基于多状态Markov模型的老年人认知损害转归研究  

Outcome Predition of Mild Cognitive Impairment Based on Multi-state Markov Model

作　　者：杨珊珊 李新毅[2] 周立业[3] 梁瑞峰[3] 罗天娥[3] 曲成毅[3] 余红梅[3] 

机构地区：[1]北京市宣武区广外社区卫生服务中心 [2]山西医科大学第一医院神经内科 [3]山西医科大学公共卫生学院

出　　处：《中国卫生统计》2012年第4期516-519,共4页Chinese Journal of Health Statistics

基　　金：国家自然科学基金资助项目(30972545);山西省留学回国人员科技活动择优项目;山西医科大学科技创新项目

摘　　要：目的将多状态Markov模型应用于轻度认知损害(mild cognitive impairment,MCI)转归研究,为慢性病转归研究提供方法学借鉴。方法通过MCI患者IQ变化反映随访人群认知功能的变化趋势,构建一个四状态模型。根据多状态Markov模型分析原理,获得各状态转移影响因素、逗留时间、生存曲线,并进行模型拟合优度评价。结果 MCI病人处于认知功能稳定、认知轻度恶化和认知好转的时间大约分别为6.4年、3.6年和5.2年;生存曲线显示预后由好到差为认知功能好转、认知功能稳定、认知功能轻度恶化;多状态Markov模型拟合结果较好。结论多状态Markov模型是多状态、多阶段慢性病转归研究的有效分析方法。Objective The aim of this study was to make out- come prediction of mild cognitive impairment based on multi-state Markov model and to provide methodological reference for the chronic disease out- come research. Methods Based on IQ changes to reflect trends in cogni- tive function among MCI subjects, a four states model was constructed. Ac- cording to multi-state Markov model theory, state transition realated factors, sojourn time, survival curve, and model fit were obtained. Results So- journ times of cognitive function stable, slight deteriorated, and improved were years of 6. 4,3.6, and 5.2, respectively. Survival curve showed that subjects of cognitive function improved, stable, and slight deteriorated had best to worst prognosis. Model fit can be accepted. Conclusion Multi- state Markov model is an effective analysis method for outcome prediction of multi-state or multi-stage chronic diseases.

关 键 词：多状态MARKOV模型 阿尔茨海默病 轻度认知损害 

分 类 号：R749.16[医药卫生—神经病学与精神病学]