基于潜变量增长混合模型的慢性乙肝患者治疗队列医疗费用的异质性研究  

Medical cost and its heterogeneity in patients with chronic hepatitis B cohort by latent growth mixture modeling

作　　者：喻嘉宏 贾卫东[2] 李章龙 李粤平 林畅琪 梁颖芳 李伟南 陈舸 周舒冬 饶淑芳 郜艳晖 李丽霞 YU Jia-hong;JIA Wei-dong;LI Zhang-long;LI Yue-ping;LIN Chang-qi;LIANG Ying-fang;LI Wei-nan;CHEN Ge;ZHOU Shu-dong;RAO Shu-fang;GAO Yan-hui;LI Li-xia(School of Public Health,Guangdong Pharmaceutical University,Guangzhou,Guangdong 510310,China;不详)

机构地区：[1]广东药科大学公共卫生学院,广东广州510310 [2]广州市第八人民医院,广东广州510310

出　　处：《现代预防医学》2020年第17期3073-3076,3108,共5页Modern Preventive Medicine

基　　金：国家自然科学基金项目(71573059)。

摘　　要：目的探讨广州市慢性乙型肝炎门诊患者医疗费用变化轨迹的异质性,以期为临床治疗方案的选择和相关部门政策的制定提供切实可行的建议。方法数据来源于广州市第八人民医院信息系统,研究对象为2008年1月至2015年12月期间在该医院就诊并诊断为“慢性乙型肝炎”的门诊患者。采用潜变量增长混合模型对患者的医疗费用进行建模,比较不同亚组的总费用、抗病毒药物使用情况以及治疗效果。结果本研究共纳入慢乙肝患者4296人,平均年龄(35.19±10.84)岁,其中男性占76.84%。基于LGMM模型识别出2个增长趋势不同的亚类:高费用组(2175人,占50.63%)和低费用组(2121人,占49.37%),结果显示“高费用组”患者的DNA应答率高于“低费用组”患者(F=7.81,P<0.05;F=5.29,P<0.05),ALT复常率也高于“低费用组”患者(F=5.40,P<0.05)。结论潜变量增长混合模型能够识别慢乙肝患者治疗队列中不可观测亚类的医疗费用变化轨迹,反映真实的慢乙肝患者治疗过程中医疗费用及疗效情况。Objective To explore the heterogeneity of medical expenses track of outpatients with chronic hepatitis B in Guangzhou City,so as to provide reasonable advice for the clinical treatment plans selecting and the related policies making.Methods A retrospective cohort study was undertaken with the patients diagnosed with chronic hepatitis B from January 2008 to December 2015.The medical expenses data were obtained from the HIS system of Guangzhou Eighth People’s Hospital.The latent growth mixture model was used to model medical expenses of outpatients with chronic hepatitis B.The total cost,drug use and treatment indicators among different subgroups were compared.Results 4296 outpatients with chronic hepatitis B were enrolled in this study,76.84%of them were males,and the average age was(35.19±10.84).The outpatients were identified two latent classes,high cost group(2175,50.63%)and low cost group(2121,49.37%).The results showed the DNA response rate of high cost group was higher than that of low cost group(F=7.81,P<0.05;F=5.29,P<0.05),and the rate of ALT recurrence rate was also higher than that of low cost group(F=5.40,P<0.05).Conclusion Latent growth mixture model can identify unobserved subgroups and their different medical expenses trajectories for the cohort of outpatients with CHB,it reflects the change of the real medical expenses and effects of CHB patients.

关 键 词：乙型病毒性肝炎 潜变量增长混合模型 医疗费用 

分 类 号：R512.62[医药卫生—内科学]