Previous trend, present prevalence and integrated knowledge, attitude and practice 1 towards malaria and use of insecticide-treated mosquito net in the rural setting of Welkait 2 district, northwest Ethiopia

Background: Malaria is still a burden to Ethiopia, especially among the productive ages of the 2 society. Although insecticide-treated mosquito nets (INTs) are widely practiced by the rural 3 community, malaria remains a leading public health issue. This study aimed to evaluate a 5-year 4 trend, current prevalence and community knowledge, attitude and practice (KAP) towards 5 malaria and use of ITNs as a strategy for prevention of malaria among the rural setting of 6 Welkait district, northwest Ethiopia. 7 Methods: A cross-sectional study was carried out on 403 households to identify the prevalence 8 rate of malaria. Besides, a 5-year (2015-2019) retrospective data was analyzed. Random 9 sampling technique was used to select representative households among the community 10 members. Well-structured questionnaires were used to collect KAP data towards the utilization 11 of ITNs against malaria infection. Blood samples were examined for identification of malaria 12 parasites following standard procedure. SPSS version 22 was used for the analysis of quantitative 13 data. Results: Of 36,219 outpatients examined, 7,309 (20.2%) malaria-positive cases were reported 15 during 2015-2019. Overall prevalence of malaria was continually declined from 2015-2019. Male N = 4,485) were significantly higher than female (38.6%, N = survey Most used during peak season. Majority 83.6% (337) of the respondents have a positive attitude 1 toward ITNs. 2 Conclusion: Although the retrospective analysis showed a declining trend of malaria in the 3 district, the current prevalence survey revealed a significant increase of malaria in the locality. 4 This was contributed by unsustainable distribution and inconsistent use of ITNs among the 5 population. We recommend the governments and stockholders to give emphasis on full coverage 6 of ITNs and educate the community about ITNs usage.


Introduction
Malaria is an infectious protozoal disease from the genus Plasmodium which is transmitted by 10 the bite of parasite-carrier female Anopheles mosquitoes. The genus plasmodium consists of four 11 well known species, namely P. falciparum, P. vivax, P. malariae and P. ovale, which commonly 12 cause malarial disease in humans [1]. Among which P. falciparum and P. vivax are very 13 common and widely distributed species of Plasmodium in Ethiopia and on average accounts for 14 60% and 40% of the whole malaria cases respectively [2]. Among the several Anopheline 15 species that transmit malaria to humans, only Anopheles arabiensis is recognized as primary 16 vector in Ethiopia, while others Anopheles pharoensis, Anopheles funestus and Anopheles nili are 17 considered as secondary vectors [3]. 18 The disease is primarily tropical diseases of most developing countries of the world; which are 19 severely affecting their economy and public health [4]. Overall world malaria cases decreased 20 from 251 million in 2010 to 228 million in 2018, however, Africa still bears the biggest burden 21 of morbidity with 93% cases [5]. Despite high morbidity, Africa also recorded the highest 22 absolute reduction of mortality in 2018 (380,000 deaths) compared to 2010 (533,000 deaths) [5]. 23 Likewise, the world malaria associated mortality rate declined in 2018 because of the expansion 1 of prevention and control measures. Nevertheless, Sub-Saharan Africa countries are largely 2 affected by malaria infection. 3 Ethiopia is one of the most malaria epidemic-prone countries in Africa with 52.7 million 4 people (68% of the population) at risk of malaria infection, especially the most productive age 5 groups [6,7]. Almost 75% of Ethiopia's mainland is endemic for malaria; furthermore, the 6 interaction of mountainous landscape with variable winds, seasonality of rainfalls, and the 7 presences of ambient temperatures creates diverse micro-climates for malaria transmission [3,7]. 8 According to the Federal Ministry of Health (FMoH) National Malaria Prevention and Control 9 Strategy, areas below 2,000 m were considered 'malarious' and prevention measures have been 10 in place half a century ago [8,9]. Among the most effective malaria control interventions, use of 11 long-lasting insecticidal nets (LLINs) and indoor residual spraying of households with 12 insecticide (IRS) are strongly recommended by WHO [5,10]. 13 Several reports indicated consistent and correct utilization of ITNs, particularly the LLINs 14 proven to reduce the transmission rate up to 90% [11][12][13]. IRS is also an extremely effective 15 means of malaria control strategy when properly applied in areas where malaria is epidemic [10,16 14]. In Ethiopia following the WHO recommendation, ITNs have been distributed in all 17 malarious areas free of charge via health workers, volunteers and local administrators. Even 18 though most households owned at least two ITNs per family [9], there is a knowledge gap among 19 the community about consistent and correct use of nets to minimize the impact of malaria in the 20 country [15]. Full coverage and proper utilization of ITNs are vital for the prevention and control 21 of malaria [16], still there are also limitations in sustainable distribution and timely replacement 22 of nets, seasonality of malaria, and poor knowledge associated with malaria, ITNs and the vector. 1 On top of it, the coverage and utilization of ITNs also differ from region to region [17].

2
The main determinants in the ownership and utilization of ITNs as reviewed by Singh et al. 3 [18] educational level, knowledge of malaria, socio-economic status and parity and community 4 participations were found to be the major factors in different settings. These factors contribute to 5 the low efficacy (60%) of the available ITNs [18][19][20]. Bearing in mind, Northern Ethiopia, the 6 Tigray region, out of the total malaria exposed households, only 74% of them received at least 7 one ITN [21]. Hence, studying the KAP towards the use of ITNs in such limited resources is 8 vital beside other efforts made to control malaria in the region. Furthermore, knowing the past, 9 providing the current prevalence of malaria as well as integrated knowledge, attitude and practice 10 towards utilization of ITNs, completes the protection and control efforts of malaria in the region, 11 especially in the district. Therefore, the current study aimed to analyze the past 5-year (2015-12 2019) retrospective data, current magnitude of malaria and explore knowledge, attitude and 13 practice towards malaria and use of ITNs as malaria prevention strategy in the rural community 14 of Welkait district, northwest Ethiopia.

16
Study design 17 The study was designed to analyze triangulated data sources. with mean annual minimum and maximum temperature of 15°C to 30°C, respectively. Malaria 13 was frequently reported annually from September to November. 14 Sample size determination 15 The sample size was calculated following single population proportion formula, 16 reported elsewhere [22]. Assuming that half of the respondents have 17 knowledge on cause and transmission of malaria and use of ITNs with an estimated malaria 18 prevalence rate of 50% (p = 0.5) at 95% confidence interval (Zα/2 = 1.96) and 5% of marginal 19 error (d = 0.05). Based upon the formula, the calculated value was 384 and plus 10% non-20 response rate, the total estimated households were 422. With a response rate of 95.5%, the actual 21 sample size 403, were included in both KAP and parasitological studies. A multi-stage cluster 22 random sampling technique with town as the first-stage administrative unit, kebele as the second-23 stage, village as the third-stage, health development group (Gujile) as the fourth-stage and 1 household as the fiveth-stage was used to select the representative sample size. First, among the 2 three towns, Maygaba town was selected using lottery method. Since Maygaba town has only 3 one kebele, three villages from six villages were randomly selected by the investigator. In each 4 selected village, four Gujiles (total 12) were randomly selected and households were selected given Gujile to determine a sampling interval for selecting households. A systematic random 9 sampling techniques was used to select every n th household. For KAP study, the head of the 10 household (male or female) or representative older than 18 years was considered as eligible for 11 interview. Household members who were unable to communicate, mentally handicapped and 12 children less than 18 years old were excluded. While for parasitological examination, any 13 member of the family older than 18 years if he/she are willing to give blood samples for malaria 14 test and no history of anti-malaria therapy within the previous two weeks were used for blood 15 sample collection. 16 Data collection 17 Retrospective health data 18 The National Malaria Prevention and Control Program of the country has established a standard 19 protocol for the detection of malaria parasites from patients' blood. Accordingly, the past five 20 year (2015-2019) malaria health examination record data have been collected from MHC using  Giemsa (3%) solution as per the standard protocol [25]. The presence of malaria parasites in 100 2 fields were examined from the thick films. The thin films were used for identification of 3 Plasmodium species. About 5% of the slides were randomly selected and re-examined by the 4 senior expert in MHC for quality control. 5 Data analysis 6 Wholeness and consistency of the data were checked twice and entered into SPSS version 20 7 software (SPSS Inc, Chicago, IL, USA) for statistical analysis. Descriptive statistics (frequencies 8 and percentages) were used to tabulate and describe the cross-sectional data. The Chi-squared 9 (χ 2 ) test was applied to analyze the retrospective data and determination of association between 10 dependent and independent variables. Statistical significance was defined at p-values < 0.05.

11
Ethical considerations 12 The study was ethically approved by the Health Bureau of Welkait district and College of 13 Natural Science Institutional Review Board (CNS-IRB), Addis Ababa University. Before 14 conducting the investigation, the researcher discussed the study with local administrative bodies 15 in the study area. All the study population were clearly informed about the purpose of the study 16 and kindly asked to participate in the study. Blood samples were collected by trained staff of 17 MHC and all malaria positive cases were carefully treated according to the national malaria 18 guidelines [26] 19  (Table 1). 4 Regarding the distribution of sex, from the total tested cases, slightly over half (54.7%, N = 5 19,797) of patients were males and (45.3%, N = 16,422) were females. While from the total 6 slide-confirmed cases, the majority (61.4%, N = 4,485) of them were males and the rest (38.6%, 7 N = 2,824) were females. Looking at the overall male to female ratio of tested and slide-8 confirmed cases, mostly males were affected more frequently than females with a ratio of 1.2:1 9 and 1:0.6 respectively. This difference was statistically significant (χ 2 = 3.923, p < 0.05). 10 Consequently each year, a higher number of malaria positive males was observed than the 11 malaria positive females, however yearly difference in the number of cases was not statistically 12 significant (p > 0.05). Comparing the annual overall prevalence of malaria under each sex 13 category, males showed higher (22.7%) prevalence than females (17.2%). The total number of malaria examined over malaria positive cases showed great variation 17 between seasons. With respect to the number of cases, total cases exhibited the following order: 18 autumn > winter > summer > spring (Fig. 1). Majority of suspected as well as infected cases 19 were observed soon after the main rainy season (September-November). A total of 34.2% of 20 malaria negative and 37.5% malaria positive cases were reported in this season (Additional file 21 1). While the smallest number of malaria-suspected (16.6%) and -infected (11.7%) cases were 22 reported during the small rainy season (March-May). Almost the same malaria-suspected cases 23 (25.5 and 23.8 %) were observed in dry and heavy rainy seasons respectively. Significantly 24 higher number of slide-confirmed malaria cases (37.5%) were observed during autumn soon 1 after the heavy rainy season (p < 0.01) compared to dry and small rainy seasons. Generally, 2 pairwise comparisons indicates that there are significant inter-seasonal variations (p < 0.001) 3 except between autumn and summer (p = 0.376). Since the actual season of malaria in the study 4 area was not clearly defined, it is not possible to calculate the prevalence of malaria for each 5 season. Nevertheless, this imperative result tells us that malaria was observed throughout all 6 seasons.

7
In terms of infectious category, P. falciparum was the major (66.1%) contributor to malaria 8 infection in the study area, while P. vivax accounted only (33.9%) of infection ( Table 1). The  Table   13 1and Fig 1). So, it is not surprising that P. falciparum was the most prevalent infection than P. 14 vivax in the past five years at the study site. Despite dominancy and prevalence, both infections 15 followed similar patterns towards seasonality (Fig 1). About 37.8, 31.2 and 19.1 of P. falciparum 16 infection were registered in September-November, December-February and June-August,  Only 403 households who provide complete information were considered in blood films 4 collection ( Table 2). From which two-hundred and forty-eight (61.5%) blood donors were male.

5
A total of nineteen slide-positives were obtained, of which twelve cases were from male and 6 seven cases were from female. Nearly half (48.9%) of the blood film providers were the most  During a cross-sectional survey, a total of 422 household heads or their representatives were 1 involved. Out of which nineteen households were excluded from the analysis due to the 2 incompleteness of the information provided. Four-hundred three participants' information were 3 completed with a response rate of 95.5%. The majority of the respondents were male (61.5%, N 4 = 248). The details of the participants' demographic data are summarized in Table 3   the participants preferred health centers as ways of treatment (Table 4).  were properly utilizing their ITNs with insecticide treatment (Table 6). The five years retrospective data analysis revealed that the overall prevalence of the malaria was 3 showing a declining trend except a slight increase observed in 2016, this was due to the 4 occurrence of unseasonal rainfall in the study area as the information obtained from the MHC 5 indicates. Although the health system data showed a continual declining of malaria prevalence in 6 the study area, the current prevalence rate obtained from community based parasitological survey 7 do not support this finding, proves that the efforts that have been made by the district as well as 8 the regional health office to control malaria in the study area was not sufficient. This was also 9 further identified by unsustainable distribution and inconsistent use of ITNs observed during the  This study revealed that the current prevalence of malaria in the Welkiat district was found to 5 be 4.7%, this finding was much higher than the prevalence reported in other malaria endemic The health record data of this study indicate that significant number of male malaria-confirmed 8 cases recorded in the MHC compared to the female, suggesting that males were extremely 9 infected by malaria in the study area. The result also coincides with the finding obtained from a 10 parasitological survey of the current study (Table 2). This finding was comparable with a similar following the rainy season that may serve as a food source for mosquito larvae to complete the 1 life cycle of mosquitoes. In both retrospective and community based cross-sectional surveys, P.

16
2 falciparum was predominantly found in the study area and become a major (66.1%) contributor 3 of morbidity and mortality. This is in agreement with the national report [3,30] as well as most 4 other regions of Ethiopia [2, 28].

5
The cross-sectional study revealed that the majority (91.3%) of the respondents had considered 6 malaria as their major public health problem and over (75%) of the respondents also identified 7 that such disease is transmitted through the biting of mosquitoes. In this study, the awareness of  does not grant the reduction of malaria related morbidity and mortality in the study area unless 4 and otherwise consistent usage and sustainable distribution of ITNs among the community has to 5 be practiced regularly. This was simply explained by high prevalence of malaria in the study area 6 obtained via community based parasitological survey and retrospective data analysis.

7
Nevertheless, the parasitological study has to be supported by advanced diagnostic methods such 8 as molecular diagnosis using polymerase chain reaction (PCR) to further confirm the results 9 obtained through microscopic examination. Additionally, it could have been better to design the 10 cross-sectional survey accompanied with direct observation of ITNs usage in the study area.

11
Finally, it is difficult to draw better and reliable conclusions from five years of health recorded 12 data, analyzing as much recorded data as possible may provide more qualified conclusions. 13 However, due to poor recordkeeping at MHC and lack of malaria related follow up data, for 14 instance morbidity and mortality data were missing and limited our study.

16
In conclusion, the overall prevalence of malaria in the study area was declined during the study   in the study area. All the study population were clearly informed about the purpose of the study 16 and kindly asked to participate in the study. Prior to administering the questionnaires and blood