Vaccination Dropout Rates Among Children Aged 12-23 Months in the Urban-Rural Health District of Mont Ngafula II, Democratic Republic of the Congo: A Cross-sectional Study

Background: Globally, 1.8 million children failed to receive the 3-dose diphtheria, tetanus and pertussis (DTP) series every year in the Democratic Republic of the Congo (DRC). Currently, an emergency plan targeting 9 provinces, including the capital of Kinshasa, is launched to reinforce routine immunization. The Mont Ngafula II health district (HD) was the only one to report high vaccination dropout rates over almost ve consecutive years. This study aimed to assess issues within the local dynamic system of vaccination services and identify associated risk factors among children aged 12-23 months. Methods: A cross-sectional household survey was conducted among 418 children in June-July 2019 using a two-stage sampling design. Socio-demographic and perception data were collected through a structured interviewer-administered questionnaire. The distribution of 2017-2018 immunization coverage and dropout rate was extracted from local authority of HD and mapped. Chi-square test and multivariate logistic regression were used to identify predictors of high vaccination dropout. Statistical signicance was dened as p < 0.05. Results: Of 14 health areas included in the Mont Ngafula II HD, four reported high vaccine coverage, only one recorded low vaccine coverage and three registered both low vaccine coverage and high dropout rate. Multivariate logistic regression showed the predictors of immunization dropout as follows: no possession of immunization card (aOR = 2.49; 95% CI = 1.25-4.93; p = 0.010), living in rural areas (aOR = 1.53; 95% CI = 1.02-2.32; p = 0.042), unavailability of seating places (aOR = 7.10; 95% CI = 1.39-36.27; p = 0.019), no respect of the order of arrival (aOR = 3.88; 95% CI = 1.48-10.16; p = 0.006) during vaccination in health facilities, and the lack of a reminder system on days before the scheduled vaccination (aOR = 2.17; 95% CI = 1.43-3.32; p < 0.001). is required to develop appropriate strategies for improving routine immunization activities. The present study aimed to assess the local system of vaccination services at health area (HA) scale and identify determinants of high vaccination dropout rates among children aged 12–23 months in the Mont Ngafula II HD.

no respect of the order of arrival, and the lack of a reminder system were the predictors of vaccination dropout rates. These results advocate for prioritizing targeted interventions and programs that will strengthen interpersonal communication between vaccination service providers and users during vaccination in health facilities as well as implement and extend a reminder system using SMS reminders on days before the scheduled vaccination.

Background
Since the establishment of the Expanded Program on Immunization (EPI) in 1974, vaccines showed a signi cant impact on vaccine preventable diseases by reducing morbidity, disability and mortality worldwide among children (1,2). However, children remained not fully immunized with the vaccines of EPI targeted diseases in many parts of the world (3). Considered as a key indicator of immunization programme performance (4,5), the third dose of vaccine against diphtheria, tetanus and pertussis (DTP3) has not been administered to 19.9 million children worldwide in the rst year of life in 2017 (6). The largest numbers of children who did not receive DTP3, 12.4 million (62%), were concentrated in ten countries: Nigeria, India, Pakistan, Indonesia, Ethiopia, the Democratic Republic of the Congo (DRC), Angola, Iraq, South Africa and Afghanistan (6). Of all the children who did not complete the 3-dose DTP series, 6.2 million (31%) started but did not complete the DTP series (dropout rate) (6).
In the DRC, 1.8 million children failed to receive the 3-dose DTP series every year (7). All that despite improvements for DTP3 coverage at the national level from 25% in 1999 to 81% in 2018 with the partnership of the Global Alliance for Vaccines and Immunization (GAVI) (7). This is even more alarming given that Congolese EPI was targeting to vaccinate 93% of children with the 3 doses of DTP before their rst birthday (8). In addition, the DRC adopted the Reaching Every District (RED) approach since 2004. This approach was developed and introduced in 2002 by World Health Organization (WHO), the United Nations Children's Fund (UNICEF) and other partners to improve immunization systems in areas with low coverage (9). However, many children are still susceptible to vaccine-preventable diseases (VPDs) and death. Measles, yellow fever and poliomyelitis outbreaks were reported across the country over the last ten years (10)(11)(12)(13).
In 2018, the DRC launched an emergency plan called Mashako to reinforce routine immunization in order to save 220.000 children within 18 months. This plan targeted 9 provinces: Ituri, Kasaï, Haut-Katanga, Mongala, Kwilu, Tanganyika, Tshuapa, Haut-Lomami and the capital of Kinshasa. In the latter that counts 35 health districts (HDs), Mont Ngafula II was the only to record high dropout rates among children aged < 12 months over almost ve consecutive years. The EPI's routine reports showed dropout rates for DTP3 dose of 15%, 19%, 10% and 11% in 2014, 2016, 2017 and 2018 respectively (Dataset V1). It is required to develop appropriate strategies for improving routine immunization activities.
The present study aimed to assess the local system of vaccination services at health area (HA) scale and identify determinants of high vaccination dropout rates among children aged 12-23 months in the Mont Ngafula II HD.  (Fig. 1). The HD has a general pediatric hospital, 107 health centers including 22 health posts. Only 14 health centers deliver EPI vaccines to the target population groups. The main local economic activities are based on agriculture, livestock and shery products.

Study area
Study design and sampling procedure A cross-sectional household survey was conducted. All children aged 12-23 months, between the 1st of June and the 1st of July 2019, were eligible. We performed a two-stage sampling design according to the WHO (14). Based on the estimated population data per village/neighborhood provided by the HD's authority, 30 clusters were rstly selected through probability proportional to their size using simple random method. The sample size was determined using the following formula: N = Z 2 × p(1-p)/d 2 , where N is the sample size, Z (1.96) is the 2-sided level of signi cant, p is the immunization coverage of the country (45%) as proposed in the study of Baguune at al. (2017) (15) and d is the desired width of the con dence interval. Given a non-response rate of 10%, we found a total sample size of 418 children aged 12-23 months. At the second stage, we used the EPI random walk method (16) to select 14 children from each of the 30 clusters. The starting point was considered as the rst household chosen randomly from the geographic centre of each cluster using the spinning bottle method and then continued to the next nearest household in the same direction until 14 eligible children were obtained.

Data collection
A structured questionnaire translated into local language was used for the data collection. Interviewers were previously recruited and trained on the study protocol and the questionnaire use. After the training, a pre-test was conducted on the eld to check the reliability and validity of our study tools. The questionnaire included information on socio-demographic characteristics, child immunization status, and perception of respondents (mothers/caregivers) on the organization of vaccinating health facilities. Doorto-door visits were conducted and only consenting respondents of selected children participated in the structured interviewer-administered questionnaire. In case of two or more children aged 12-23 months were found in the same household, the youngest child was selected (16). Information on child vaccination such as dates and doses of each vaccine received was obtained from vaccination cards or respondents' recall if the card was unavailable. Data on routine immunization coverage and dropout rate at HA scale was also extracted from 2017-2018 annual datasets of HD (Dataset V2). According to Congolese routine immunization schedule summarized in Table 1, dropout rate was considered as the percentage of children aged 12-23 months who received the 1st but not the 3rd dose of DTP. It was calculated according to WHO method: [( rst dose -last dose)/ rst dose] × 100 (4). Collected data were recorded on Microsoft Excel® 2013 les. All statistical analyses were performed in Statistical Package for Social Sciences (SPSS) version 20.0. Descriptive statistics such as frequencies and percentages were calculated and presented into tables. Children, respondents, household characteristics and respondents' perception differences associated with immunization status (cards and respondents' recall) were tested by the Chi-square test. Adjusted Odds Ratios (aOR) with their 95% con dence intervals (95% CI) were calculated using a multivariate logistic regression model to measure the strength of association between independent predictors and immunization dropout. The threshold of statistical signi cance was de ned as p < 0.05.

Results
A total of 418 children and their mother/caregivers were assessed in the study. Overall, 7.7% of the respondents were ≤ 19 years and 0.7% were ≥ 50 years of age. All respondents had a formal education while 69.9% up to secondary level and 12.4% up to tertiary level. More than half (54.5%) of the respondents were unemployed. Nearly, two-third (63.6%) of the respondents lived in urban settings and 92.1% had less than 3 children in their households. The majority of children (90.0%) possessed vaccination card, 50.2% were female and 53.6% were the 3rd or later born in the family (Table 2).  Table 3 shows the respondents' perceptions on the organization of vaccinating health facilities of the Mont Ngafula II HD. The large majority indicated a good attitude of health staff during vaccination (94.0%), the availability of seating (95.0%), the respect of the order of arrival during vaccination (90.0%), and a long waiting time during vaccination (83.0%). More than half (58.9%) of the participants reported the lack of a reminder system on days before the scheduled vaccination. Spatial distribution of routine immunization coverage and dropout rate Sociodemographic characteristics and perceptions of participants associated with dropout status The analysis of sociodemographic characteristics and perceptions of participants with dropout status based on vaccination cards and respondents' recall is shown in Table 4. Children with vaccination cards had a lower dropout rate compared to those without it (p = 0.008). There were no statistical differences in other sociodemographic characteristics. Children of respondents who reported positive attitude of health staff (p = 0.005), availability of seating places (p < 0.001) and respect of the order of arrival (p < 0.001) during vaccination in health facilities were more likely fully immunized. The lack of a reminder system on days before the scheduled vaccination was found associated with higher dropout (p < 0.001). Predictors of vaccination dropout rates  More than two-fth of them also registered low vaccine coverage while only one HA reported exclusively low vaccine coverage.
Concerning the predictors, residing in rural areas and the absence of immunization card, no seating places and no respect of the order of arrival during vaccination in health facilities as well as the lack of a reminder system on days before the scheduled vaccination were signi cantly associated with high dropout rates among children aged 12-23 months.
Concerning the local system of vaccination services, the poor utilization of vaccine services seemed as the main health concern followed by the poor access at HA scale. Among the HAs that recorded high dropout rates, more than half reported good access but poor utilization of vaccination services (high vaccine coverage and high dropout) while two-fth registered both poor access and poor utilization vaccination services (low vaccine coverage and high dropout). Furthermore, only one HA reported both poor access and high utilization vaccination services (low vaccine coverage and low dropout). Thus, these results may represent a useful indicator for the risk assessment of vaccination services in the study area.
In the present study, children without immunization card had higher susceptibility to drop out from vaccination services. Not having an immunization card was also found as a predictor of incomplete immunization in other studies from low-income African countries (15,17,18). This situation may be related to perception of attitudes of the health staff. As suggested by Baguune et al. (2017), in case of loss of immunization cards, negative attitudes from some health workers unable to provide new vaccination cards induce a reluctance feeling of mothers for new vaccination in health facilities (15).
This evidence is also supported by the results from a cross-sectional survey conducted during a polio outbreak in Cameroon in which good perception of immunization services was considered as a determinant of vaccination completeness (19). Thus, we hypothesize that a good reception of health staff can strengthen interpersonal communication and be a major asset for the subsequent use of the immunization services.
This study showed that children residing in rural areas were more likely to drop out compared to their urban counterparts. However, discordant results have been reported in other studies, especially conducted in Asia and Africa. At the contextual level, living in rural areas was associated with breaks in childhood vaccination in India (20) while this characteristic was found protective in a multilevel analysis involving 24 African countries (21). As suggested previously, discrepancies observed would call into question the urban advantage probably due to urban demographic growth and related di culties of access to health services (20,(22)(23)(24).
Respondents perceived that unavailability of seating places and no respect of the order of arrival during vaccination in health facilities were associated with immunization dropout rate. In addition, the lack of a reminder system on days before the scheduled vaccination was also perceived as a predictor to high dropout. This result is similar to studies conducted by Bangure et al. (2015) in Zimbabwe (25), Haji et al.
(2016) and Gibson et al. (2017) in Kenya (26,27). They found that complete immunization was signi cantly higher among children of mothers/caregivers who received short message service (SMS) reminders than those in control groups. Thus, our ndings support the hypothesis that implementing and extending a reminder system using SMS-type messages may contribute to improve substantially immunization services in our context.
Other socio-demographic characteristics considered in the present study were not statistically associated with dropout rate while they were found as predictors of incomplete vaccination. Children (birth order, sex), parents (education level, religion), and household (number of children) characteristics were signi cantly related to the children immunization status in several African studies (15,19,26,(28)(29)(30)(31)(32)(33)(34).
Growing number of children, sex discrimination, low parents' education level, and misconceptions from certain religious groups affecting vaccination uptake may lead to immunization dropout (15,19,26). The discrepancies observed on these characteristics in our study represent the rst limit. They could be explained by our small sample size and some speci cs characterizing the objectives assigned and the sample procedures comparatively to other studies.
A second study limitation is that determinants of vaccination dropout rates were assessed from the perspective of the users of vaccination services. Evaluation from the perspective of health systems and providers was not considered. However, determinants of childhood immunization coverage are complex and interrelated. Vaccination depends on a real need for vaccines and health seeking service from users and high-quality service of vaccines offer under the optimal technical, logistical, and operational conditions from providers (35).
Nonetheless, there has been a limited number of research in DRC concerning vaccination dropout (36). To our knowledge, the present study is the rst conducted at both ne spatial scale and individual level to understand the main health concern of vaccine services and identify factors related to low vaccination completeness among children aged 12-23 months. An additional strength of this survey is that the result of this study can be extrapolated to the population of the DRC because the survey used a probability sampling technique and the minimum sample size was computed taking into account the formula required to allow each person to be part of the sample.

Conclusion
The poor utilization of immunization services seemed as the main health concern followed by the poor access. Residing in rural areas, no immunization card possession, absence of seating places and no respect of the order of arrival during vaccination in health facilities, and the lack of a reminder system on days before the scheduled vaccination were the predictors of high dropout rates among children aged 12-23 months. These results will serve as a useful tool to prioritize targeted vaccination interventions and programs that will strengthen interpersonal communication between vaccination service providers and users during vaccination in health facilities. In addition, the latter nding advocates for implementing and extending a reminder system using SMS reminders on days before the scheduled vaccination.
Adopting such an approach would be even more bene cial for rapid immunization in ghting child morbidity and mortality, especially since it has been shown that the cost is affordable (25,37,38).
Further studies including determinants attributable to health workers and immunization services will help to thoroughly explore factors associated with routine vaccination dropout rates among children aged 12-23 months. Kinshasa)". After being informed, the respondents' consent was obtained verbally. Data collection was conducted con dentially.

Consent for publication
Our manuscript does not contain any individual person's data. Study area. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 2
Spatial distribution of 2017-2018 routine immunization coverage and dropout rate. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.