This was a cross-sectional study that utilized a questionnaire survey to collect qualitative, independent, and explanatory variables and malaria rapid diagnostic test (mRDT) kits to determine the outcome variable, which was asymptomatic malaria in adults living in Nnewi North LGA. Regression analysis was used to determine which of the qualitative variable is predictive or protective of asymptomatic malaria in Nnewi North LGA, Anambra State.

This study was conducted across the four sub-towns ( Figure 1A) of the Nnewi North LGA, Figure 1B of Anambra State, and the southeast geopolitical enclave of Nigeria. The Nnewi kingdom was founded in four quarters (large villages or sub-towns): Otolo, Uruagu, Umudim, and Nnewi-Ichi. Most people in this area are of the Igbo ethnic origin. The commonly spoken language is Igbo, while the majority of people practice Christianity. Nnewi is home to many major indigenous manufacturing industries, including the Ibeto Group of Companies, Cutix and AD switch, Uru Industries Ltd, and Omata Holdings Ltd. The majority of industrialists in the cluster of spare parts factories in Nnewi are also traders. Major trading centers include the Nkwo Nnewi, Nwafor, and Eke Ichi markets. The traditional ruler of Nnewi North LGA is referred to as the Igwe of Nnewi, and popular community festivals include the Afiaolu and Ofala festivals. Nnewi hosts several institutions and places for learning and healing, including Nnamdi Azikiwe University Teaching Hospital (NAUTH). NAUTH provides specialized and comprehensive medical care ^{18, 19} . Malaria is endemic and perennial in Anambra, with transmission occurring mainly from April to December for approximately 9 months ²⁰ .

Participants were recruited using a stratified convenience sampling technique from markets, banks, offices, and churches within the sub-towns, with sub-towns as strata and ensuring diversity in terms of socioeconomic status and access to healthcare. Recruitment was performed after obtaining ethical approval, permission from the chairman of the Nnewi North LGA, and consent from each participant. Recruitment was conducted for two weeks in November, from 7 November 2024 to 20 November 2024, at the end of the rainy season.

Participants were considered eligible for inclusion if they were healthy, 18 years of age or older, and living in the sub-towns of Nnewi North LGA in Anambra State for at least the past 3 years. People who had lived outside a malaria-endemic country for most of their life and within the last three years and those who did not give consent were excluded. A balanced representation of participants from urban and rural areas was ensured.

A detailed, structured, and validated questionnaire Consent form and RESEARCH QUESTIONNAIRE_Malaria_Epi Study), comprising 19 items divided into six sections assessing the participant's demographics (seven items), knowledge of malaria disease (five items), knowledge about malaria prevention (13 items), knowledge of malaria treatment (two items), perception, attitudes, and beliefs towards malaria (17 items), and community activity towards malaria (five items) was developed to collect information on the independent variables from the participants during the study. The questionnaire was presented in English and was self-administered or researcher-administered if the respondent could not fully understand English. Pre-tests and re-tests were conducted with four volunteers who were not part of the study to determine the reliability of the questionnaire items before the questionnaire was finalized. The pre-test was carried out on the 12th of September 2024 and the retest was carried out on the 18th of September 2024. Face validity was assessed by an expert and a non-expert in epidemiology. The overall Cronbach's alpha for the questionnaire during the pretest was 0.88, ( Pre-Test and Re-Test) indicating good internal consistency. Cohen's kappa values were calculated to assess the reliability of the responses between the pre-test and retest for each section of the questionnaire. Overall, the questionnaire demonstrates good to excellent reliability, with Kappa values consistently falling between 0.72 and 0.82, indicating strong consistency in responses across the pre-test and re-test. The questionnaires were modified, as suggested, before they were included in the application for ethical approval. Consent was obtained before obtaining any data or blood samples.

A finger prick was used to obtain 5 μl of blood from the respondents to detect the presence of infection with P. falciparum using a rapid lateral flow immuno-chromatographic in vitro antigen detection test kit for detecting P. falciparum HRP2 (histidine-rich protein 2) (First Response Malaria Antigen P. falciparum (HRP2) Card Test, Premier Medical Corporation Limited and Standard Q Malaria P.f Ag, SD BIOSENSOR) according to the manufacturer’s instructions. The results were read within 20 minutes, during which time the respondents were able to complete the questionnaire.

This study was prone to selection bias, as the sample of this population who were not able to give 20 minutes of their time were excluded from the study. This study was also prone to recall bias, as some participants may have recalled past events differently over time. The researcher-administered questionnaire may have prompted favorable answers, thus modifying our primary data in some way.

This study aimed to determine the prevalence of malaria among adults in Nnewi North LGA and the associated social, economic, environmental, and behavioral determinants. The sample size was calculated based on the prevalence of malaria in children under 5 years of age. This was appropriate because we intended to determine whether the prevalence in a population that serves as a reservoir for continual transmission is related to the prevalence in individuals under 5 years of age. The latest Nigeria Demographic and Health Survey (NDHS) conducted in 2018 showed that the prevalence of malaria in under 5’s in Anambra State was 8.8% ² . The prevalence data obtained for children during the national survey in Anambra were used to calculate the sample size of the adults in this study.

The sample size was determined using the simple Cochran sample size formula ²¹ :

n = Z ² P(1− P)/ d ²

where n = sample size, Z = Z statistic for a level of confidence (1.96), P = expected prevalence or proportion (0.088*), and d = precision (0.05; 95% CI). n was calculated to be 123, and when we allowed for 10% attrition, the sample size was 137 (123/0.9). In total, 140 questionnaires were used in this study. 35 questionnaires were to be used per sub-town, but we used slightly more in some sub-towns due to non-responses and time constraints.

Ethical approval for the study was obtained from the Ethics Committee, Nnamdi Azikiwe University Teaching Hospital (NAUTH) – NAUTH/CS/66/VOL.17/VER.3/103/2024/081- on 9 ^th October 9, 2024, ( ethical approval). Written informed consent was sought from all participants, and the data was de-identified using the ‘Safe Harbor’ method and is publicly available ( Questionnaire responses (ALL ITEMS). The participants were informed of their right to withdraw from the study at any time.

The obtained data were analyzed using Stata 17/BE, and significant associations were measured at the 5% alpha level (p < 0.05). Descriptive statistics were used to summarize the data, while inferential statistics (chi-squared tests and logistic regression) were used to identify significant associations between the factors studied and malaria prevalence. Qualitative data were analyzed to identify recurring patterns and insights that complemented the quantitative findings. Bivariate statistical analyses were performed. Bivariate analysis - Chi-square (χ2) - was used to determine significant dependence between the categorical independent variables and malaria-positive outcomes. The independent variables with a significant relationship with malaria were fitted into the binary logistic regression model, and malaria status as the dependent variable was binary. The results of logistic regression are presented as odds ratios (OR). These ratios represent the magnitude of the malaria risk. OR, also known as crude odds, represents the likelihood of an event when other variables are not taken into consideration. The likelihood of contracting malaria increases when the OR is higher than one (OR > 1). In situations where the OR is less than 1 (OR < 1), the chance of malaria is reduced. All results were presented as 95% confidence intervals.

Outcome variable. The dependent or outcome variable in this study was the asymptomatic malaria status determined through malaria rapid diagnostic tests (mRDT). This variable was defined in a binary manner, with a positive malaria result of 1 and negative result of 0.

Independent potential explanatory variables. These include sociodemographic characteristics, socioeconomic factors, healthcare-seeking behavior, use of malaria prevention measures, environmental conditions, and perceptions of malaria risk, obtained through answers to the items in the questionnaire. These independent variables have already been categorized. All responses were double-checked, and some were converted into dummy variables for modelling purposes.

A total of 160 adults were informed of the aim of the study and approached, and 140 adults were eligible for the study. Of the 140 patients who were confirmed as eligible and gave their consent to be included in this study, only 130 completed the questionnaires and were tested for malaria parasites using the mRDT, giving a response rate of 92.86%. However, with this response rate, we met our enrolment target of 127 participants. Data from all subjects were analyzed. At every stage of our recruitment process, non-response was due to exclusion criteria, time constraints and phobia for needles on the part of the potential participants even after they were calmed.

Most participants were aged between eighteen and thirty-seven years, Table 1, followed by those that were thirty-eight to fifty-seven years. Very few (7) participants above seventy-eight years old were recruited into the study. More males than females participated in the study. Very few (4) persons with no formal education were recruited, and trading was the most popular profession among this adult population, followed by farming, and then being a student. In addition, there were few participants working in a professional role and as civil servants, even though more participants had attained post-primary education, with those attaining secondary education being the highest. The participants resided mainly in a rural area, with most living in households of one to four persons, including themselves, followed by those living in a household of five to eight persons. We tried to equalize the number of participants from each sub-town, but there were more people from Nnewichi and Umudin who participated in the study. However, the number of participants was evenly distributed across the sub-towns.

Of the 130 participants examined for malaria infection using the mRDT, 34 tested positive, thus giving a prevalence rate of 26.15%, as shown in Figure 2. All parasites detected were P. falciparum, using an HRP2-based mRDT kit.

The major contributors to the prevalence of asymptomatic malaria among the different demographics and socioeconomic categories were adults of fifty-five to seventy-seven years of age, Table 1, adults with professional jobs, adults living in urban areas, adults living in households with nine to twelve occupants, and residents from Otolo sub-town, even though the number of participants in these sub-groups was smaller in the category. Other subgroups majorly contributing to the malaria prevalence in this community include males and those with secondary education, even though their numbers were smaller compared to other subgroups of their category.

All participants were aware of malaria as a disease ( Table 2), with the majority (60.77%) having malaria in the year of study recruitment. Those who had previously had malaria in the year contributed more to the prevalence of asymptomatic malaria than those who did not have malaria in the year, but the difference between these subgroups was not significant.

However, the data on malaria may be affected by their knowledge of the signs and symptoms of malaria and the transmission of malaria. Figure 3 shows that most participants relied on personal feelings and assumptions as the criteria for determining that they had malaria, with a minority relying on confirmatory laboratory tests.

For preventive strategies, we have a greater contribution to malaria prevalence than those who never or rarely use insecticide-treated bed nets ( Supplemenary data: Supplement 1: Participants knowledge on malaria prevention. The greatest contribution is from those who often use mosquito repellents/coils. Those that rarely sprayed insecticides contributed the most to the prevalence of malaria, whereas those that never fumigated their environment were the greatest contributors. Counterintuitively, the more malaria preventive activities that were carried out, the greater the possibility that individuals who carried them were infected with P. falciparum. This was reflected in the use of antimalarial chemoprevention, nets on the windows, or prayer. The findings presented for these preventive strategies thus far have not been found to be significant. Among the various categories, the only malaria preventive strategy that was significant was the covering of water containers around the house. Evidently, those that always covered water containers did not have asymptomatic falciparum malaria Supplementary data: Supplement 1: Participants knowledge on malaria prevention.

During a malaria episode most participants would use antimalarial drugs Supplementary data: Supplement 2: Knowledge and attitude for malaria treatment, and did contribute less to asymptomatic malaria, however, their contribution to asymptomatic malaria was not significantly different compared to those that did not use antimalarial drugs. Herbal remedies were commonly used, and those that used them contributed less to asymptomatic malaria. Using local concoctions, taking lots of fruit, prayers, and visiting the hospital made no difference to the carriage of P. falciparum. None of the three participants that did nothing had asymptomatic malaria at the time of testing. None of the differences were statistically significant.

There is no consensus about the effectiveness of most malaria control strategies deployed in the communities in Nnewi North LGA ( Supplementary data: Supplement 3: Perception, attitudes and beliefs towards malaria control. This lack of consensus may not correspond with asymptomatic malaria. For example, most participants agreed or strongly agreed that sleeping under an insecticide-treated net (ITN) effectively prevented malaria, but those who strongly agreed contributed the most to asymptomatic malaria, showing a discrepancy in belief and practice. On the other hand, those who strongly disagreed that their community was actively involved in malaria prevention efforts contributed the most to asymptomatic malaria. Only one range of perceptions and beliefs was different and concurrent. And, this is shown in the agreement that ‘the cost of malaria prevention tools, such as Insecticide-treated nets (ITNs), insecticides and mosquito repellents, is reasonable,’ those that believed this contributed significantly less to asymptomatic malaria; with those that agreed or strongly agreed having a lower odds of having asymptomatic malaria, with an odd ratio of 0.29 or 0.27 (95% CI = 0.07/0.06 – 1.24), compared to those that strongly disagree, Table 3.

Importantly, most participants were optimistic that malaria could be eradicated from their community but did not think that the malaria level in their community was currently low ( Supplementary data: Supplement 4: Community activity towards malaria control, and they had a higher odds of having asymptomatic malaria ( Table 3). It was encouraging that most of them know that their community faces challenges in preventing and treating malaria and that they need to do more to prevent the transmission of malaria.