**4. Discussion**

This research reveals a zero-dose proportion of 91.7% (255/278), almost ten times the Cameroon national zero-dose proportion of 9.7% reported in 2020 [23]. This low vaccination coverage contributing to the low national EPI coverage could be explained partly by factors peculiar to its hard-to-reach characteristics. These factors include the absence of health facilities in the study health areas, distance from the lone health facility, multiple poorly accessible communities (islets), and frequent diurnal flooding, making access an uphill task. Our findings are consistent with a publication by Ozawa et al. in 2019 on the characteristics of hard-to-reach communities—based on an extensive literature review from 2000 to 2018 [24]. In our study, the primary service delivery approach employed in vaccinating children was mass vaccination campaigns—78.3% of vaccinated children were vaccinated through this approach. This highlights the importance of Supplementary Immunization Activities (SIA) and vaccination campaigns in improving vaccination coverage in hard-to-vaccinate communities, similar to the role of SIA in preventing measles outbreaks during the COVID-19 pandemic in Kenya [25].

Based on the multivariable logistic regression analysis, the log odds of being a zerodose child decreased with the child's age and being born in a health facility. However, children born to immigrant fathers and non-Christian mothers had higher odds of being zero-dose children than those born to Cameroonian fathers and Christian mothers, respectively. Younger children are likely to be unvaccinated compared to their older peers

(AOR: 0.90, 95% CI: 0.82–1.00, *p* = 0.0401). This can be explained by the fact that this population depends solely on outreach and mobile strategies for vaccination and most often have to wait for a national vaccination campaign or an interventional vaccination program during an epidemic to receive routine vaccines. By reviewing demographic and health surveys in sub-Saharan Africa, Mutua et al. showed that on-time vaccination was relatively low in sub-Saharan Africa and varied depending on different factors, including place of residence [1]. This implies younger children are likely to miss their vaccines and only get them at an older age. This is also consistent with a study by Stein-Zamir et al. in Israel, which showed that age-specific vaccine delays would lead to fewer vaccination cases at younger ages compared to older children [26]. A study in 2018 showed the impact of a mobile vaccination strategy in hard-to-reach communities, with children of older ages having higher vaccination coverages than those of younger age groups, similar to the findings in this study [27]. It is, therefore, of significant value to design tailored approaches that permit routine vaccination of children from birth to ensure all children benefit from vaccine protection throughout childhood.

Children born in health facilities were less likely to be unvaccinated than those delivered at home (AOR: 0.07, 95% CI: 0.02–0.30, *p* = 0.0003). Since women deliver at home, they miss the vaccines given to the child at birth, including vaccination-related counseling and scheduling, which can explain this finding. Also, 93.5% of children in this study did not have birth certificates, which presents a challenge in determining a child's age, posing a problem in terms of logistics, routine vaccination micro-planning, and the vaccination activity itself, as it relies on the ages of the children. This is consistent with other studies, though they did not focus on zero-dose cases, but were more interested in incomplete and complete vaccination cases [24,28–30]. In missed communities, a context-specific approach, such as setting up micro-health facilities or collaborating with traditional birth attendants to identify, track and vaccinate children from birth, will significantly improve immunization coverage and the fight against VPDs.

Children born to immigrant fathers were likely to be zero-dose children compared to children whose fathers were native Cameroonians (AOR: 2.60, 95% CI = 0.65–10.35, *p* = 0.0016). Most immigrants do not have a residence permit and as such, they cannot easily access essential health services outside their current residence. As a result, they tend to depend on traditional healers, birth attendants, roadside drug vendors, and unregistered private dispensaries for their healthcare needs. As such, even if parents are willing to vaccinate their children, they would have no choice but to wait for an outreach vaccination program since they cannot travel to get vaccines outside of this setting. A systemic review of studies in sub-Saharan Africa and the European region revealed migration as a factor associated with low vaccination coverage [31,32]. Also, comparatively lower vaccination coverage was found among immigrants in India compared to the locals because of the high prevalence of home births, lack of awareness of the location of health facilities, mobility, and fear of vaccine side effects [33].

In the same line, this study reveals that children born to minority non-Christian mothers are likelier to be zero-dose children than those born by Christian mothers (AOR: 6.55, 95% CI: 1.04–41.25, *p* = 0.0453). The non-Christian communities in MHD represent a minority population, with only 5.4% of mothers belonging to this population as opposed to their Christian counterparts, 93.8%. To leave no child unvaccinated, the finding in this study further emphasizes the need to identify minority communities; employ human-centered design and tools, such as the WHO framework of behavioral and social drivers (BeSD), to have in-depth knowledge on supply and demand barriers specific to minority populations; and develop context-specific strategies.

Unlike most studies, birth order was not significantly associated with zero-dose vaccination status, AOR, 1.33, 95% CI: 0.97–1.81, *p* = 0.0753 [28,34,35]. For instance, a nested case-control study conducted on a cohort of 110,902 Israeli children under the age of 5 revealed that birth order progression is inversely associated with vaccine utilization [36]. The critical explanation is that previous parental vaccination service delivery experiences

with their firstborns tend to shape parents' new attitudes towards vaccination [36]. Birth order was probably insignificant in this study because vaccination coverage was too low in these health areas to significantly impact subsequent parental attitudes toward vaccination.
