Factors associated with malaria vaccine uptake in Sunyani Municipality, Ghana

Sunyani Municipality is one of the 27 administrative districts in the Bono region of Ghana. Its estimated population for 2019 was 151,378. Sunyani is the Municipality’s capital. It lies between Latitudes 70º20′N and 70º05′N, and Longitudes 20º30′W and 20º10′W. It shares boundaries with Sunyani West district to the north, Asutifi district to the south, Tano North district to the east, and Dormaa East district to the west. The Municipality has a total land area of 829.3 square kilometres. One-third of the total land area is not inhabited or cultivated which provides arable land for development [9].

The Municipal Health Directorate (MHD) is in charge of health administration in the district. There are 33 health facilities that provide care to inhabitants of the Municipality. The Municipality is zoned into 34 functional Community-based Health Planning and Service (CHPS) zones under six sub-Municipalities. The sub-Municipalities are Abesim, Antwikrom, Newtown/Baakoniaba, New Dormaa, Sunyani central, and Penkwase. The CHPS zones are demarcated areas for the delivery of health services [9].

Childhood vaccines are delivered at health facilities and outreach points within the Municipality. The outreach points serve to deliver vaccines at the doorsteps of beneficiaries so that physical access does not hinder uptake. The outreach points are organized under the various CHPS zones in the 6 sub-Municipalities. The Municipality is noted for being one of the best in terms of vaccine coverage in the country. It is one of the reasons why it was selected as one of the implementing districts under the MVIP [9].

The study was an analytical cross-sectional study employing a quantitative approach.

Structured questionnaire and an observation checklist were used to collect data on malaria vaccine uptake and related factors. The assessment was done at a point in time giving a snapshot of the situation. Data was collected from parents/caregivers in Sunyani Municipality about themselves and their children on factors related to malaria vaccine uptake.

Cochrane’s formula was used in to calculate the sample size; \({\text{n}} = \frac{{Z^{2} Pq}}{{d^{2} }}\) where n  =  sample size; Z  =  standard normal variate for margin of error; p  =  proportion of children who take RTS,S vaccine; q  =  1-p; d  =  margin of error.

Since this study was on uptake of a new vaccine, an assumed proportion of uptake of 50% was employed; using a margin of error of 5% and a 10% adjustment for non-response, the calculated sample size was 424.

Stratified sampling technique proportionate to size was used in selecting participants from the six sub-Municipalities which served as strata. There are differences in the characteristics of the sub-Municipalities. Employing stratified sampling technique ensured that all sub-Municipalities were represented in the sample. It also allowed comparison to be made across sub-Municipalities.

In determining the proportionate sample size for each sub-Municipality, the formula below was used:where A is the sampling fraction, ‘y’ is the sample size  =  424, and ‘z’ is the 2019 monthly target coverage for vaccines in Sunyani Municipality  =  505.

The sampling fraction was applied to the monthly target of each sub-Municipality to determine the sample size for each sub-Municipality as shown in Table 1.

Systematic sampling was then used to select respondents from each stratum. A sampling frame was constructed using the Child Welfare Clinic (CWC) registers at the various CHPS zones. The CWC registers contain the official records of each vaccinated child in a particular CHPS zone. The frame contained the names of children who were eligible to have taken the first 3 doses of the malaria vaccine by December 2019, for each sub-Municipality. A sampling interval was determined for each sampling frame using the formula:where K  =  sampling interval, N  =  the number of children in the sampling frame, and n  =  sample size for the sub-Municipality. Simple random sampling was used to select the first sample by writing the names of the children from one to the sampling interval, folded and mixed up in a bowl. One piece of paper was selected and the name on the paper represented the first sample. Subsequent samples were drawn by adding the sampling interval to the number of the first drawn sample until all samples required for the sub-Municipality were drawn.

The parents/caregivers of the selected children were contacted and those who agreed to be part of the study were interviewed.

Data was collected through the administration of questionnaires to respondents and observation of CWC cards. Parents/caregivers were contacted at CWCs or in their houses depending on where they were available to respond to the questionnaire. Questions centred on socio-demographic factors and other independent variables known to affect vaccine uptake. The other independent variables assessed were: knowledge about malaria vaccine, previous experience with vaccines and vaccination, affordability, and accessibility of vaccines in Sunyani Municipality.

The observation checklist centred on the uptake of malaria vaccine. The CWC card of the children provided this information.

Each questionnaire administration and CWC card observation lasted about 20 min.

To ensure voluntary participation in the study, informed consent was obtained from each parent/caregiver before data collection. None of the parents/caregivers contacted refused to participate in the study, indicating a 100% response rate.

The data was cleaned and entered into Microsoft excel. Entries were double-checked for errors and corrections made. It was then imported to STATA version 15 and analysed. Frequencies and percentages were generated for sociodemographic characteristics such as age, occupation, marital status, religion, and sex. Median and ranges were generated for the continuous variables.

Uptake of the malaria vaccine was measured as levels: no uptake (no dose received), partial uptake (either first or second dose received), and full uptake (all first three doses received). Ordinal logistic regression analysis was done to determine the association between the independent variables and the level of malaria vaccine uptake. The regression was done first at the univariate level. Independent variables with significant p values at the univariate level were used in a multivariate analysis and the model with the best Akaike’s Information Criterion (AIC) and Bayesian’s Information Criterion (BIC) was selected. For all associations, significance level was set at 5%.

A total of 424 parents/caregivers and 424 children participated in the study They were drawn from the six sub-Municipalities in the Sunyani Municipality. The study lasted approximately 10 months.

The median age of parents/caregivers was 29 years (27, 32 years). It ranged from 17 to 45 years. Majority of them (99.3%) were parents with almost all being females (99.5%). Most respondents (60.9%) were married with the rest being either single or cohabiting. Up to 43.2% of respondents had up to secondary education, whiles up to 41.5% of their partners, mostly males had up to tertiary education. Most of the respondents were Christians (75%). Up to 55.2% of respondents were self-employed, whiles 20.5% of respondents were unemployed. However, only 2.6% of their partners were unemployed.

Details of the distribution of the socio-demographic characteristics of respondents are shown in the Table 2.

Out of the 424 children, 66.3% were aged 15–16 months. Their ages ranged from 13 to 18 months. The median age was 15 months (15, 16 months). Up to 55.4% of them were males. Almost all of them were delivered at a health facility (94.6%).

Details of the characteristics of children studied are shown in Table 3.

While 94.1% (95% CI 91.4–96.0%) of the children had been administered the first dose of the malaria vaccine, 90.6% (95% CI 87.4–93.0%) had been administered both the first and the second dose with a reduced percentage of 78.1 (95% CI 73.9–83.8%) having been administered all the three doses.

The reasons given for receiving some but not all the doses of the vaccine were: “did not know when the next one was due” − 45.6%, “was not around”, − 23.5%, and “not comfortable with issues surrounding vaccine” − 13.2%. For those who had received no dose of the vaccine, 60% of the mothers said it was their partner’s (husband) decision not to allow their children to be administered the vaccine whiles the rest said it was their own decision to refuse the vaccine.

Distribution of uptake of malaria vaccine in Sunyani Municipality is shown in Table 4.

The uptake of malaria vaccine in Sunyani Municipality shows a declining uptake for the subsequent doses of the vaccine. Whiles uptake for the first dose was 94.1%, it reduced to 90.6% for the second dose and to 78.1% for the third dose. RTS,S 1 and RTS,S 2 uptake met the WHO target of 90% but uptake of RTS,S 3 did not.

A chart of the uptake of the first three doses of malaria vaccine in Sunyani Municipality is shown in Fig. 1.

There was a significant association between the sub-Municipality where one resided and level of uptake. Having a parent who had up to tertiary level education was associated with significantly increased odds of uptake. Details of association between independent variables and level of uptake is shown in Table 5.

Multiple ordered logistic regression analysis using variables that were significant at 5% in the univariate analysis demonstrated that adjusted odds ratio for uptake per sub-Municipality was not significant.

As compared to secondary education, children with a parent who had been educated up to the tertiary level had an increased odds of 4.72 times of completing uptake. Children with parents/caregivers who thought vaccines were becoming too many for them with the addition of the malaria vaccine had 71% reduced odds of full uptake as compared to those who thought otherwise. This association was significant with a p value of 0.001.

Additionally, children who had suffered fever as an adverse reaction had an increased odds of 2.27 of their children completing uptake as compared to those whose children suffered abscess. Details of the multivariate analysis are depicted in Table 6.

Findings from this study indicated an uptake of 94.1% for RTS,S 1; 90.6% for RTS,S 2; and 78.1% for RTS,S 3. Uptake of RTS,S 1 and RTS,S 2 thus met the target of 90% coverage for vaccines set by WHO [10]. RTS,S 3 coverage, however, did not meet the set target.

There was a reduction in uptake of subsequent doses of the vaccine. This observed reduction is similar to that observed in Senegal, Cameroun, Nigeria, Togo, Congo, and in the Kwabre East district of Ghana [11‐16]. The over 90% uptake recorded for the RTS,S 1 and RTS,S 2 indicates that the anti-vaccine campaigns that greeted the introduction of the malaria vaccine did not impact negatively on the uptake of the vaccine in Sunyani Municipality [6]. This may have been so because the messages were largely on social media and did not really seep down to negatively influence parents/caregivers. It may also have been due to effective public education and other community mobilization strategies employed by the Municipality’s health directorate to create awareness about the vaccine when it was introduced.

Out of the 5.9% of children who had not been administered any dose of the vaccine, most (60%) were attributed to a partner’s decision to refuse the vaccine (Table 4). Almost all the respondents were females indicating that it was the fathers who prevented their children from being administered the vaccine. Fathers play a major role in the family and are usually the decision-makers. Those who prevented their children from being given the vaccine may have been influenced by the anti-vaccine campaigns. Fathers are usually not present at CWCs and so are not likely to benefit from education about vaccines which are usually delivered there.

It may also be the case that mothers were unwilling to admit during the interview that they themselves did not want their children to be vaccinated considering the fact that fathers were not around to respond.

The trend of reduced coverages for subsequent doses of the malaria vaccine may be due to poor knowledge of parent/caregivers about the schedule of the vaccines. This could result in parents/caregivers not presenting their children for the subsequent doses on time or not presenting them at all as was the case of 45.6% of respondents (Table 4). Up to 23.5% of children had not been administered all three doses because their parents/caregivers had travelled when they were due. This can be attributed to the fact that not all districts in the country are administering the vaccine (only districts on the MVIP). Therefore, when parents/caregivers travel to these non-implementing districts, their children may not be administered the vaccine at all or on time.

The findings of higher education and occupation being positive predictors of vaccine uptake are consistent with findings made by Adu, Ofosu, and Mukthar et al. [17‐19]. Similarly, the findings of Acharya et al. [15] of higher education being associated with complete uptake was consistent with findings from this study.

Having a higher educated parent was associated with higher odds of complete uptake both in the univariate analysis and the multivariate analysis (AOR: 4.72, 95% CI 1.27–17.55). This could be because highly educated parents have access to more information about the vaccine and were better placed to understand the implementation programme. Since most parents/caregivers who send their wards for vaccination services are women (99.5%), having a partner who has higher education could mean that as the decision-maker, he is more likely to accept the vaccine. Having a higher education is associated with better occupation, the possible reason why civil servants had higher odds of their children completing uptake when compared.

Additionally, having a parent with primary education was also found to be associated with increased odds of full uptake in the multivariate analysis (AOR: 4.10, 95% CI 1.02–16.47) as compared to having a parent with secondary education. This can be attributed to middle level educated parents being more susceptible to misinformation as compared to lower level educated parents. Whiles lower educated parents may rely on official communication such as health education at child welfare clinics and public announcements, middle level educated parents are more likely to be influenced by the anti-vaccine campaigns which were mainly on social media platforms.

The findings of parents/caregivers having the perception that vaccines are becoming too many for their children is unique to this study per available literature reviewed. Those who thought vaccines for children (32.3%) are becoming many had lower odds of completing uptake (AOR: 0.29, 95% CI 0.14–0.61). This could be that parents/caregivers do not see the benefits of the child being vaccinated overriding the potential adverse effect that could occur when the vaccine is given.

Additionally, parents/caregivers who have children who have ever had fever as an adverse reaction (148/223) following immunization had a higher odds of completing uptake as compared to those who had abscess as an adverse reaction (64/223) (AOR: 3.09, 95% CI 1.56–6.09). This could be related to the fact that most parents/caregivers consider fever to be a minor immediate side effect of vaccines as compared to developing an abscess. They were therefore not likely to ‘risk’ going for a new vaccine the safety of which has been questioned.