Knowledge, awareness and practices regarding malaria in Western Area Rural District, Sierra Leone: a household-based cross-sectional survey at the midpoint of the National Malaria Strategic Plan (2016-2020)

Background Sierra Leone is a highly endemic area for malaria, and the implementation of the National Malaria Strategic Plan (2016-2020) has reached its midpoint. To provide more specic guidance for interventions in the future, a household-based cross-sectional survey was conducted to elucidate the knowledge, attitudes and practices regarding malaria and malarial control measures among the general public. Methods Three communities (Grafton, Jui and Kossoh) in the Western Area Rural District that were in close proximity to Sierra Leone’s capital city of Freetown were included. Households were randomly selected and interviewed with a structured questionnaire covering malaria infection, diagnosis, treatment and prevention, as well as knowledge of malaria prevention, in collaboration with the National Malaria Control Program of the Ministry of Health and Sanitation, Sierra Leone. Results A total of 262 qualied questionnaires were included. The average cost for meals per day is approximately 30000 Leones in each household. The rate of awareness, indicated by reporting having heard of malaria, was 98.1% (257/262), and 86.6% (227/262) of the respondents knew that mosquito bites are the main route of transmission. In addition, 80.9% (212/262) of the respondents sought health advice or treatment for the illness, and a similar percentage of respondents had been tested for malaria, mostly with rapid diagnostic tests (RDTs). A hight demand for long-lasting insecticidal nets (72.1%) matched the serious shortage (61.8%, 162/262), and of the households that reported a lack of nets, 66 had children younger than ve years old. these supprt


Background
Malaria is endemic in Sierra Leone; the entire population is at risk of exposure because Sierra Leone is an area of stable malarial endemicity [1]. Sierra Leone's high malaria disease burden accounted for approximately 48% of outpatient morbidity and approximately 38% of mortality in children younger than ve years according to the national Malaria Indicator Survey conducted in 2016 [2]. Although signi cant progress with regard to reducing mortality has been made, it is still a major public health concern. The control of malaria remains a priority in the national health agenda of Sierra Leone.
A multipronged strategy to contrl malaria, including vector control interventions, diagnostic testing, treatment with quality-assured artemisinin-based combination therapies, and standardized case management training manuals for community and facility treatment providers, has been developed and implemented in Sierra Leone based on the World Health Organization (WHO) recommendations [3]. The goals are for 80% of the population to have appropriate levels of knowledge and uptake/practice of malaria prevention and management strategies, all the at-risk population to have access to the appropriate preventive interventions, 80% of individuals with suspected malaria cases to have access to con rmatory diagnostic testing, and all individuals with malaria to receive effective treatment by 2018, according to the current Sierra Leone National Malaria Strategic Plan (2016-2020) (SL-NMSP) [3].
The awareness of malaria symptoms, transmission, prevention and treatments is closely related to the implementations of malaria control programs [4][5][6]. A knowledge, attitudes and practices (KAP) survey about malaria could help policy makers better understand the current situation and guide malaria control strategies. Therefore, this study is conducted to determine the knowledge, attitudes and practices regarding malaria and its control measures in the general public through a household-based survey at the midpoint of the implementation of the SL-NMSP.

Study site
Three communities (Grafton, Jui and Kossoh) in Western Area Rural District within close proximity to Sierra Leone's capital city Freetown were included in this household-based survey on malaria. This district is home to a very religiously and ethnically diverse population. Grafton is a trade center. Jui and Kossoh are neighborhood towns. Kossoh town is surrounded by a large forest reserve, and its major industries are farming and coal mining.
Participants, questionnaire and data collection A total of 300 households in the three designated communities were randomly selected as survey respondents using a cluster sampling design. One adult member from each selected household was interviewed with a structured questionnaire covering demographic information, economic status, basic knowledge of malaria, and malaria prevention and control practices. When medicines were discussed, packages of various drugs were displayed to the participants.
The survey was conducted between 17 December and 19 December 2018 in collaboration with the National Malaria Control Program of the Ministry of Health and Sanitation, Sierra Leone. Local staff from Sierra Leone-China Friendship Biological Safety Laboratory [7] were trained in prior to their involvement in the administration of the survey.

Data management and statistical analysis
Data were input and cross-checked in EpiData version 3.1 [8,9]. Then, descriptive statistics and percentages were obtained using Microsoft Excel 2010 software. Differences in distributions of variables according to the daily cost of meals in each household were evaluated using the chi-square (χ 2 ) test or Fisher's exact test by SPSS software version 20.0 (IBM, USA) and P < 0.05 was considered statistically signi cant.

Results
A total of 300 households were investigated, and 262 quali ed questionnaires were nally included in the analysis after checking for questionnaire completion and logic. The survey covered 1645 residents, with approximately 6 persons per family. The average cost of meals per day was approximately 30000 Leones in each household (one US dollar was equal to 8000 Leones in December 2018). Three groups were established according to the cost of meals. Group 1 was de ned as families paying ≤ 20000 Leones/household per day for meals, and this group contained 475 persons in 97 households; Group 2 was de ned as families paying ≤ 40000 and ≤ 20000 Leones/household per day, and this group contained 778 persons in 123 households; and Group 3 was de ned as families paying > 40000 Leones/household per day, and this group contained 392 persons from 42 households.

Malaria infection, diagnosis and treatment
A total of 308 persons in 87 households, 430 persons in 106 households and 191 persons in 39 households in Group 1, 2 and 3, respectively reported having had a fever in 2018 (P = 0.594, Fisher's exact test). The distribution of individuals with malaria among the groups in 2018 is shown in Table 1.   Furthermore, 84.5%, 84.6% and 83.3% of the respondents in Groups 1, 2 and 3, respectively, reported that they and their family members always visited a health care worker or doctor when they suspected they had contracted malaria (P = 0.487, Fisher's exact test). Among the individuals who went to see a health care worker or doctor, they reported visiting a professional every time or most of the times they suspected they had contracted malaria, and there were no differences among the three groups (χ 2 = 2.882, df = 6, P = 0.830) ( Table 4). Artesunate and lumefantrine (51.7%) and artesunate andamodiaquine (19.1%) were the top two choices of combination therapy. The most common reason given for the early termination of a course of treatment was having recovered prior to nishing all the doses (P = 0.815, Fisher's exact test). Moreover, approximately 80.2% of the respondents reported starting anti-malarial medicine on the same day that they noted the fever (P = 0.095, Fisher's exact test), and approximately 64.5% of them responded that not all kinds of anti-malarial medicine were free (P = 0.399, Fisher's exact test) ( Table 4).

Malaria prevention measures
In this survey, 188 households had long-lasting insecticidal nets (LLINs) (χ 2 = 1.466, df = 2, P = 0.482), but more nets were required because 189 respondents said that they did not have enough LLINs in their households (P = 0.253, Fisher's exact test), so resulting in children under the age of 5 years in 66 households not having LLINs to sleep under (P = 0.410, Fisher's exact test). Only 146 respondents had slept under nets the night before the survey (P = 0.331, Fisher's exact test). Furthermore, some members in 162 households with nets did not sleep under LLINs, which could lead to cross-infection (χ 2 = 4.359, df = 4, P = 0.363). The LLINs were mainly distributed by the governmental hospital/health centers and community health centers (Table 5). In addition, a total of 143 households never sprayed insecticide indoors to kill mosquitoes (P = 0.217, Fisher's exact test), and 225 households never sprayed insecticide outside (P = 0.003, Fisher's exact test); the proportions of households that did not take implement these preventive measures were especially high in Group 1 (89.7%) and Group 2 (89.4%). Furthermore, more than three quarters of the respondents reported that they did not have a door or window screen to prevent mosquitoes from entering their houses (P = 0.020, Fisher's exact test), and the proportion of respondents without screens was especially high in Group 1 (85.6%). Additionally, most families reported staying outside of the house at night (P = 0.800, Fisher's exact test) ( Table 5).

Knowledge of malaria
In this survey, almost all the respondents had heard about malaria (98.1%, 257/262) and knew that the main route of transmission was through mosquito biting (86.6%, 227/262). Fever, body aches or joint pain, and loss of appetite were the top three clinical manifestations they reported (Table 6). A total of 173 respondents reported that they would always go to see a doctor or health care worker when they suspected they had malaria, while 48 interviewees said they would take some of the anti-malarial medicine kept in their houses rst. The expensive costs (mentioned by 176 respondents) and the long distance (mentioned by 26 respondents) were the main factors preventing people from visiting the doctor or health care worker (Table 6). Sleeping under LLINs (198), keeping their surrounding clean (84) and using insecticide spray intended for use against mosquitoes (20) were the most common methods of malaria prevention employed (Table 6).

Discussion
Malaria remains one of the most serious public health issues and is responsible for high proportions of morbidity and mortality in Sierra Leone. In the present study, overall high levels of knowledge regarding the causes of malaria, prevention mechanisms and symptoms, and active seeking of treatment for malaria from health care providers were found among Sierra Leoneans, similar to the results in the MIS 2013 & 2016 [2,10]; these results with regards to knowledge and practices support malaria control [11][12][13].
Globally, vector control is the most commonly accomplished through the use of LLINs and indoor residual spraying (IRS), with the aim of reducing the transmission by preventing human-vector contact and killing vector mosquitoes [14,15]. The goal was for the entire at-risk population to emply preventive measures, including the use of LLINs, IRS and larval source management, by 2017 in Sierra Leone. However, 72.1% of the respondents required more nets, which may explain the high percentage (61.8%, 162/262) of households and the high proportion of children under ve years old who did not sleep under the nets at night according to this survey. This indicates a large gap in obtaining and using nets by the most vulnerable groups who are not bene ting as much as hoped from preventive malaria interventions. In addition, screens for doors and windows that are cost-effective to install and maintain are a supplementary public health intervention to prevent humans from being bitten by mosquito vectors indoors, thereby signi cantly reducing the transmission of malaria.
However, screens seem to have been neglected in the regular malaria control campaigns [16,17], resulting in fewer than 25% of households having door or window screen in this survey, and a general low awareness of the role of screens. In addition, a low level of implementation of IRS and outside spraying and a high level of engagement in activities outside of the house at night were found, despite the presenece of many mosquitoes.
The early diagnosis and prompt treatment of malaria reduces the transmission of the disease and prevents deaths. It is critical for people to seek diagnosis and care as soon as they experience any symptoms of malaria. It is encouraging that more than 80% of the respondents reported that they visited the doctor in a timely manner, and a similar percentage of respondents reported being tested for malaria, usually with RDTs. However, approximately three-quarters of the respondents said that the doctors did not interpret the test results for the patients. Therefore, a good opportunity to provide health education and promote malaria control and prevention is being missed. A home-based RDT that could be performed by trained family members rather than at a healthcare facility may improve the timeliness of the diagnosis of malaria [18,19]. Moreover, the reported adherence to completing the course of antimalarial treatment was close to 80% at the clinic or if self-administration at home. Reasons for the lack of adherence were reported to be sickness after the rst dose, insu cient food to take the medicine, forgetting to take the medicine, and poor instructions provided by the community health center (CHC) [20]. In addition, the cost of medical treatment and the distance from the hospital must also be considered.
However, there were some limitations of this study. First, this was a small-sample cross-sectional study conducted in 3 communities, and the ndings are not yet nationally representative. Second, the actual differences in the cost of meals among the three groups in this study were not signi cant. Third, the collected data were based on respondent recall, and the data may be biased. Therefore, the results of the study highlight the need for more extensive investigations of the KAP regarding malaria in Sierra Leone.

Conclusion
In conclusion, in these three communities, public awareness of malaria prevention measures was relatively high in this survey, but the malaria burden was still higher than because the lack of su cient anti-malarial supplies and the limited use of preventive measures against malaria. Thus, the existing knowledge of malaria should be sustained and reinforced, and the availability and use of malaria prevention measures should be promoted to achieve the goals of the SL-NMSP.