Incentivizing Value in Managed Care Plans: Impact of a Value-Based Formulary

描述（由申请人提供）：大多数美国雇主赞助的计划使用处方药的共同支付等级来控制医疗保健支出。这些共同支付等级通常是根据药物的成本而不是药物提供的价值来确定的。然而，根据成本而不是价值设定共同支付可能导致药物使用效率低下。基于价值的处方集（VBF）是设计处方药效益的一项发展，其中使用成本效益分析（CEA）估计每种药物的价值。明确使用CEA来指导共同付费层的设置不同于以前的基于价值的保险设计（VBIDs），后者统一减少了治疗区域内所有药物的共同付费，指定治疗区域具有高价值，即使治疗区域内并非所有药物具有相同的价值。相反，VBF使用CEA来估计每种药物的价值。这样，共同费用就可以根据单个药物的价值而不是治疗类别来分配。此外，该方法可以定义低价值药物（以前的VBIDs只定义高价值治疗领域），并增加相应的共付额。虽然有希望，但VBF的影响无法预测，需要实证评估。首先，CEA估计是从广泛的来源得出的，这种估计的有效性是不确定的。其次，基于人口平均价值估计的共同支付变化的影响可能对药物的边际选择产生不可预测的影响。2010年，太平洋西北地区的一家大型非营利健康计划——普雷梅拉蓝十字（Premera Blue Cross）实施了VBF。我们利用这个自然实验来评估由商业健康保险计划实施的VBF的影响。此外，在VBF的所有药物的共同支付的变化提供了一个独特的机会，探索需求的交叉价格弹性的个人与多种慢性疾病。近三分之一的美国人患有两种或两种以上的慢性病，这些人占全国卫生支出的71%左右。然而，用于治疗许多慢性病的药物的依从率并不理想，在35%- 72%之间。对交叉价格弹性的评估将补充我们对这些个体自身价格弹性的理解。本研究将采用中断时间序列设计与同步对照组，以检验以下3个目的：(1)评估VBF对药物使用的影响；(2)评估VBF对成员自付费用和总医疗费用的影响；(3)评估多重慢性疾病个体需求的交叉价格弹性。为了实现这些目标，我们建议使用以下3种分段回归模型的方法：(1)分位数回归的两部分模型(2)扩展估计方程(3)选择性特定条件逻辑回归。

DESCRIPTION (provided by applicant): A majority of US employer-sponsored plans use copay tiers for prescription drugs to control healthcare expenditures. These copay tiers are generally determined based on the cost of the drugs rather than the value provided by the drugs. However, setting copays based on cost and not value can lead to inefficient use of drugs. A value-based formulary (VBF) is a development in designing prescription drug benefits in which the value of each medication is estimated using cost-effectiveness analysis (CEA). The explicit use of CEA to guide copay tier placement differs from previous value-based insurance designs (VBIDs) which have uniformly reduced copays for all drugs within a therapeutic area, designating the therapeutic area to be of high value, even though not all medications within a therapeutic area have the same value. In contrast, the VBF uses CEA to estimate the value of each individual drug. Copays can then be assigned based on individual drug value rather than by therapeutic class. Furthermore, this method can define low value drugs (previous VBIDs have only defined high value therapeutic areas) and increase their corresponding copays. While promising, the impact of the VBF cannot be predicted and requires empirical evaluation. First, CEA estimates were drawn from a wide range of sources and the validity of such estimates is uncertain. Second, the effect of copay changes based on population average estimates of value can have unpredictable effects on the marginal choice of medication. In 2010, Premera Blue Cross, a large non-profit health plan in the Pacific Northwest, implemented a VBF. We exploit this natural experiment to evaluate the impact of a VBF as implemented by a commercial health insurance plan. Furthermore, the changes in copay for all drugs in the VBF provide a unique opportunity to explore cross- price elasticities of demand among individuals with multiple chronic conditions. Almost a third of Americans have 2 or more chronic conditions and these individuals account for about 71% of national health expenditures. Yet, adherence rates for medications used to treat many chronic conditions are suboptimal-ranging between 35%- 72%. Evaluation of cross-price elasticities would complement our understanding of own-price elasticities for these individuals. This study will utilize an interrupted time series design with concurrent control grou in order to examine the following 3 aims: (1) assess the impact of the VBF on medication utilization (2) assess the impact of the VBF on both member out of pocket costs and total healthcare costs and (3) estimate cross-price elasticities of demand for individuals with multiple chronic conditions. In order to accomplish these aims, we propose to use the following 3 methods with segmented regression models: (1) two-part model with quantile regression (2) extended estimating equations (3) alternative-specific conditional logistic regression.