Factors Inuencing Caregivers’ Health Seeking Behavior for Malaria Treatment of Children Under 5 Years in Busia Municipality, Uganda

Background: Malaria remains a serious cause of under-ve mortality and morbidity worldwide and Uganda inclusive. This burden can be minimized by promptly seeking health care. In Uganda, however, studies around malaria health-seeking behaviors for under-ve children in the most malaria prevalent areas are very few. This study aimed at determining the factors inuencing caregivers’ health-seeking behavior for malaria treatment of children under ve years in Busia Municipality, Uganda. Methods: A cross-sectional research design was used with a structured questionnaire to collect data. Data were analyzed using SPSS Version 22 to establish relationships between the variables. Results: The results showed that the current health-seeking behaviors of the caregivers of under-ve children in Busia municipality are associated with caregiver education level (p= 0.008), the health worker’s behavior towards the client (p=0.015), the severity of fever (p<0.001), the severity of last malaria episode (p<0.001), waiting time (p=0.001), the quality of health services (p= 0.001) and age of caregiver (p<0.001). Traditional medicine and home remedies are the most utilized means of malaria management in under-ve children in Busia Municipality. Conclusions: Caregivers need to be sensitized about the proper health-seeking behaviors for the management of malaria in children under-ve years through radio shows, television, community engagement meetings among others to enhance the knowledge and understanding of communities about the recommended malaria treatment-seeking practices. There is a need to continuously train medical workers on client engagement skills to promote a good relationship with patients and encourage their return. a signicant relationship caregivers’ health-seeking

gestational anemia which is associated with abortion [16]. These adverse effects can be prevented if malaria is treated promptly in approved medical facilities within 24 hours of the appearance of the rst symptoms.
Unfortunately, there is reported poor health-seeking behavior by caregivers for their children as far as malaria infection is concerned. Since 2015, hospital admissions have remained almost constant within the country at 60% of the con rmed laboratory diseases [11]. The poor health-seeking behaviors related to malaria infection have been largely attributed to the low caregiver education, low economic status of the family members [14], low knowledge on malaria, low perceived malaria severity, increased belief in traditional care [17], negative experiences with the healthcare facilities [8], and unavailability of the required health services in the nearby health facilities [9]. Given these factors, more than 60% of all suspected malaria cases are treated outside the formal health sector, making the projected number of malaria cases to be as high as 60million per annum [5]. Usually, caregivers rst resort to self-medication, use of traditional medicine like herbs, and informal facilities [6].
Malaria affects productivity and adds to the already high costs of care at the household and national levels. Also, malaria has a noteworthy deleterious impact on Uganda's economy due to the loss of productivity from decreased school attendance and sickness. Infection from malaria is reported to cost a family more than 2% of their annual revenues [10]. Overall, Uganda's GDP is highly affected and will continue being affected unless a permanent solution to malaria infection is found. As such, several measures have been put in place to deal with the malaria-related burden. Uganda accepted the Roll Back Malaria (RBM) and foresaw the enhancement of control practices as a basis to attain the regional targets for malaria control. The RBM initiative emphasized admittance to the most appropriate treatment for malaria within 24 hours of onset by both under-ve children and pregnant women [6]. Additionally, Uganda's efforts against malaria were recently guided by the 6-year Uganda malaria reduction strategic plan for 2014-2020. Through this plan, the ministry of health targeted reducing malaria morbidity to 30 cases per 1,000 by 2020 [6] through a rapid and synchronized nationwide scale-up of cost-effective interventions to achieve universal coverage of malaria prevention and treatment.
Despite the existence of this plan and the RBM initiative, anecdotal evidence showed a surge in malaria cases in Uganda by over one million between June and August 2019, with some border districts like Busia being highly affected [10]. Therefore, the purpose of this study was to identify the factors in uencing caregivers' health-seeking behavior for malaria treatment of children under ve years in Busia district, Uganda. The following objectives guided this study; To establish the caregiver characteristics in uencing their health-seeking behavior for malaria treatment in children under ve years in Busia municipality.
To determine the health system factors in uencing caregivers' health-seeking behavior for the treatment of malaria in children under ve years in Busia municipality.
To identify the treatment-seeking patterns for malaria in children under ve years in Busia municipality, so as to make appropriate recommendations.

Methods
This was a cross sectional study design in Busia Municipality, Uganda. 236 primary caregivers responsible and caring for at least one child aged 5 years and below participated in this study. These caregivers permanently stay in Busia municipality. The sample size was randomly determined using [7].
Where: n = desired sample size if target population > 10,000 Z = the standard normal deviation (1.96) and it corresponds to a 95% con dence level. The child to the selected caregivers had ever suffered from fever within the last 6 months. The fever had been con rmed by a quali ed health professional as malaria-related. Caregivers below 18 years were excluded from this study. Data was collected using a structured researcher administered questionnaire and analyzed using SPSS version 22. Logistic regression was conducted to establish whether the independent variables predicted the observations on the dependent variables. To ensure data reliability, the research tool was pre-tested on 24 participants outside Busia district. Also, data collectors were trained and orientated about the correct use of the research tool. Table 1 shows the caregiver characteristics as described below. In this table, most of the respondents (41.5%) were in the age group between 26-35 while the lowest was 14.0% for 46 years or more. 28.0% of respondents were between 18-25 years while 16.5% lie between 36 and 45 years. Table 1 below also indicates that 53% of the caregivers are female while 47.0% are male. Most of the respondents (51.3%) have a secondary level of education, followed by 25.8% who have a primary level of education. 14.8% are uneducated while only 5.9% have achieved a tertiary level education. 2.1% of the respondents reported that they had technical skills through various training in areas like tailoring, handicrafts. And only 5.9% of them had attained a tertiary level education. 42.8% of caregivers of under-ve children are married, followed by 32.6% who are cohabiting, 13.6% are single, 8.1% are separated/divorced while 3.0% are widows/widowers. 28.4% of the caregivers of under-ve children had rst heard of malaria through health care providers, followed by 26.3% from television, 23.3% had heard from the radio, 18.6% from family members while the lowest at 3.4% had heard of malaria from newspapers. 79.2% of the caregivers of under-ve children are aware that fever/high temperature is a sign and symptom of malaria. 68.2% know that loss of appetite is a sign and symptom of malaria while only 25% do know that convulsions can be a sign of malaria. 36.9% of the respondents stated that cough is a sign and symptom of malaria which is a misconception while 27.5% indicated that bloody diarrhea is a sign of malaria which is a misconception as well.

Results
However, Table 1 results also indicate that there are misconceptions such as bloody diarrhea and cough being signs and symptoms of malaria in under-ve children. The highest percentage of caregivers at 35.2% believe that malaria is caused by mosquito bites, 23.3% believe it is caused by cold or changing weather, followed by 14.8% who believe that it is due to other causes. 11.9% believe that malaria is caused by drinking dirty water, 10.2% believe it is caused by eating dirty food while the lowest (4.7%) believe it is caused by birds.
Majority of respondents (44.1%) decried the quality of health services in Busia municipal council rating them as poor, followed by 30.9% who rated them as fair, 16.1% as good while 8.9% rated them as excellent. These results are not exciting and point out that most caregivers of under-ve children are dissatis ed with health system quality in Busia municipal council. Many caregivers at 35.2% stated that the physical accessibility of health facilities is fair, followed by 28.4% who rated it as poor, 23.3% rated it as good while 13.1% rated it as excellent. This shows that health facilities in the Busia district are generally inaccessible physically. Most of caregivers (38.6%) stated that the waiting times at the health facilities are fair, followed by 30.1% who rated it as poor, 21.6% rated it as good while 9.7% rated it as excellent.   Table 3 indicate that the severity of the last malaria episode is found to have a signi cant (p-0.001) in uence on health-seeking behavior for malaria treatment in the under-ve child. These results also indicate that for children with very severe malaria, 65.3% of them were taken to a formal health facility, while 33.3% of those with mild illness, and 17.9% of those with very mild illness attended a formal healthcare facility respectively. These ndings in Table 3 also show that whether a blood test was conducted to con rm malaria in the under-ve child or not and whether the severity of malaria in uenced the decision to seek medical assistance were statistically signi cant at p=0.031 and p-0.001 respectively.

Discussion
The above ndings have largely contributed to meeting the three objectives of this study. The results of this study show a signi cant relationship between the caregivers' health-seeking behavior and their level of education. These ndings are similar to those of [12] who established that caregiver education is a serious predictor of treatment-seeking behavior. Caregivers with at least post-secondary school education are more likely to initiate treatment in the formal sector for their febrile children compared to the less educated respondents who mostly utilize avenues such as home treatment and traditional healers.
Although the majority (79.2%) of the caregivers are aware of the common signs and symptoms of malaria in their children, a reasonable percentage of the caregivers do not know the exact symptoms of malaria. For example, 25% of the caregiver do know that convulsions can be a sign of malaria, 36.9% and 27.5% of the caregivers think cough and bloody diarrhea respectively are symptoms of malaria. These results indicate that misconceptions about malaria exist among caregivers and some caregivers are not even well informed.
The ndings of this study expose the treatment-seeking patterns for malaria in the under-ve child in developing countries like Uganda. Many caregivers do self-management of malaria at home using home remedies and herbs and might even consider seeking treatment from traditionalists before the health facilities. Most caregivers (38.6%) resort to traditional medicine when they realize that a child has a fever that is likely to be caused by malaria. Few (18.6%) caregivers use Western medicines like artemisininbased combination therapy (ACT) as their rst choice. Traditional medicines and home remedies are the commonly used means of malaria management in under-ve children. Some of the traditional medicines used involve concoctions of medicinal plants locally known as "halulu" which is bitter due to the quinine content, "mulusa", concoctions of moringa leaves among others. Home remedies also involve the use of packaged Aloe vera products or locally concocted ones.
Although the MOH, Uganda recommends solely seeking consultations in health facilities for any child found to be ill, these results indicate that these recommendations have largely not been honored by caregivers, hence, putting under-ve children at great risk of complicated malaria and fatality. These ndings agree with those of [15] who noted that the current behavior patterns in Africa indicate that a signi cant proportion of suspected malaria cases are self-treated at home or in informal health facilities where malaria diagnoses are rarely available. These ndings are also similar to those of [13] who noted that home treatment is seen as a type of rst aid for commonly experienced symptoms, such as fever. In Malawi and Tanzania, mothers prefer to rst treat their febrile child with antipyretics, and further manage the fever with tepid sponging, and then seek antimalaria drugs if the fever persists. Similarly, in Nigeria, malaria is not considered life-threatening, and mothers seek care at higher-level facilities when selftreatment has failed [14]. This trial-and-error process acts as a home-based diagnosis. If the fever continues after the application of home-based fever reduction methods, then the caregiver presumes their child has a more severe case of malaria and seeks treatment in the formal sector.
WHO [2] recommends that febrile children are tested and treated appropriately with ACTs within 24-48 hours on symptom onset. Unfortunately, 59.7% of the caregivers of under-ve children in this study indicated that no blood test was carried out before the initiation of antimalaria treatment for their children. This nding exposes the health system weakness that hinders appropriate malaria management. The initiation of malaria treatment in children without due procedures is a serious risk to their health as it compromises the quality of treatment and can lead to mortality especially when the cause of the fever is from a disease other than malaria.
On the side of the caregivers, the majority of the children are taken to nd treatment for malaria when severity has been recognized. The results of this study show that majority of the caregivers are reactive to malaria severity, hence, posing a serious challenge to the health of the under-ve children. The caregivers go to the health facilities late when the disease has progressed to complicated levels. Additional evidence shows that severity is one of the strongest and most common predictors of seeking immediate care in the formal healthcare sector by caregivers. Caregivers are more likely to seek care in the formal sector if the child's illness is perceived to be severe [15].
Even though nding treatment for a caregiver giver is in uenced by caregiver factors, also health system factors have a role to play. The quality of health services as perceived by the caregivers signi cantly impacts their health-seeking behavior for the malaria treatment of their under-ve child in a formal health facility. These ndings agree with those of [8]. These authors noted that the quality of service that is provided in a formal health facility in uences whether caregivers will bring their febrile child or not. Formal healthcare facilities are perceived as providing more services such as malaria testing as well as having trained and experienced doctors and personnel [15]. Negative experiences through long waiting times can have the opposite effects and dissuade caregivers from returning to a speci c health facility [8]. Subsequently, health workers' behavior was found to be signi cantly associated with the caregivers' health-seeking behavior. Caregivers who rated the health worker behavior as excellent were 2.2 times more likely to take their under-ve children to a formal health facility than those who rate them as poor.
These ndings conquer with ndings from studies by [13] who noted that government facilities are often criticized for treating patients poorly, which can create negative experiences that can dissuade caregivers from seeking care from them in the future.
As evidenced in the ndings of this study poor malaria treatment-seeking behaviors of the under-ve children's caregivers are aggravated by caregiver education and age. These study ndings indicated that caregivers of under-ve children in Busia municipality are generally under-educated as only 5.9% of them have achieved a tertiary level of education. This low level of education signi cantly affects their decision to seek the recommended care for the malaria treatment of their under-ve child. Additionally, caregivers who were older (36-45 years) had signi cant treatment-seeking behavior for malaria treatment (p<.001) compared to their younger counterparts.
This research was however limited by recall bias. Some of the analyses relied on the respondent's ability to recall details about health-seeking behavior for malaria treatment. Additionally, this study was limited by scope. This study was limited to the perceived "malaria", as reported by respondents and what health workers typically and presumptively diagnose as "malaria." It is concerned with what people perceive to be malaria illness episodes, as opposed to the actual prevalence of malaria as con rmed by clinical laboratory tests. This approach was ideal for the study area that lacked reliable malaria diagnostic facilities that are also consistent with the WHO presumptive diagnosis and treatment guidelines [2].