Bayesian Model Averaging to Account for Model Uncertainty in Estimates of a Vaccine's Effectiveness.
Bayesian Model Averaging to Account for Model Uncertainty in Estimates of a Vaccine's Effectiveness.
复制标题
DOI：
10.2147/clep.s378039
复制
发表时间：
2022
影响因子：
3.9
通讯作者：
中科院分区：
文献类型：
--
作者：
疫苗有效性（VE）研究通常在引入新疫苗后进行，以确保它们在现实环境中提供保护。在分析过程中经常需要控制混杂因素，而多变量建模是最有效的方法。当考虑到许多混杂因素时，知道哪些变量需要包含在最终模型中可能是具有挑战性的。我们提出了一个直观的贝叶斯模型平均（BMA）框架。数据来自一项匹配的病例对照研究，旨在评估莱姆病疫苗获得许可后的有效性。病例为康涅狄格州居民，年龄15-70岁，确诊患有莱姆病。每个病例受试者按年龄匹配最多2名健康对照。对所有参与者进行了访谈，并审查了医疗记录，以确定免疫史并评估潜在的混杂因素。利用BMA系统地搜索潜在模型，并从模型的顶部子集计算加权平均VE估计。对比了三种传统的单最佳模型选择方法（两阶段选择、逐步淘汰和跳跃算法）的性能。分析包括358例病例和554例匹配的对照。VE的范围在56%到73%之间，95%的置信区间在<5%的候选模型中过零。前15款车型的平均BMA VE为69% （95% CI: 18-88%）。两阶段、逐步和跳跃式算法的VE分别为71% （95% CI: 21-90%）、73% （95% CI: 26-90%）和74% （95% CI: 27-91%）。本文强调了如何使用BMA框架来生成透明和稳健的VE估计。bma导出的VE和置信区间与使用传统方法估计的相似。然而，通过将模型不确定性纳入参数估计，BMA可以为设计良好的研究提供额外的严谨性和可信度。
Vaccine effectiveness (VE) studies are often conducted after the introduction of new vaccines to ensure they provide protection in real-world settings. Control of confounding is often needed during the analyses, which is most efficiently done through multivariable modeling. When many confounders are being considered, it can be challenging to know which variables need to be included in the final model. We propose an intuitive Bayesian model averaging (BMA) framework for this task. Data were used from a matched case–control study that aimed to assess the effectiveness of the Lyme vaccine post-licensure. Cases were residents of Connecticut, 15–70 years of age with confirmed Lyme disease. Up to 2 healthy controls were matched to each case subject by age. All participants were interviewed, and medical records were reviewed to ascertain immunization history and evaluate potential confounders. BMA was used to systematically search for potential models and calculate the weighted average VE estimate from the top subset of models. The performance of BMA was compared to three traditional single-best-model-selection methods: two-stage selection, stepwise elimination, and the leaps and bounds algorithm. The analysis included 358 cases and 554 matched controls. VE ranged between 56% and 73% and 95% confidence intervals crossed zero in <5% of all candidate models. Averaging across the top 15 models, the BMA VE was 69% (95% CI: 18–88%). The two-stage, stepwise, and leaps and bounds algorithm yielded VE of 71% (95% CI: 21–90%), 73% (95% CI: 26–90%), and 74% (95% CI: 27–91%), respectively. This paper highlights how the BMA framework can be used to generate transparent and robust estimates of VE. The BMA-derived VE and confidence intervals were similar to those estimated using traditional methods. However, by incorporating model uncertainty into the parameter estimation, BMA can lend additional rigor and credibility to a well-designed study.
影响因子：
4.2
作者：
Oliveira CR;Massad C;Shapiro ED;Vazquez M
通讯作者：
Vazquez M