Burgers Laura T, Redekop William K, Severens Johan L
Institute of Health Policy and Management, Erasmus University Rotterdam, Rotterdam, The Netherlands,
Pharmacoeconomics. 2014 Jul;32(7):627-37. doi: 10.1007/s40273-014-0155-9.
Decision analytic modelling is essential in performing cost-effectiveness analyses (CEAs) of interventions in cardiovascular disease (CVD). However, modelling inherently poses challenges that need to be dealt with since models always represent a simplification of reality. The aim of this study was to identify and explore the challenges in modelling CVD interventions.
A document analysis was performed of 40 model-based CEAs of CVD interventions published in high-impact journals. We analysed the systematically selected papers to identify challenges per type of intervention (test, non-drug, drug, disease management programme, and public health intervention), and a questionnaire was sent to the corresponding authors to obtain a more thorough overview. Ideas for possible solutions for the challenges were based on the papers, responses, modelling guidelines, and other sources.
The systematic literature search identified 1,720 potentially relevant articles. Forty authors were identified after screening the most recent 294 papers. Besides the challenge of lack of data, the challenges encountered in the review suggest that it was difficult to obtain a sufficiently valid and accurate cost-effectiveness estimate, mainly due to lack of data or extrapolating from intermediate outcomes. Despite the low response rate of the questionnaire, it confirmed our results.
This combination of a review and a survey showed examples of CVD modelling challenges found in studies published in high-impact journals. Modelling guidelines do not provide sufficient guidance in resolving all challenges. Some of the reported challenges are specific to the type of intervention and disease, while some are independent of intervention and disease.
决策分析模型对于开展心血管疾病（CVD）干预措施的成本效益分析（CEA）至关重要。然而，建模本身带来了一些需要应对的挑战，因为模型始终是对现实的简化。本研究的目的是识别和探讨CVD干预措施建模中的挑战。
对发表在高影响力期刊上的40篇基于模型的CVD干预措施CEA进行了文献分析。我们系统地分析了所选论文，以确定每种干预类型（检测、非药物、药物、疾病管理计划和公共卫生干预）的挑战，并向通讯作者发送了一份问卷，以获得更全面的概述。基于论文、回复、建模指南和其他来源，提出了应对这些挑战的可能解决方案。
系统文献检索确定了1720篇潜在相关文章。在筛选了最近的294篇论文后，确定了40位作者。除了数据缺乏的挑战外，综述中遇到的挑战表明，主要由于缺乏数据或从中期结果进行推断，很难获得足够有效和准确的成本效益估计。尽管问卷回复率较低，但证实了我们的结果。
这项综述和调查相结合的研究展示了在高影响力期刊发表的研究中发现的CVD建模挑战的实例。建模指南在解决所有挑战方面没有提供足够的指导。一些报告的挑战特定于干预类型和疾病，而一些则与干预和疾病无关。
