卫生体系的复杂性决定了卫生技术在现实环境中的效果评估需借助于数学模型。卫生决策常用的Markov状态转移模型因其无后效性假设在慢性病疾病进展过程中不成立而广受诟病；基于个体水平的Markov模型也存在不能有效地记录慢性疾病患者的疾病自然史中各种随时间变化的患者信息，不能考虑卫生服务资源利用等缺陷，导致弱化实际疾病进程的随机性和不确定性，忽略卫生服务利用时的多样性，使模型相对于疾病进程和卫生体系实际状况来说过于简化，不仅影响结果精确性，甚至可能对卫生决策产生误导。离散事件模拟模型克服了Markov模型的这些缺陷，模型通过内置仿真钟，灵活地记录所有发生的离散事件，同时允许考虑疾病相关重要危险因素和资源利用。因此本研究拟构建慢性疾病离散事件模拟模型，并用现有脑卒中二级预防数据进行实证，以证实模型的可行性，为慢性疾病决策分析提供参考依据。

The randomised controlled trials can be considered as the gold standard in evaluation of health intervention. However, additional modelling and observational data seem to be the most promising since the health systems are inherently complex. More critical still is that the Markovian assumption of Markov state transition model is not correct for most of the chronic diseases. Individual level Markov model still have two limitations. One is that they may not be suitable to tracking patient’s disease history properly. The other is the resources use could not be considered in the model. Discrete event model resolve these weaknesses and that its flexibility allows patients with different attributes to move from one event to another in sequential order while simultaneously taking into account important risk factors such as age, gender disease history and disease related events, together with resources use. Based on these reasons, we put forward a new model named ‘Chronic disease discrete event microsimulation model’ and plan to demonstrate it in decision analysis of secondary prevention of ischemic stroke. The model should depict real word disease progression as closely as possible.