Utazi C Edson, Pannell Oliver, Aheto Justice M K, Wigley Adelle, Tejedor-Garavito Natalia, Wunderlich Josh, Hagedorn Brittany, Hogan Dan, Tatem Andrew J
WorldPop, School of Geography and Environmental Science, University of Southampton, Southampton, United Kingdom.
Flowminder Foundation and WorldPop, School of Geography and Environmental Science, University of Southampton, Southampton, United Kingdom.
PLOS Glob Public Health. 2022 Apr 27;2(4):e0000244. doi: 10.1371/journal.pgph.0000244. eCollection 2022.
Achieving equity in vaccination coverage has been a critical priority within the global health community. Despite increased efforts recently, certain populations still have a high proportion of un- and under-vaccinated children in many low- and middle-income countries (LMICs). These populations are often assumed to reside in remote-rural areas, urban slums and conflict-affected areas. Here, we investigate the effects of these key community-level factors, alongside a wide range of other individual, household and community level factors, on vaccination coverage. Using geospatial datasets, including cross-sectional data from the most recent Demographic and Health Surveys conducted between 2008 and 2018 in nine LMICs, we fitted Bayesian multi-level binary logistic regression models to determine key community-level and other factors significantly associated with non- and under-vaccination. We analyzed the odds of receipt of the first doses of diphtheria-tetanus-pertussis (DTP1) vaccine and measles-containing vaccine (MCV1), and receipt of all three recommended DTP doses (DTP3) independently, in children aged 12-23 months. In bivariate analyses, we found that remoteness increased the odds of non- and under-vaccination in nearly all the study countries. We also found evidence that living in conflict and urban slum areas reduced the odds of vaccination, but not in most cases as expected. However, the odds of vaccination were more likely to be lower in urban slums than formal urban areas. Our multivariate analyses revealed that the key community variables-remoteness, conflict and urban slum-were sometimes associated with non- and under-vaccination, but they were not frequently predictors of these outcomes after controlling for other factors. Individual and household factors such as maternal utilization of health services, maternal education and ethnicity, were more common predictors of vaccination. Reaching the Immunisation Agenda 2030 target of reducing the number of zero-dose children by 50% by 2030 will require country tailored analyses and strategies to identify and reach missed communities with reliable immunisation services.
在全球卫生界,实现疫苗接种覆盖率的公平性一直是一项关键优先事项。尽管最近做出了更多努力,但在许多低收入和中等收入国家(LMICs),某些人群中未接种和未充分接种疫苗的儿童比例仍然很高。这些人群通常被认为居住在偏远农村地区、城市贫民窟和受冲突影响地区。在此,我们研究这些关键社区层面因素以及一系列其他个人、家庭和社区层面因素对疫苗接种覆盖率的影响。利用地理空间数据集,包括2008年至2018年期间在9个低收入和中等收入国家进行的最新人口与健康调查的横断面数据,我们拟合了贝叶斯多层次二元逻辑回归模型,以确定与未接种和未充分接种显著相关的关键社区层面及其他因素。我们独立分析了12至23个月大儿童接种第一剂白喉-破伤风-百日咳(DTP1)疫苗和含麻疹疫苗(MCV1)以及接种所有三剂推荐的DTP疫苗(DTP3)的几率。在双变量分析中,我们发现,在几乎所有研究国家,偏远地区都会增加未接种和未充分接种疫苗的几率。我们还发现,生活在冲突地区和城市贫民窟会降低接种疫苗的几率,但在大多数情况下并非如预期那样。然而,城市贫民窟接种疫苗的几率比正规城市地区更有可能更低。我们的多变量分析表明,关键社区变量——偏远地区、冲突和城市贫民窟——有时与未接种和未充分接种有关,但在控制其他因素后,它们并不经常是这些结果的预测因素。个人和家庭因素,如母亲对卫生服务的利用、母亲的教育程度和种族,是更常见的疫苗接种预测因素。要实现《2030年免疫议程》的目标,即到2030年将零剂量儿童数量减少50%,将需要根据各国情况进行分析并制定战略,以识别并为错过接种机会的社区提供可靠的免疫服务。