Background: Although a safe and effective vaccine is available, measles remains an important cause of mortality and morbidity among young children in Sub-Saharan Africa (SSA). The WHO and UNICEF recommended measles-containing vaccine dose 2 (MCV2) in addition to measles-containing vaccine dose 1 (MCV1) through routine services strategies. Many factors could contribute to the routine dose of MCV2 coverage remaining far below targets in many countries of this region. This study aimed to assess the prevalence of MCV2 utilization among children aged 24–35 months and analyze factors associated with it by using recent nationally representative surveys of SSA countries. Methods: Secondary data analysis was done based on recent Demographic and Health Surveys (DHS) data from eight Sub-Saharan African countries. In this region, only eight countries have a record of routine doses of measles-containing vaccine dose 2 in their DHS dataset. The multilevel binary logistic regression model was fitted to identify significantly associated factors. Variables were extracted from each of the eight country’s KR files. Adjusted Odds Ratios (AOR) with a 95% Confidence Interval (CI) and p-value ≤ 0.05 in the multivariable model were used to declare significant factors associated with measles-containing vaccine dose 2 utilization. Result: The pooled prevalence of MCV2 utilization in SSA was 44.77% (95% CI: 27.10–62.43%). In the multilevel analysis, mothers aged 25–34 years [AOR = 1.15,95% CI (1.05–1.26), mothers aged 35 years and above [AOR = 1.26, 95% CI (1.14–1.41)], maternal secondary education and above [AOR = 1.27, 95% CI (1.13–1.43)], not big problem to access health facilities [AOR = 1.21, 95% CI (1.12–1.31)], four and above ANC visit [AOR = 2.75, 95% CI (2.35–3.24)], PNC visit [AOR = 1.13, 95% CI (1.04–1.23)], health facility delivery [AOR = 2.24, 95% CI (2.04–2.46)], were positively associated with MCV2 utilization. In contrast, multiple twin [AOR = 0.70, 95% CI (0.53–0.95)], rural residence [AOR = 0.69, 95% CI (0.57–0.82)] and high community poverty [AOR = 0.66, 95% CI (0.54–0.80)] were found to be negatively associated with MCV2 utilization. Conclusions and recommendations: Measles-containing vaccine doses 2 utilization in Sub-Saharan Africa was relatively low. Individual-level factors and community-level factors were significantly associated with low measles-containing vaccine dose 2 utilization. The MCV2 utilization could be improved through public health intervention by targeting rural residents, children of uneducated mothers, economically poor women, and other significant factors this study revealed.
Secondary analysis was performed based on the recent Demographic and Health Surveys (DHS) of eight Sub-Saharan African countries. Generally, there are thirty-six countries in the Sub-Saharan Africa region, among them only eight countries have a record of routine dose of MCV2 in their DHS dataset. Those countries were Angola, Burundi, Malawi, Nigeria, Sierra Leone, Tanzania, South Africa, and Zambia. The DHS used a cross-sectional survey study design to collect the data and the study was conducted in those eight Sub-Saharan African countries. The DHS survey in those countries was conducted from 2015 to 2019 (Table 1). Year of the survey by countries The DHS is a nationally representative survey conducted in countries with low and middle income. Eight countries’ datasets were appended together to investigate MCV2 utilization and associated factors among children aged 24–35 months in Sub-Saharan Africa. We used the Kids record dataset (KR file) and children who have data on MCV2 utilization at any time before the survey according to vaccination card, mother’s report, either vaccination card or mother’s reports were included. DHS selected the study participants by using two stages of the stratified sampling technique. In the first stage, Enumeration Areas (EAs) were randomly selected while in the second stage households were selected. In most DHS surveys the sample is selected with unequal probability to increase cases available for certain areas for which statistics are needed. We weighted the sample using the individual weight of women (v005) to produce the proper representation. Hence sample weights were generated by dividing (v005) by 1,000,000 and the total weighted sample of 15,090 children was used for the analysis (Fig. 1). Diagrammatic representation of sample selection in the study The source population for this study was children who lived in Sub-Saharan Africa. Our study population was children aged 24–35 months, who lived in Sub-Saharan Africa. The WHO recommended MCV1 to be given at 9 months of age, and MCV2 at age 15–18 months through routine services strategies [18]. The DHS data reported the percentage of children aged 12–23 months and 24–35 months who received MCV2. Therefore the age from 24 to 35 is the ideal age category for our study to get a complete record of MCV2 vaccination. We weighted (v005/1,000,000) our sample to correct over- and under-sampling and sound the findings. MCV2 vaccination status of children aged 24–35 months was our response variable. The outcome variable was binary and was coded as “1” if children received MCV2 and”0″ otherwise. Independent variables were considered at two levels (individual level and community level). Individual-level (level-I) variables were maternal age, maternal education, mother’s marital status, wealth index, media exposure, sex of the child, distance to health facilities, twin status, child size at birth, sex of household head, ANC visit, PNC visits, place of delivery. Community-level (level-II) variables such as community media exposure, community women’s education, community poverty, country, and place of residence were included. We did an aggregation of individual-level variables at the cluster level and categorized them as higher or lower based on median value to generate community-level variables except for residence and country (Table 2). Independent variables of MCV2 utilization among children aged 24–35 months in SSA A multilevel binary logistic regression model was fitted to identify significantly associated factors. Variables were extracted from each of the eight country’s KR files and STATA version 14.2 was used to clean, recode and analyze the data. Pooled data were generated by appending the extracted data from the 8 Sub-Saharan African countries and weighted to draw valid inferences. Four models were applied, comprising the null model (model 0) without any explanatory variables, to test the random effect of between-cluster variability and check the existence of variation (ICC) on random intercept, Model I with individual-level variables only, to assess the impact of individual-level variables on the outcome, Model II with community-level factors only assesses the impact of community-level factors on the outcome, and Model III with both individual-level and community-level variables fitted to reveal their net fixed and random effects on the outcome variable. Because the models were nested, we used deviance (− 2LLR) for model comparison. The intra-cluster Correlation Coefficient (ICC) was used to quantify the degree of heterogeneity of MCV2 between clusters. In addition, the Likelihood Ratio test (LR), Proportional Change in Variance (PCV), and Median Odds Ratio (MOR) were computed to measure the variation between clusters. Both community and individual-level variables with a p-value ≤ 0.2 in the bi-variable analysis were included in the multivariable model [32]. Adjusted OR (AOR) with 95% CI and p < 0.05 were applied to determine significantly associated factors. We used the variance inflation factor (VIF) test to check multicollinearity, and multicollinearity was not found because all variables have VIF < 5, and model III’s VIF was 1.49.
N/A