FACTORS ASSOCIATED WITH THE ADHERENCE TO UNSUPERVISED DAILY DOSE TO SEASONAL MALARIA CHEMOPREVENTION IN BUILSA NORTH DISTRICT, UPPER EAST REGION OF GHANA

Background: Since 2015, SMC with amodiaquine–sulfadoxine-pyrimethamine (AQ–SP) has been implemented during the high malaria transmission season in three regions of Ghana. Adherence to the Seasonal Malaria Chemoprevention (SMC) is one of the critical determinants for successfully protecting children under five. In Ghana, SMC In Ghana, the SMC implementation was started in 2015. The purpose of this study was to determine the level of Adherence to SMC and the associated factors among caregivers in Builsa North District. Methods: We conducted a cross-sectional study in four (4) sub-districts in the Builsa North District. The district was stratified into six, and four subdistricts were selected using simple random sampling. We started recruitment at the community chief’s house the researcher randomly selected the direction of movement. The 435 participants were recruited via balloting using a random sampling procedure. The knowledge of caregivers on Malaria and SMC was assessed by scoring correct answers to questions related to malaria and SMC. The association between SMC adherence and independent variables was tested using Pearson's chi-square test and Fisher Exact Test. Simple and multiple logistic regressions were performed to determine associations between SMC adherence and the independent variables, with all results interpreted at 95 % confidence level. Results: The SMC adherence rates among the caregivers in the Builsa North District was 95.36% (416/435) with an awareness level of 97.94%. The reasons reported for non-adherence were the child's refusal (38.46%), the child's vomiting of the drug (33.33%), the occurrence of an illness within the period of medication (15.38%), the forgetting to give the subsequent doses (12.82%). About half (49.31%) of caregivers had a good knowledge of malaria, while 66.51% had a fair knowledge of SMC. Significant predictors of SMC on multiple logistic regression were the place of residence aOR=3.59, 95% CI [1.02-12.56] caregivers dissatisfied aOR=0.10, 95% CI [0.01-0.74], and being informed by a friend aOR=0.04, 95% CI [0.00-0.51]. Conclusion: This study found a very high adherence rate to SMC in Builsa North District. The key factors associated with SMC Adherence were being aware of SMC through non-health professional source, place of residence, and caregivers' satisfaction with the previous SMC campaign.

1, and a wrong answer was scored as 0. Then total scores were categorized into four groups: poor knowledge, insufficient knowledge, good knowledge, and very good knowledge. The results were presented in frequencies and percentages using tables and graphs. Frequencies were generated to describe categorical variables. Continuous variables were expressed as arithmetic means and standard deviations for normally distributed variables, median, and quartiles for variables not following the binominal distribution.
Pearson's chi-square (ꭓ 2 ) test was used to compare the proportions in cross tables with high observed observation, and the Fisher exact test was used to compare the proportions with low observed observation. The bivariate analysis (Fisher exact test and Chi-square test) was used to assess the crude association between potential predictors of the adherence, while univariate analysis was performed to estimate the strength of associations between the adherence and its predictors after controlling for the covariates. After then, all the independent variables with p-value from 0% to 50% were used to perform different regressions models.

Results
A summary of the various characteristics of participants is in Table 1.

Adherence level to the SMC first round in 2020
Most of the caregivers interviewed, 90.87% (396/435) reported that all their children received the AQ+SP from drug distributors during the first round of 2020. The adherence rate to SMC among caregivers of those whose children were covered during the first-round drug distribution was 95.36 (416/435) % (Fgure1). Figure 1: Adherence level of caregivers to SMC during the first in Builsa North District 2020 Reasons for non-adherence Four main reasons were reported by the 18 caregivers who did not adhere to SMC during the first round of drug distribution. The primary cause they said for non-adherence was the child's refusal which represented 38.46% (7/18), followed by the child's vomiting of the drug administration, the occurrence of an illness within the medication period. Besides, 12.82% (2/18) reported that they forgot to give the drugs the subsequent two days. These results are shown in Figure 2 below.    Nearly half of the respondents, 43.56%, reported that local radios are their source of knowledge on SMC, followed by health centers 29.04% and community health workers 11.71%. However, some minorities reported public announcements, friends, and TV as their source of knowledge on SMC presented in Figure 4.

Sociodemographic and adherence to unsupervised daily treatment doses
Respondents under the age of 20 (5.26 %), Muslims (10.42 %), married (61.34 %) trained (35.51%), single and widowed (8.33 %), those with primary education (7.04 %), housewife (5.38 %), those residing in smaller households (5.88 %) and those with two or more children were more likely to be non-adherent. However, there was no statistically significant relationship between age, gender, marital status, race, educational degree, profession, household size, number of children under five, SMC knowledge level, Malaria knowledge level, and adherence (p-value?). There was statistical significance for adherence with source of knowledge (p-value = 0.007) and satisfaction level (p-value = 0.000). These results are illustrated below in Table 4.  Table 5 shows the results of MLR, SMC adherence analysis, and selected independent variables. Data from the 388 caregivers in Builsa North District who said they received the AQ+SP in July 2020 were used to perform this logistic regression model. Place of residence, level of satisfaction, marital status, number of children under five, occupation, SMC Level of knowledge, knowledge of SMC's purpose, source of knowledge on SMC, household size was used to perform a simple and logistic regression. The place of residence, the level of satisfaction, knowing SMC through a friend, residing in a household of at least five other persons, affected the SMC adherence. The odd of adherence is 2.62 times higher among caregivers residing in urban areas than caregivers residing in rural areas. This finding was not found to be significant while unadjusted, but it was found to be statistically significant when adjusting for other factors (Adjusted OR = 2.62, 95% CI= [0.96, 7.14]). Participants who reported being very dissatisfied with drug distributors had 98% lower adherence odds than those who reported being very satisfied (Crude OR=0.02; 95% CI = [0.00-0.27]). Respondents reporting their knowledge source as a friend had 93% lower odds of adherence than those reporting their knowledge source as health centers (Crude OR=0.07, 95% CI = [0.01-0.41]). Respondents who indicated that their household composition is six or more is 3.24 times more likely to adhere than those who indicated that their household size is less than six (Crude OR=3.24, 95% CI = [1.04-10.04]).

Discussion
This study has identified some factors that significantly influenced the adherence to unsupervised daily doses of SMC among caregivers in Builsa North District during the first SMC campaign in 2020. The study shows that the adherence rate to the SMC equal to 95.36%. This study also provides further evidence that having more than one child under five, inquiring information on SMC through any other sources deafferents from health professionals and being dissatisfied with the SMC are associated with a decrease of the odds of adherence to SMC. In this study, the caregivers who adhered to SMC represented 95.36% Diawara et al. reported that self-reported adherence to SMC is similar to 95%, but the reliability in this regard is uncertain [6]. While this was not in SMC's setting, a study comparing various malaria preventive regimens in Ugandan children showed that adherence to a 3-day course of dihydroartemisinin-piperaquine was much higher when reported by the caregiver (~100%) compared to the unbiased concentration of drugs [7]. A lower adherence was found using a longitudinal study in Nigeria by Ward et al. Their research found that the adherence was 83.8% [8]. Several factors may explain these observations, and a mix of these reasons rather than one alone is likely responsible. First, there may have been a good understanding of SMC in the Ghanaian community than Nigerian because the Ghanaian health system is better in terms of coverage than Nigerian's [9]. It is possible that the SP-AQ intervention improvements over the years contributed to increasing the adherence in Builsa North District by reducing the frequency and/or severity of infections among participating children. A similar study in southern Ghana and other African countries using AS+AQ also found a high adherence level (81% -97%) among caregivers [10]. These findings indicate that high adherence to AS+AQ can be reached in remote areas where a successful treatment education program is in place. However, the adherence rate found in this sample could be an overestimation. A randomized, placebo-controlled SMC trial in Ghana showed nearly 100% self-reported adherence to SMC's 3day course across all research communities. However, some caregivers were found to have residual SMC tablets that had not been administered [11]. Another way to measure the adherence is by testing the levels of medications in children's blood during a follow-up period [11]. To reduce malaria incidence in children under five within an endemic community, a high proportion of adherence to SMC is indispensable. The adherence demonstrated by this study is associated with mainly to the good knowledge of SMC, the knowledge of SMC's purpose, the larger household size, the place of residence, the satisfaction of caregivers with previous SMC, the source of knowledge of caregivers about SMC. The study revealed that the respondents demonstrated a better awareness of malaria and SMC. This result is not different from what Mazigo et al. and Ingabire et al. reported on the knowledge of SMC [12]. Having good knowledge is a basic predisposition for committing to adhere to all three doses of AQ-SP. There is a tremendous need to intervene in the communication channels related to the SMC. The caregivers who reported that there had been introduced to SMC through their friend are likely to not adhere to the 3-day protocol. In contrast, those introduced through professional sources are likely to adhere. This factor may be explained by a lot of rumors that go from one to another about all the mass drug administration (MDA). Some people do not trust any free medication. In some remote areas, people still trusting only traditional medicine. This knowledge gap may explain the lack of importance of those who reported they forgot to give medicine to children. There was a difference in household size observed in this study. Households composed of more than 5 people are more likely to adhere to SMC; this may be due to the increased number of people (sisters, brothers) to take care of children, the primary caregivers. The perceptive factor assessed in this study is the satisfaction level with SMC. The majority of caregivers reported that they are satisfied with SMC. However, the study found that the tendency to adherence varied significantly with caregivers' level of satisfaction, the more they are satisfied more they adhered to SMC. In Mali Diawara et al. found that parents' opinions about SMC were very positive, with 99.9% of parents reporting that they felt the strategy was good or very good, and 99% of them were in favor of continuing the intervention. SMC's strong support and favorable opinion are major assets for continuation and scale-up [13] The reasons for non-adherence in this study were the child's refusal to swallow the drug, vomiting, illness after the first dose, and the mother's forgetting about the medication. In Niger, Ding et al., 2020b found that children refusing to take the drug; suboptimal health worker instructions; small children spitting out medication, even if it is dissolved; vomiting within 30 minutes of dosing; caregiver saving medicines for the treatment of another family member with acute malaria later on; caregivers sharing/giving medication to older children who were not eligible for SMC; fathers not allowing the medicine to be taken; and fatigue of giving medication, mainly when the child is not sick were the main reason for non-adherence cited by study participants were the main reasons of nonadherence to SMC [14] Ultimately, an integrated approach to funders, policymakers, health workers, applied health social scientists, drug distributors, community members, and all stakeholders will be appropriate for achieving the National Malaria Control Program's goal in the Upper East region of Ghana.

Limitations of the study
Study results were based on the opinions and experiences of a representative Builsa North District subdistricts, not for all the Upper East Region of Ghana. Therefore, the results can be generalized only to the sampled population in the district and not to the whole regional or national level. The present study adopted a quantitative approach only. A mixed approach (both quantitative and qualitative) would help define the deep reasons behind the non-adherence influencing factors.

Strength of the study
The strengths of this study include the fact that the study was carried out by a research team that was not involved in the implementation of the SMC. There is a possibility for selection and observer bias as with any observational study. Study participants were selected at random to eliminate these biases. The potential of recall bias could not be omitted; to minimize them, the survey was conducted in the four subdistricts about one week after the first round of SMC 2020.

Conclusion
The present study found the level of SMC adherence in Builsa North District, Upper East Ghana. The main reason for non-adherence is the child's refusal to take the drug, the child vomiting, the occurrence of a disease, the caregiver's forgetting about medication. Caregivers with good knowledge, those who have obtained extensive information about SMC from health care providers, those who have only one child, those who live in Town are likely to adhere to SMC. The results of this study indicate that the caregivers in the district of Builsa North have a very good knowledge of the cause of malaria, its consequences on children, and its preventive methods. It also shows that more than half of caregivers in the district are aware of Seasonal Malaria Chemoprevention. There has been considerable progress in Ghana's implementation of SMC at the district level. However, some factors regarding SMC, like child refusal to take medicine, still have a bad effect on adherence rate to unsupervised doses. If specific strategies are not implemented to control them, they could spread and undermine all attempts made so far to achieve the reduction of the burden of malaria in children under five may be spread.

Ethical Consideration
Ethical approval for the protocol of this survey was obtained from Ghana Health Service Ethical Committee [GHS-ERC 034/02/20]. In August 2020, community meetings were held with caregivers and local community representatives to explain the study's purpose, after which communities were offered the choice to participate in the survey. Signed consent before the interview was also obtained from all the caregivers interviewed during the study. The study had been carried out during the COVID-19 pandemic, so participating in this study presented a minimal risk of COVID-19 transmission. To ensure the study participants' and research assistants' protection, the study adopted safety measures such as PPEs (Personal Preventive Equipment) by study participants and the interviewers. The study ensured that the interviews maintain a social distance of at least 6 feet (about two arm lengths).

Funding information:
This study is aligned to a master's degree in Public Health funded by the World Health Organization-Special Program on Tropical Disease Research (WHO/TDR) through the African Regional Training Center (ARTC) and Office of Research Innovation and Development (ORID) at the University of Ghana.