Department of Accounting and Finance, Kingston University, London, UK.
Med Care. 2012 Dec;50(12):1076-85. doi: 10.1097/MLR.0b013e318269e06d.
The development of human papillomavirus (HPV)-related diseases is not understood perfectly and uncertainties associated with commonly utilized probabilistic models must be considered. The study assessed the cost-effectiveness of a quadrivalent-based multicohort HPV vaccination strategy within a Bayesian framework.
A full Bayesian multicohort Markov model was used, in which all unknown quantities were associated with suitable probability distributions reflecting the state of currently available knowledge. These distributions were informed by observed data or expert opinion. The model cycle lasted 1 year, whereas the follow-up time horizon was 90 years. Precancerous cervical lesions, cervical cancers, and anogenital warts were considered as outcomes.
The base case scenario (2 cohorts of girls aged 12 and 15 y) and other multicohort vaccination strategies (additional cohorts aged 18 and 25 y) were cost-effective, with a discounted cost per quality-adjusted life-year gained that corresponded to €12,013, €13,232, and €15,890 for vaccination programs based on 2, 3, and 4 cohorts, respectively. With multicohort vaccination strategies, the reduction in the number of HPV-related events occurred earlier (range, 3.8-6.4 y) when compared with a single cohort. The analysis of the expected value of information showed that the results of the model were subject to limited uncertainty (cost per patient = €12.6).
This methodological approach is designed to incorporate the uncertainty associated with HPV vaccination. Modeling the cost-effectiveness of a multicohort vaccination program with Bayesian statistics confirmed the value for money of quadrivalent-based HPV vaccination. The expected value of information gave the most appropriate and feasible representation of the true value of this program.
人乳头瘤病毒（HPV）相关疾病的发展尚不完全清楚，必须考虑到常用概率模型相关的不确定性。本研究在贝叶斯框架内评估了基于四价疫苗的多队列 HPV 疫苗接种策略的成本效益。
使用完全贝叶斯多队列马尔可夫模型，其中所有未知量都与合适的概率分布相关联，这些分布反映了当前知识的状态。这些分布的信息来源于观察数据或专家意见。模型周期为 1 年，随访时间为 90 年。癌前宫颈病变、宫颈癌和生殖器疣被视为结局。
基础情况（两批 12 岁和 15 岁的女孩）和其他多队列疫苗接种策略（增加 18 岁和 25 岁的队列）具有成本效益，每获得一个质量调整生命年的贴现成本分别为 12013 欧元、13232 欧元和 15890 欧元，对应的疫苗接种计划分别基于 2、3 和 4 个队列。与单队列疫苗接种策略相比，多队列疫苗接种策略更早地减少了 HPV 相关事件的数量（范围为 3.8-6.4 年）。信息期望价值的分析表明，模型的结果存在有限的不确定性（每位患者的成本为 12.6 欧元）。
这种方法旨在纳入 HPV 疫苗接种相关的不确定性。使用贝叶斯统计学对多队列疫苗接种计划的成本效益进行建模，证实了基于四价疫苗的 HPV 疫苗接种具有良好的性价比。信息期望价值为该计划的真实价值提供了最合适和最可行的表示。
