Urban malaria prevalence and its associated risk factors: the case of Batu town, East Shoa, Oromia Regional State, Ethiopia

Background- Malaria is one of the leading causes of morbidity and mortality particularly in sub-Saharan Africa including Ethiopia. Approximately, 68% of its population is at a risk of malaria infection and nearly three-fourth of the Ethiopia landmass is considered as malarious. Most of the malaria related researches are being conducted in the rural areas; however, urban malaria should be also targeted without which malaria elimination is impossible. Therefore, the aim of this study was to assess the magnitude of malaria, prevalence of P. falciparum and P. vivax and its associated risk factors during low transmission season (April to July, 2018) in Batu town, East Shoa, Oromia Regional state, Ethiopia. Methods – Finger pricked blood samples were collected from 356 febrile individuals visiting health facilities in Batu town from April to July, 2018 supplemented with questionnaire survey to assess socio-demographic and risk factors associated with malaria infection.


Background
Malaria is a life threatening protozoan disease caused by one of the ve Plasmodium parasites: P. falciparum, P. vivax, P. ovale, P. malariae and P. knowlesi and transmitted by the bites of female Anopheles (An.) mosquitoes. P. falciparum is the deadliest parasite in terms of its morbidity and mortality and it is the most prevalent malaria parasite in sub-Saharan Africa (SSA) accounting for 99% malaria cases in 2016, whereas, P. vivax is the most prevalent in Asia and South America [1]. Worldwide an estimated 219 million cases of malaria occurred in 2017 which are slightly higher in trend compared to 216 million cases of infection in 2016 in 91 countries and the global total of malaria deaths reached 435,000 deaths, compared with 451,000 estimated deaths in 2016 [1]. Mainly (90%) malaria deaths occur in SSA, where children under ve years of age account for 78% for all malaria deaths [2].
Ethiopia is one of the sub-Saharan Africa mostly affected by malaria. It ranks top of the communicable infectious diseases in the country [3]. It is one of the leading causes of morbidity and mortality in Ethiopia [4]. Approximately, 68% of its population is at a risk of malaria infection [5] and nearly three-fourth of the Ethiopia landmass is considered as malarious, with malaria mainly associated with rainfall and altitude [6]. Malaria transmission is seasonal and unstable; major transmission occurs from September to December while minor transmission occurs following short rainfall from April to May [7]. In Ethiopia, P. falciparum and P. vivax are the two dominating species accounting for 60% and 40% malaria cases, respectively, reported in the country [7,8]. However, this proportion varies from place to place and season to season [7]. For example, recent studies have showed that there is a trend of shift of dominance from P. falciparum to P. vivax [9][10][11]. Since 2004, arthemisinin-lumefantrine (AL) (Coartem®) and chloroquine (CQ) are used as rst line drugs for the treatment of P. falciparum and P. vivax, respectively, in the country [7]. An. arabiensis, a member of An. gambiae species complex, is the main malaria vector while An. pharoensis, An. funestus and An. nili are secondary vectors in some areas [12,13]. Recently, a new malaria vector, An. stenphensi, was detected in Ethiopia through molecular and morphological methods [14].
Compared to rural areas, urban areas are considered at low risk to malaria because of the improved housing, socioeconomic status, expanded personal protection, effective diagnosis and treatment and limited breeding sites for Anopheles mosquitoes [15,16]. However, reports showed that urban malaria cases accounted for 6-28% estimated global malaria incidence [15], therefore in the era of malaria elimination urban malaria should receive due attention. Urban agricultural practices and small irrigations surrounding towns in African cities might create conducive environment for the Anopheles mosquitoes breeding [11,16]. In addition most of African cities including those of Ethiopia are characterized by poor housing conditions, lack of proper sanitation and poor water drainage system which could likely increase humans and malaria vector contacts and thereby facilitate urban malaria transmission [17]. On the other hand, Ethiopia in its national malaria elimination roadmap set target to eliminate malaria by 2030 [18].
However, malaria researches in Ethiopia mainly focused on rural areas, as a result there have been scarcity of studies addressing magnitude of malaria, species distribution and associated factors in Ethiopian towns located in malarious areas including Batu town. The nature of rural areas is different in many aspects from that of urban areas. Research ndings of the urban areas could help to design and implement appropriate cost-effective intervention strategies to realize malaria elimination set by the country. Therefore, the aim of the current study was to assess the magnitude of malaria, prevalence of P. falciparum and P. vivax and the associated risk factors during low transmission season (April to July, 2018) in Batu town, East Shoa, Oromia Regional State, Ethiopia. Understanding the magnitude of malaria situation and risk factors during minor transmission season enables us to deduce larger picture of malaria during major malaria transmission season. The ndings of this study will augment policy makers, program managers and implementers during the design of strategies for the control and elimination of malaria in Ethiopia.

Description of study area
Batu, formerly Zeway town, is located in the central part of Ethiopia, 165 Km South of Addis Ababa, capital of Ethiopia, in the middle of Ethiopian Rift Valley ( Figure 1) and described elsewhere [11]. Brie y, the town's geographical coordinates are 7 0 56'03" N latitude and 38 0 42'56" E longitude. The town is found at an average altitude of 1657m above sea level. The town has a total population of 78,784 (40,180 males and 38,604 females) in two kebeles (government administrative entities below district level). The town has two hospitals (one government and one private), two government health centers and eight private medium clinics (Batu town health o ce, 2018). In the town there are gorges, ditches, poor sanitations (tidiness) and vegetation covered areas which favour mosquito breeding conditions. Although there are also poorly constructed houses in the town, its periphery is especially characterized by such houses which could allow indoor mosquito biting during the nights. Around the town there is Lake Batu (Zeway), which covers an area of about 434Km 2 and an average depth of 4m [19]. The lake serves for shing, recreation and irrigation of small farms. The lake area maintains malaria transmission even during dry season by creating conducive breeding ground for Anopheles mosquitoes at the lake shoreline [20]. The area receives between 700 -800 mm of annual rainfall, with the heavy rains during the months of June to September and short rains in April and May. The mean temperature of four months was 22 0 C; with a mean maximum and minimum temperatures of 29 0 C and 15 0 C, respectively (South Ethiopia District Meteorological Agency, 2018). Malaria transmission in Batu area is generally unstable (seasonal), with peak transmission occurring between the months of September and November, immediately after the main rainy season, while the second less pronounced transmission period falls between April and May in the short-rainy season. The major malaria vector in the area is An. arabiensis while An. pharoensis plays secondary role [19].

Study design
Health institution based cross-sectional study was conducted among patients from Batu town attending governmental and private health facilities from April to July, 2018. Patients suspected for malaria from the outpatient departments of health facilities were included in the study. Malaria suspected patients who came from out of the town were excluded from the study.
Based on the above assumptions, the sample size was calculated as follows: Accordingly, 356 individuals were included in the study. Study subjects were selected randomly for parasitological examination from all 12 health facilities (two hospitals, two health centers and eight private medium clinics) found in Batu town.

Blood Collection
At each health facilities, patients were rst clinically diagnosed for the presence of fever at outpatient departments. Study subjects with eligible criteria were invited for parasitological examination. Febrile individuals who are older than 18 years and parents of those children less than 18 years were interviewed for socio-demographic and risk factors associated with malaria infections.
Blood collection was done according to the standard operating procedure developed by WHO [23]. Brie y, before blood collection the tip of the nger was cleaned by cotton moistened with alcohol. Using blood disposable lancet two drops of blood was taken on slide. Thick and thin blood smears were prepared on the same slide side by side. After slides were air dried on horizontal position, the thin blood smears were xed with methanol for about 30 sec. The thick blood smears were stained with 10% Giemsa solution for 20 minutes. Blood slides were read under oil emersion objectives by experienced laboratory technologists.
Parasites positivity was detected from thick blood smears while Plasmodium species identi cation was done from thin smears. For quality control blood slides were read by two technicians at each health facilities. At each health facilities laboratory technicians re-read the samples by being blind to the slide test results, no discordant results were obtained. Slides were regarded as negatives after 100 elds were carefully examined.

Ethical consideration
Ethical clearance for this research was obtained from Adama Science and Technology University Research and Ethical Committee. The aims, risks and bene ts of the study were explained for the study participants. Written informed consent was obtained from the study participants and in cases of children under 18 years age assent was also obtained from their parents. Patients showing positive results for P. falciparum and mixed infections were treated with AL while those infected with P. vivax were treated with CQ according to the national malaria treatment guide line [24].

Questionnaire Survey
Structured questionnaire involving socio-demographic and risk factors associated with malarial infections was developed rst in English and then translated to local language (Afan Oromo) and administered to voluntary study participants from whom blood was taken. Questionnaire was rst explained to laboratory technicians who collected the data at each health facilities. After the purpose of the study was explained to the study participants/children's parents questionnaire was lled out during blood lm collection.

Data Analysis
Collected data were entered into Microsoft Excel. The data were analyzed using SPSS version 25 (SPSS, Chicago, USA). Chi-square test statistics was conducted to determine association between age, gender, season and prevalence of malaria using 95% con dence interval and P-value. P-value < 0.05 was considered statistically signi cant. Descriptive statistics was used to calculate frequencies and percentages. Tables and graphs were also used to present the results. The associations of different risk factors with malaria infection were analyzed using logistic regression analysis together with their corresponding 95% con dence intervals and odds ratio.

Results
Socio-demographic characteristics of the study participants All 356 individuals who were included in the study gave response to the questionnaire. Among study participants two-hundred sixteen (60.7%) were males where most of them were in the age groups above fourteen years old (46.90%). Educational status for most of the study respondents (24.72%) were above grade twelve and most of them (41.57%) were private (daily laborers) ( Table 1).

Malaria prevalence and its associated risk factors
Out of 356 subjects involved in the current study, 17.13% (61/356) of them were con rmed malaria positive ( Table 2). Variable test positivity was recorded at different health facilities which might be due to spatial heterogeneity of mosquito breeding sites in Batu town. Of the total malaria cases identi ed, higher cases (50.8%, 31/61) were occurred due to P. vivax than P.falciparum (45.9%, 28/61). Mixed infections were occurred in 3.3% (2/61) of the total malaria cases ( Figure 2).
With regard to the distribution of malaria cases with sex, out of 17.13% overall prevalence of malaria, we observed 9% and 8.13% in males and females, respectively (Table 3). Sex had no statistical signi cance association with malaria infection (χ2 = 2.026, P = 0.155). Malaria infection occurred among all age groups. We observed higher prevalence of malaria, 11.8%, among age groups of greater than 14 years old (adolescents and adults); whereas for age groups between 0 -4 and 5 -14 years old, it was 2.5% and 2.8%, respectively. There was statistically signi cant association between age and malaria infection (Table 3). The average monthly prevalence of malaria was 16.9%. However, the number of patients visiting health facilities showed uctuating during the months the study was conducted. Relatively greater number of malaria cases (36; 59.0%) were treated during June and July months (wet season) while lower number (25; 41.0%) during April and May months (Dry season) ( Figure 3). However, there was no statistical signi cance difference in malaria prevalence between seasons (χ2 = 3.118, P = 0.077) ( Table 4). Among patients treated for malaria at health facilities in Batu town 82% of patients were treated at private health facilities while 18% at government health facilities (Figure 4).
Logistic regression analysis showed that individuals who use ITN were less likely to get malaria infection (OR = 0.38, 95% CI (0.194, 0.743), and P = 0.005). Individuals who had been living closer to stagnant water were almost three-fold times to get malaria infection than those who were found at a distant from reservoir water (OR= 2.97, 95% CI (1.68, 5.26) and P < 0.001). Similarly, individuals whose houses were sprayed with insecticides were approximately 0.2 times less chance of getting infection with malaria as compared to those whose houses were not sprayed with insecticides (OR = 0.18, 95% CI (0.097,0.34) and P < 0.001) ( Table 5).

Discussion
The nding of the current four months (April -July, 2018) study revealed that malaria infection in the study participants was 61(17.13%). This prevalence is larger than that reported from other parts of Ethiopia such as 0.93% from Butajira [25] and 7% from Arba Minch hospital [5]. However, the current result is less than that of patients attending Wonago health centers, 28.1% [21], Hadiya, 25.8% [26] and Hallaba, 82.8% [9]. The observed differences might be due to altitudinal, seasonal and other climatological variations that contribute for Anopheles mosquito breeding and malaria control measures implemented in the study areas. The lower prevalence of malaria at Buta Jira [25] and Arba Minch [5]compared to Batu town might be Buta Jira is found at higher altitude and at Arba Minch the study was conducted almost during dry season when very low malaria prevalence is expected in the country. On the other hand higher malaria prevalence was recorded at Wonago health center because the study was conducted during major malaria transmission season. However, the reason why higher prevalence was recorded at Hadiya [26] and Halaba [9], almost irrespective of season (similar to Batu town) and altitude (higher than Batu town) might be due to climatic and vector factors that might favour more malaria transmission compared to Batu town. But this needs further investigation.
With respect to Plasmodium species prevalence, P. vivax accounted for the highest cases. This nding is in agreement with the study conducted at Halaba health center, with 70.41% for P. vivax, 23.08% for P. falciparum and the remaining, 6.51% mixed infection [9]. It is also similar with the study conducted at Aleta Wondo that revealed 66% of P. vivax and 34% of P. falciparum infection [27]. The result of the study conducted in Dilla town [6] at health centers also showed P. vivax accounting for 62.5%, followed by P. falciparum for 26.8% and mixed infection with both P. vivax and P. falciparum for 10.7% agreeing with the present nding. From the surrounding area of Dilla town 85% prevalence of P.vivax was also reported [28]. The higher prevalence of P. vivax over P. falciparum might be attributed to the relapsing dormant liver stage and CQ resistance nature of the P. vivax. Study conducted by Ketema et al. [29] on the therapeutic e cacy of CQ treatment for P. vivax showed a two-fold increase in the prevalence of CQ resistant P. vivax in South Ethiopia.
In the present study, relatively males were more infected with malaria than their female counterparts, but did not show statistically signi cant association which is in agreement with the result of retrospective of study of Batu town health facilities [11]. It is also in agreement with the ndings of Regasa [9] at Arba Minch hospital, Alemu et al. [17] from Jimma town who reported higher infection rates of malaria among males than females. The higher prevalence of malaria among males might be males engage in outdoor activities and recreation during the nights outside home which make them more prone to Anopheles mosquitoes breeding sites. Study conducted by Kenea et al. [30] at Adami-Tulu Jido Kombolcha (close to our study area) showed that greater proportion (76.6%) of human biting activities of Anopheles mosquitoes occurs outdoors than indoors during early part of the night. Peak biting times for An. arabiensis (the major vector in the area) [19] that begins early nights [31] coincide with life activities of male individuals. Similar study conducted at Bioko Island, Equatorial Guinea, showed high level of outdoor biting of An. gambiae (s.s) occurred throughout the night [32].
Malaria infection also occurred among all age groups. However, the highest malaria infection occurred in age groups older than fourteen having statistically signi cant difference. This result is in agreement with the ndings of Regessa [9], Molla and Ayele [6]. The highest prevalence of malaria in this age group might be attributed to their stay out of home during biting time of Anopheles mosquitoes.
In the current study area, during the four months, though there was no statistically signi cant difference, more malaria cases were detected during June and July than in April and May. This might be because during April and May it was dry season but rain occurred during June and July which might create conducive breeding ground for Anopheles mosquitoes. Generally, seasonal variation in malaria transmission is a well established feature of unstable malaria where in Ethiopia [33] 2.6% in dry season (April/May) and 5.8% during wet season (September-November) was reported.
In the context of malaria elimination, involving private health sector is essential for completely and timely reporting of malaria cases [34]. Investigation of the role of private health facilities in the town in diagnosing and treating malaria infected patients showed that more number of patients was attending these facilities. Survey study conducted by Jerene and et al. [35] showed that 86% private health facilities in Oromia regional state of Ethiopia were providing malaria diagnosing and treatment services. At private health facilities because of fear of pro t oriented and burden of drug pressure which cause anti-malarial drug resistance, the town health o ce and regional health bureau forces them to treat malaria cases only once con rmed by microscopic examination, but the o ces allows public health facilities to treat malaria cases con rmed by microsope and presumptively as well. Because of this reason more parasitologically con rmed cases were treated at private health facilities. On the other hand, preference of some patients to be treated at private health sectors than government health facilities might be another reason.
In the present study living in the nearby stagnant water was identi ed as a risk factor for malaria infection. Among those infected with malaria, more cases occurred in patients who live nearby stagnant water. This is in line with study conducted in Dilla town and its surroundings by Molla and Ayele [6]. This can be explained by the fact that stagnant water serves as a suitable breeding ground for Anopheles mosquitoes. Those individuals whose houses were not sprayed with insecticide get more likely malaria infection than those whose houses were sprayed. Individuals having and using insecticide treated nets were less likely to get malaria infection than those not having bed nets which is in agreement the ndings of Molla and Ayele [6] and Belete and Roro [22].
The present study is not without limitations. Firstly, the study was conducted during the low malaria transmission season (only within four months) before major transmission season (September to December), therefore, the actual picture of malaria prevalence could not be addressed. Secondly, the study is facility based which involved only symptomatic patients who visited health facilities to seek treatments for some disease conditions. Asymptomatic malaria carriers from the community of the town who do not visit health facilities and seek treatments were not involved due to lack of logistics and nancial constraints. Thirdly, malaria diagnosis was conducted using only with gold standard, microscopic examination. Robust molecular diagnostic techniques such as polymerase chain reaction, loop-mediated isothermal ampli cations which could detect low density infections especially during low transmission season were not employed due to lack of budget, logistics and molecular laboratory set-up. If these techniques were used it would have been possible to detect malaria cases from some of the patients whose samples were reported negative by microscopic examination.

Conclusion
From the result of the present study it can be concluded that both P. falciparum and P. vivax were the dominant species in the Batu town. However, the prevalence of P. vivax is slightly higher than that of P. falciparum. In the current study malaria detection from febrile patients attending health facilities was conducted only by using microscopic examination. However, in order to get real burden of malaria in the town, both symptomatic and asymptomatic studies should be carried out by microscopic examination supplemented with molecular techniques such as polymerase chain reaction and loop mediated isothermal ampli cation. The study also revealed that there is association between malaria and several risk factors. Generally, malaria is still public health problem in Batu town which might be true for other urban areas in the country, therefore appropriate control measures should be scaled-up in the town in order to minimize malaria related morbidity and mortality.