Anton S F, Revicki D A
Bristol-Myers Squibb Company, Pharmaceutical Research Institute, Wallingford, CT 06492-7660, USA.
Psychopharmacol Bull. 1995;31(2):249-58.
A decision model can be a useful tool for evaluating the cost-effectiveness of a drug during research and development, before the drug is marketed. Decision analysis provides a structured process for comparing the costs and consequences of a new drug with those of standard drug therapy. This article introduces the use of a decision model and discusses the advantages and disadvantages of such a model. A cost-utility study of nefazodone, a new antidepressant, illustrates the application and interpretation of decision models. Decision analyses using data from clinical trials are a great potential source of information on the economic impact of new drugs. However, the development and use of such models require tolerance of uncertainty, the ability to represent complex relationships accurately, awareness of all factors that might influence the results, and a willingness to validate the model to the extent that validation is possible. The disadvantage of clinical decision models is that the findings are only as reliable as the information on which the model is based. Nevertheless, sensitivity analyses can be used to vary any information in question and assess the impact of that variation on the findings of the analyses. The advantage of clinical decision models is that they encourage the consideration and explicit representation of all pertinent inputs and outcomes. They clearly differentiate knowledge supported by data from assumptions, and compel assessment of the effect of those assumptions on the findings. The cost-effectiveness of nefazodone relative to standard treatment for major depression was evaluated using a decision model that simulated the lifetime direct medical costs and health outcomes associated with nefazodone and two comparators, imipramine and fluoxetine. The model included the possible clinical management pathways and the use of medical resources in treating major depression from early adulthood until death. Sensitivity analyses were performed to determine the impact of variation in model parameters on the results of the model. Using the base case assumptions of the model, which included parity of pricing with fluoxetine, nefazodone was found to be a cost-effective treatment for major depression.
在药物上市前的研发阶段，决策模型可作为评估药物成本效益的有用工具。决策分析提供了一个结构化的过程，用于比较新药与标准药物治疗的成本和后果。本文介绍了决策模型的应用，并讨论了该模型的优缺点。对新型抗抑郁药奈法唑酮的成本效用研究，阐明了决策模型的应用和解读。利用临床试验数据进行的决策分析，是获取新药经济影响信息的巨大潜在来源。然而，此类模型的开发和使用需要容忍不确定性、准确呈现复杂关系的能力、意识到所有可能影响结果的因素，以及在可能的范围内愿意对模型进行验证。临床决策模型的缺点是，其结果的可靠性仅与模型所基于的信息相同。尽管如此，敏感性分析可用于改变任何有疑问的信息，并评估该变化对分析结果的影响。临床决策模型的优点是，它们鼓励考虑并明确呈现所有相关的投入和结果。它们能清楚地区分由数据支持的知识与假设，并促使评估这些假设对结果的影响。使用一个决策模型评估了奈法唑酮相对于重度抑郁症标准治疗的成本效益，该模型模拟了与奈法唑酮以及两种对照药物丙咪嗪和氟西汀相关的终身直接医疗成本和健康结果。该模型纳入了从成年早期直至死亡治疗重度抑郁症可能的临床管理途径和医疗资源使用情况。进行了敏感性分析，以确定模型参数变化对模型结果的影响。根据该模型的基础案例假设，包括与氟西汀价格相当，发现奈法唑酮是治疗重度抑郁症的一种具有成本效益的疗法。
