Study design and study setting
This was a community-based descriptive cross-sectional study using a quantitative data collection method. The study therefore used community structures to access the study participants.
The study was carried out in the Amolatar district, which is located in the Lango sub region of northern Uganda. The Amolatar district is bordered by the Apac district, Kwania district to the North, Dokolo district to the Northeast, Kaberamaido district to the East, Buyende district to the Southeast, Kayunga district to the South and Nakasongola to the West. The administrative headquarters of the district of Amolatar is 85 kilometers (53 mi) by road in southern Lira city. This location lies approximately 185 kilometers (115 mi) by road, Northwest Kampala, the capital of Uganda and the largest city in the country. Amolatar is covered by two large water bodies, Lake Kyoga and Lake Kwania. Amolatar experiences two rainy seasons per year, with heavy rains from March to May and sometimes light rains between September and December. The peak incidence of clinical malaria follows the peak of rain, with a delay of approximately 4 to 6 weeks.
Study population
The study population included children <5 years of age, and the target population included mothers/caretakers of children <5 years of age in the Amolatar district from which the sample was drawn.
Inclusion and exclusion criteria
The mothers/caretakers of children under the age of 5 years who were true residents of the household or who had migrated from another area for six months or more were included in the study. The consent of the caretaker and assent of the children was ideal, whereas mothers/caretakers of children under the age of 5 years who were severely ill talking during the interview were excluded.
Sampling size determination
The sample size was estimated using the Kish Leslie 1965 formula for a single proportion [20].
Where N is the sample size.
Z - Z score at a confidence level of 95% (1.96)
P – Estimate of the prevalence of malaria in children under 5 years of age
e - Margin of error (+/-5%).
According to a study on the age-specific prevalence of hospitalized paediatric malaria patients in Uganda, the average malaria incidence in children aged 0-59 months in the country was 16.9% in 2018-2019 [25, 26]. Substituting the value in the formula,
N = 216 (sample size)
To account for nonresponses, 10% of the sample size was added to obtain the actual sample size. This implied that N= (10/100) *216 + 216 = 237.6 = 238 respondents (mothers/caretakers of children under 5 years) and 238 of their children under 5 years were tested.
Study variables
Dependent variables
The prevalence of malaria among children under 5 years of age was detected using Plasmodium falciparum malaria rapid diagnostic test (PF. mRDT). The following procedure was followed:
The Research Assistants began by checking the expiry date on the test packet, putting on the new pair of gloves for each child, opening the packet and removing the test, pipetting and then writing the name of the child on the test. Then, the alcohol swab was opened, the child’s 4th finger was grasped on the left hand, and the finger was cleaned using an alcohol swab. The finger was allowed to dry before pricking, the lancet was opened, and the child’s finger was pricked to obtain a drop of blood; the lancet was then discarded in the safety box. The data collector then gently squeezed the bulb of the pipette and touched its tip to the drop of blood. The bulb of the pipette was gently released, and the blood was drawn to the first line of the pipette; the tip of the pipette was touched to the sample hole marked “s”, which was gently squeezed to transfer the blood, and the pipette was discarded in the safety box. Then, two (2) drops of buffer were added to the assay well marked “A”. The samples were incubated for 20 minutes, after which the results were read. For positive results, a line in “C” and a line in “T” indicated that the child had P. falciparum malaria, and for negative results, a line in “C” and no line in “T” indicated that the child had no P. falciparum malaria. For invalid results, no line in “C” and a line or no line in “T” indicated that the test was invalid. The gloves and alcohol swabs were also discarded in the non-sharp biohazard pill. The children who tested positive for P. falciparum malaria were referred to the nearest health facility for malaria treatment, and the mothers appreciated their time. In all the steps, the COVID-19 guidelines were followed.
Independent variables
Factors associated with the prevalence of malaria and mothers’/caretakers’ perceptions of the use of LLINs and IRSs as interventions for malaria prevention and elimination were assessed using a semi-structured questionnaire and responses summarized as proportions.
Data collection methods, tools and procedure
Both semi-structured questionnaires and observation checklists for quantitative data were used. The questionnaires were pretested by a researcher in the field before actual data collection to ensure data validity. Pretesting was carried out on 9 participants in one of the villages in the Amolatar town council that was not part of the study setting to help determine the effectiveness of the questionnaires and adjustments made where necessary. The questionnaires were written in English and translated to the Lango language with the help of traditional leaders. The accuracy of the translated questionnaires was checked through back translation. Eight (8) data collectors comprising 4 health workers and 4 VHTs were trained on data collection, blood sample collection and screening for Plasmodium spp., a causative agent for malaria disease, using PF. mRDT of children under 5 years. All the collected data were entered into an Excel spreadsheet, saved and kept under lock and key.
Quantitative data collected
Semi-structured questionnaires and observation check lists were used to collect data on sociodemographic characteristics, the use of LLINs and the IRS, and mothers’/caretakers’ perceptions of the use of LLINs and the IRS for malaria prevention. The investigator used face-to-face interviews to gather participants’ views on the use of LLINs and IRSs as interventions for malaria elimination, and a blood sample was obtained from the respondents’ children under 5 years of age to determine the prevalence of malaria among children under 5 years of age in the Amolatar district. After the interview, the respondents were debriefed and appreciated for their participation. At most, 8 interviews were conducted by each data collector under the close supervision of the investigator each day.
Data management and analysis
The principal investigator reviewed the field questionnaires to ensure the completeness and accuracy of the collected information. The coded data were entered into Epidata software and cleaned. The data were analysed using STATA version 15. During the analysis, the principal investigator conducted univariate, bivariate and multivariate analyses. The Hosmer–Lemeshow test was used to test for the goodness of fit of the final logistic regression model by the backwards elimination method
Univariate analysis was used for both categorical and continuous variables. Categorical variables such as sex, education, marital status, and religion were summarized using proportions, percentages and frequencies, while continuous variables such as age were summarized using the mean, mode, range and standard deviation. This is presented in the table and graph.
Bivariate analysis was used to assess the associations between independent and dependent variables. The results are expressed as chi-square values and p values at 95% confidence intervals, and statistical significance was set at a P value of 0.2, as presented in the table.
Multivariate analysis: Variables that were significant in the bivariate analysis were reconsidered for multivariate analysis. The results are expressed as adjusted odds ratios at 95% confidence intervals, and independent variables with p values <0.05 were considered to be significantly associated with the prevalence of malaria among those under five years of age. Significant variables from bivariate analysis were run in a logistic model using the backwards elimination method, and the results are presented in the table.
Ethics approval
Approval to conduct the study was obtained from Gulu University Research Ethics Committee (GUREC) number GUREC-2021-118, permission to collect data was sought from the Chief Administrative Officer (CAO) through the District Health Officer (DHO) of Amolatar district; through to the sub county accounting officers (SACO), the parish local council II (LCII) and the village local council I (LCI); where required number of households with children under five years in each village were reached for interviews and a drop of blood sample removed from the children’s 4th finger. The risk involved of children crying when collecting blood sample was well explained to the participants. They were also informed that, they were free to withdraw at any time without coercion. To confirm their participation in the study, a written informed consent and assent forms were obtained from the eligible research participants (mothers/caretakers and children under 5 years) after thorough explanation of the research objectives, and the main goal of the study, risks from the study and how to mitigate the risks in the language (English/Lango) they could understand and why their participation was necessary. To guarantee confidentiality of the respondents, each was assigned a unique identifier code that was used instead of their names. All the necessary steps were followed sticking to research and ethical procedures, guidelines and other relevant regulations.