtained the anti-malaria drugs from the hospital compared with those obtaining from the pharmacy or chemist. This could be because health practitioners in the hospital settings prescribe antimalarial drugs in relation to laboratory results. Furthermore, patients can purchase antimalarial drugs directly from the pharmacies and chemists without any medical prescription. The rate of self-medication practice was found to be 19.1 % which was low compared to other studies. For example a study conducted in northern Uganda, a geographical area of high malaria transmission intensity reported slightly at a higher prevalence of 36.2% self-medications among the study population [38]. Moreover, it has been previously reported in a study done in Apac, northern Uganda that 75.7% of adult (≥18 years) community members accessed and used antimicrobial drugs generally without taking a laboratory test and nor a prescription from health care providers [39]. The probable reason for the reduced selfmedication might be because of the current scaling of free malaria treatment offered by the government of Uganda and the availability of many health facilities across the district. The low self-medications reported in the current study area might also be due to the fact that majority of the population have access to health facilities which are located not more than ten kilometers. Another study conducted in Nigeria reported a high percentage of 33.7% for self-medication compared to 17.4% for doctor recommendations and 12.5% for nurse recommendation [40]. Anti-malaria drug use pattern in uences the development and spread of resistance in the community due to continued sub-optimal use of the drugs. Self-medication uses treatment without any diagnosis and this has been implicated in the anti-malaria resistance [41; 42]. Appropriate use of the correct prescribed dose is an important factor in malaria treatment. From the current study, 31.2% of the participants did not complete their prescribed doses while 67% completed the prescribed anti-malaria doses in the recent malaria illness. These ndings are contrary to those from other studies such as one that was carried out in Bangladesh which reported that 68% of the participants did not complete the treatment doses because patients would stop taking medications as soon as they would feel well [43]. Failure to complete treatment doses leads to sub-therapeutic drug levels in patient blood which encourages selection for resistant parasites and treatment failure [44].
The current study recorded a substantially low usage of indoor residual spraying at 42.3%. Contrary to previous ndings by Uganda demographic health survey [45] which found out 67.2% of the population was using IRS. However, it was established that majority of the participants (78.6%) used government to access the IRS. This is because currently the government of Uganda has scaled out nationwide IRS in all households present in the high malaria transmission areas [46] In 2012, 88 countries in Africa were implementing IRS, and the number of people protected by IRS increased from 5% of those at risk in 2005 to 11% in 2010, but fell to 8% in 2012. Our study has shown signi cant reduction by IRS on the occurrence of parasitemia after treatment with anti-malaria drugs. This is in agreement with other studies conducted elsewhere in Uganda [47; 48; 49; 50], as well as in other studies conducted outside Uganda [51; 52].
In 2007, WHO recommended universal coverage for ITNs use by all persons in all households [53]. The present study found out that 83.5% of the participants slept under mosquito nets. This may be due to the fact that the government of Uganda has distributed free nets in previous years as well as the continued malaria campaigns that have been taking place. But surprisingly only 45.1% of those who owned nets were insecticides treated nets (ITNs). These ndings are in contrast to a study which showed that 54% of African households owned at least one ITN and only 36% of the population slept under ITNs (54). According to the Uganda Demographic Health Survey done in 2018/2019, 83% of the total population was in possession of at least an ITN per house hold [55]. West Nile region had the highest possession of ITNs at 92%, with Karamoja region at 58% and Ankole region where the current study area is located at 86%.Our ndings did not show any signi cant difference on the protective effect against malaria between ITNs and noninsecticides treated nets. The results however, showed that sleeping under any type of net reduced the rate of malaria occurrence after treatment with anti-malarial agents. Our ndings are in agreement with a study conducted in Uganda from 2009 to 2014 which showed a strong ITNs effect on parasitaemia risk reduction [56]. Other studies done outside Uganda also reported similar trends on usage of ITNs in the reduction of malaria incidences [57; 58; 59; 60]. In relation to the in uence of ITNs on treatment outcomes, our study recorded an ACPR of 30.2% which disagrees with previous studies conducted outside Uganda that recorded lower ACPR. For example, a study conducted in Tanzania among malaria positive pregnant women recorded a 6% prevalence of ACPR [61]. Other related studies have also reported lower ACPR among the participants at 9.1 % [62]. However, our study is in agreement with some previous studies which recorded a higher prevalence of ACPR of 42% [63; 64].

Conclusion
This study has demonstrated that two major malaria control interventions i.e. IRS and ITNs had a positive impact on reducing of parasitaemia risk after treatment of patients with antimalaria drugs.

Recommendation
We hereby recommend that the government of Uganda with other private sectors need to increase the universal ITN and IRS coverage. Uganda Ministry of Health should promote malaria prevention awareness campaigns so as to educate the public about the importance of ITNs and IRS even after being treated with drugs. ITNs and IRS should be integrated together with ACTs in patients under malaria treatment so as to prevent future malaria occurrences. In addition we recommend that more studies covering larger geographical regions should be conducted in Uganda.

Authors' Contributions
This work was carried out in collaboration between all authors. Authors JNM, ADK, and MCM designed the study, managed literature review, performed the statistical analysis, wrote the protocol, and wrote the rst draft of the manuscript and proof reading of the nal draft. All Authors managed the analyses of the study, proof read the rst draft of the manuscript and performed the statistical analysis. All authors read and approved the nal manuscript.

Competing interest
All authors declare that there are no con icts of interest existing in regard to this study and publication of the ndings.