Health economic models simulate the costs and effects of health technologies for use in health technology assessment (HTA) to inform efficient use of scarce resources. Models have historically been developed using spreadsheet software (e.g., Microsoft Excel) and while use of R is growing, general purpose modeling software is still limited. hesim helps fill this gap by facilitating parameterization, simulation, and analysis of economic models in an integrated manner. Supported model types include cohort discrete time state transition models (cDTSTMs), individual continuous time state transition models (iCTSTMs), and partitioned survival models (PSMs), encompassing Markov (time-homogeneous and time-inhomogeneous) and semi-Markov processes. A modular design based on R6 and S3 classes allows users to combine separate submodels for disease progression, costs, and utility in a flexible way. Probabilistic sensitivity analysis (PSA) is used to propagate uncertainty in model parameters to model outputs. Simulation code is written in C++ so complex simulations such as those combining PSA and individual simulation can be run much more quickly than previously possible. Decision analysis within a cost-effectiveness framework is performed using simulated costs and quality-adjusted life years (QALYs) from a PSA.
翻译:健康经济模型模拟保健技术的成本和影响,供卫生技术评估使用,为有效使用稀缺资源提供信息; 模型历来使用电子表格软件(例如微软Excel)开发,而R的使用正在增加,通用模型软件仍然有限; 健康经济模型通过综合促进经济模型的参数化、模拟和分析,帮助填补这一空白; 支持的模型类型包括组群离散时间状态过渡模型(cDTSTMS)、个别连续时间国家过渡模型(ICTSTMS)和分离的生存模型(PSMS),其中包括Markov(时间-同源和时间-不相容)和半马尔科夫进程; 以R6和S3等级为基础的模块设计允许用户以灵活的方式将疾病演变、成本和效用的子模型结合起来。 概率敏感性分析(PSA)用于在模型参数中传播不确定性; 模拟代码以C++的形式编写,例如将PSA与个人模拟相结合的复杂模拟可以比以往更快地运行。 使用成本-质量框架进行决策分析。