MRC Centre for Global Infectious Disease Analysis, School of Public Health, Imperial College London, London, United Kingdom.
Center for Disease Analysis Foundation, Lafayette, Colorado, United States of America.
PLoS One. 2020 Aug 10;15(8):e0237525. doi: 10.1371/journal.pone.0237525. eCollection 2020.
Hepatitis B is a global epidemic that requires carefully orchestrated vaccination initiatives in geographical regions of medium to high endemicity to reach the World Health Organization's elimination targets by 2030. This study compares two widely-used deterministic hepatitis B models-the Imperial HBV model and the CDA Foundation's PRoGReSs-based on their predicted outcomes in four countries. The impact of scaling up of the timely birth dose of the hepatitis B vaccine is also investigated. The two models predicted largely similar outcomes for the impact of vaccination programmes on the projected numbers of new cases and deaths under high levels of the infant hepatitis B vaccine series. However, scenarios for the scaling up of the infant hepatitis B vaccine series had a larger impact in the PRoGReSs model than in the Imperial model due to the infant vaccine series directly leading to the reduction of perinatal transmission in the PRoGReSs model, but not in the Imperial model. Meanwhile, scaling up of the timely birth dose vaccine had a greater impact on the outcomes of the Imperial hepatitis B model than in the PRoGReSs model due to the greater protection that the birth dose vaccine confers to infants in the Imperial model compared to the PRoGReSs model. These differences underlie the differences in projections made by the models under some circumstances. Both sets of assumptions are consistent with available data and reveal a structural uncertainty that was not apparent in either model in isolation. Those relying on projections from models should consider outputs from both models and this analysis provides further evidence of the benefits of systematic model comparison for enhancing modelling analyses.
乙型肝炎是一种全球性的流行疾病，需要在中高度流行地区精心策划疫苗接种计划，以实现世界卫生组织到 2030 年消除乙型肝炎的目标。本研究比较了两种广泛使用的确定性乙型肝炎模型——帝国 HBV 模型和 CDA 基金会的 PRoGReSs 模型——根据它们在四个国家的预测结果。还研究了扩大乙型肝炎疫苗及时出生剂量的影响。这两种模型对疫苗接种计划对新发病例和死亡人数的预测结果基本相似，假设婴儿乙型肝炎疫苗系列达到高水平。然而，由于婴儿疫苗系列直接导致 PRoGReSs 模型中围产期传播的减少，而在帝国模型中则没有，因此扩大婴儿乙型肝炎疫苗系列的方案在 PRoGReSs 模型中的影响大于在帝国模型中的影响。同时，由于在帝国模型中，出生时疫苗对婴儿的保护作用大于 PRoGReSs 模型，因此扩大及时出生剂量疫苗对帝国乙型肝炎模型的结果影响大于 PRoGReSs 模型。这些差异是模型在某些情况下做出预测的差异的基础。这两套假设都与现有数据一致，并揭示了一种结构上的不确定性，这种不确定性在单独的模型中并不明显。那些依赖模型预测的人应该考虑两个模型的输出，本分析进一步证明了系统模型比较对于增强模型分析的好处。
