【作者】 赵文静；

【导师】 尹爱田； Lizheng Shi；

【作者基本信息】 山东大学 ， 社会医学与卫生事业管理， 2014， 博士

【摘要】 研究背景糖尿病与结核病均是我国重大的公共卫生问题。2009年,全国抽样调查结果显示我国糖尿病患病率已高达9.7‰并有不断上升的趋势。据WH02013年全球结核病控制报告可知,中国的结核病负担位居全球第二,耐多药结核病患者数量位居全球首位,2012年我国结核病患病人数高达90万。两种疾病给患者、家庭以及社会均造成了沉重的经济负担。各个国家的多项研究证实糖尿病患者是结核病发病的危险人群。因此,在我国糖尿病合并结核病的发生会产生大量的结核病患者,造成我国结核病更加沉重的疾病负担。同时,糖尿病患者合并结核病病情严重,给结核病的治疗带来巨大的困难,会导致痰培养阴转时间延长,治疗失败率、耐多药发生率以及死亡率增加。在我国,现有的结核病控制策略中最有效的是早期病例发现,早期发现结核病病例利于结核病的早期治疗和转归,降低结核病的经济负担,控制结核病的传染,将结核病控制的关口前移。在结核病危险人群中进行结核病的病例发现易于发现更多病例,取得更好的效果。因此,基于我国庞大的糖尿病人口基数和两种疾病之间的联系,在我国糖尿病患者中进行结核病病例的早期发现对我国结核病控制工作会产生重大意义。我国现有主要的糖尿病患者中结核病病例发现策略包括被动发现策略和主动发现策略。被动发现策略即为因症就诊,以患者为导向,患者由于出现结核病可疑症状到结核病防治机构进行结核病的诊断和治疗。主动发现策略是以项目或机构为导向,在特定人群中加强结核病的病例发现,我国糖尿病患者中结核病病例主动发现策略包括社区主动发现策略和门诊主动发现策略。社区主动发现策略即由卫生机构人员在社区范围内最大限度纳入糖尿病患者进行结核病病例发现。门诊主动发现策略即由卫生机构人员在医院内分泌科门诊对前来就诊的糖尿病患者进行结核病的病例发现。通过在结核病危险人群中进行结核病病例的主动发现能够实现结核病的早期发现,从政府决策的角度考虑,如何合理、定量的对糖尿病患者中结核病病例的主动发现策略的经济学效果进行评价成为决策者关注的重要问题。病例发现策略是病例发现方式和病例发现工具的组合模式,虽然国外已经开始开展侧重于结核病病例发现方式的病例发现策略的经济学评价,但研究的核心领域还是对病例发现工具的经济学评价研究,国内结核病病例发现策略的经济学评价对象集中于病例发现工具的经济学评价,国内外均缺乏糖尿病患者中结核病病例发现策略的经济学评价研究。国内外研究表明决策树模型用于结核病控制策略优化具有直观、多元和量化的特点,辅助以卫生经济学相关的指标容易被理解和接受,马尔可夫模型可通过模拟疾病慢性演进的复杂过程评价不同控制策略下研究对象的生命年数、质量调整生命年等效果,由于结核病是慢性传染性疾病,决策树-马尔可夫模型能够综合两种模型的优点,应用于结核病长期效果的评价研究。结核病控制领域中使用决策树-马尔可夫模型进行经济学评价在国外已有较广泛的使用,但在国内多使用决策树模型,对决策树-马尔可夫模型的联合使用尚比较欠缺。本研究通过构建我国现有的糖尿病患者中结核病病例发现策略的决策树-马尔可夫模型,进而进行效果评价和经济学评价,对现有的病例发现策略进行优选,为卫生决策者制定结核病病例早期发现策略提供循证依据。研究目的总体研究目标：通过全面分析我国现有主要的糖尿病患者中结核病病例主动发现策略的特点以及策略所依托的项目调查资料和文献资料构建病例发现策略的决策树-马尔可夫模型,定量地对糖尿病患者中结核病病例主动发现策略进行效果、效用评价和经济学评价,为我国结核病病例早期发现策略的优选提供循证依据。具体研究目的：(1)全面分析我国现有主要的糖尿病患者中结核病病例主动发现策略的特点及依托的项目调查结果。(2)利用项目调查数据及文献资料提供重要参数,构建我国糖尿病患者中结核病病例主动发现策略的决策树-马尔可夫模型。(3)利用所构建的决策树-马尔可夫模型对现有主要的病例主动发现策略进行效果评价和经济学评价,进行病例发现策略的优选,为结核病的早期发现与防治策略提供科学依据。研究方法(1)经济学评价方法本研究的经济学评价采用最广泛也是最合理的社会角度,成本分析立足于社会角度,不同病例发现策略所产生的成本包括筛查成本和感染成本。筛查成本即不同病例发现策略所依托项目的发现成本,本研究所测量的为直接成本,包括直接卫生服务成本和直接非卫生服务成本。感染成本即结核病感染所造成的疾病社会经济负担。本研究所选取的结果指标包括效果指标和效用指标,效果指标包括短期结果和长期结果,短期结果指标包括结核病患者发现数、预防的结核病患者数,长期结果指标包括避免的结核病患者数、降低的结核病发病率、避免的人群死亡数、降低的人群死亡率、挽救生命年,效用指标为额外获得的质量调整生命年,除结核病患者发现数外,所有指标均以被动发现策略为分析对照。本研究对成本和健康结果以同样的率进行贴现。本研究针对短期结果构建不同病例发现策略的决策树模型,针对长期结果和效用指标构建决策树-马尔可夫模型,模拟糖尿病人群队列在不同病例发现策略下,经过20年结核病潜伏发病期所产生的期望成本和期望结果,进行经济学评价,所采用经济学分析方法为成本-效果分析与成本-效用分析。最后对研究的不确定性进行分析,即敏感性分析,分析重要参数的变化对不同病例发现策略的成本效果比、成本效用比的影响,评估模型的稳定度。(2)资料来源与方法我国糖尿病患者中结核病病例的被动发现策略即为我国结核病病例发现的常规工作。社区主动发现策略依托于世界卫生组织西太地区办公室结核病控制项目所资助的《加强糖尿病患者中结核病病例发现的实施性研究》,是本课题组于2012年5月-7月在我国山东省选取两个样本地区作为研究现场进行的结核病病例发现研究,共纳入4,085名糖尿病患者,新发现14名结核病病例。门诊主动发现策略依托于世界糖尿病基金会所资助的在我国5个城市所进行的加强我国糖尿病患者中结核病病例发现的前瞻性实施研究,该研究根据经济发展水平、地理位置在我国选择了5个城市,每个城市分别选择一家医院进行现场调查,在医院内分泌科门诊对前来就诊的糖尿病患者纳入结核病的筛查,现场调查持续时间为2011年9月到2012年5月,共纳入11,330名门诊糖尿病患者,新发现48名结核病病例。所构建决策树与决策树-马尔可夫模型中成本参数包括筛查成本和感染成本参数,筛查成本参数主要来源于项目的现场调查,个别参数参考文献分析数据,感染成本参数主要参考文献分析数据及专家咨询结果。筛查过程参数包括糖尿病患者的纳入率和糖尿病患者中结核病病例发现率,是模型中的核心参数,社区和门诊主动发现策略下筛查过程参数均来源于现场调查结果,被动发现策略下参数来源于世界卫生组织所公布的我国国家层面数据。马尔可夫模型模拟疾病慢性演进的复杂过程,即结核病各种疾病状态之间的相互转换过程,结核病感染参数即各疾病状态之间相互转换的概率,主要来源于世界卫生组织报告、我国结核病发病预测模型及相关文献资料数据和专家咨询结果。研究结果(1)成本分析：包括对糖尿病患者中结核病病例不同主动发现策略的筛查成本和感染成本进行分析。社区主动发现策略下所花费的总成本为57,970元,共筛查4,085名糖尿病患者,因此,社区主动发现策略下对糖尿病患者进行筛查的例均成本为14.19元；门诊主动发现策略下所花费的总成本为129,704元,共筛查11,330名糖尿病患者,因此,社区主动发现策略下对糖尿病患者进行筛查的例均成本为11.45元。不同病例发现策略下感染成本相同,非耐药结核病感染成本为2,908.32元,耐多药结核病感染成本为5,561.67元。将成本参数纳入决策树-马尔可夫模型,模拟10,000名糖尿病患者,循环20个周期即结核病的潜伏发病期,得到循环周期末不同病例发现策略下所发生的期望成本,社区主动发现策略下10,000名糖尿病患者所发生的筛查成本和感染成本总和为2,729,141.79元；门诊主动发现策略下10,000名糖尿病患者所发生的筛查成本和感染成本总和为2,397,796.63元。(2)效果、效用分析：除结核病患者发现数外,其它指标分析均以被动发现策略的模型结果作为分析对照,结核病患者发现数为主动发现的人数,因此不需纳入被动发现策略进行分析。中间结果即短期效果分析：1)结核病患者发现数根据社区主动发现策略和门诊主动发现策略所依托的项目调查结果,两种主动发现策略分别发现14例和48例结核病患者,从社会层面分析不同病例发现策略的效果,将不同方式对糖尿病患者的纳入情况纳入决策树模型,模拟10,000名糖尿病患者,结果显示,社区主动发现策略能够发现26名结核病患者,门诊主动发现策略能够发现13名结核病患者,理论状态即常规的被动发现策略下能够发现7名结核病患者。2)预防的结核病患者数以理论状态即常规的被动发现策略作为对照,社区主动发现策略能够预防的结核病患者数为19人,门诊主动发现策略能够预防的结核病患者数为6人。最终结果即长期效果分析：1)避免的结核病患者数以理论状态即常规的被动发现策略作为对照,社区主动发现策略下10,000名糖尿病患者循环20年后能够避免的结核病患者数为31人,门诊主动发现策略能够避免的结核病患者数为10人。2)降低的结核病发病率以理论状态即常规的被动发现策略作为对照,社区主动发现策略下10,000名糖尿病患者循环20年后能够降低的结核病发病率为26.2/10万,门诊主动发现策略能够降低的结核病发病率为8.5/10万。3)避免的人群死亡数以理论状态即常规的被动发现策略作为对照,社区主动发现策略下10,000名糖尿病患者循环20年后能够避免的人群死亡数为83人,门诊主动发现策略能够避免的人群死亡数为30人。4)降低的人群死亡率以理论状态即常规的被动发现策略作为对照,社区主动发现策略下10,000名糖尿病患者循环20年后能够降低的人群死亡率为0.7‰,门诊主动发现策略能够降低的人群死亡率为0.2‰。5)挽救生命年以理论状态即常规的被动发现策略作为对照,社区主动发现策略下10,000名糖尿病患者循环20年后能够挽救的生命年数为167年,门诊主动发现策略能够挽救的生命年数为65年。效用分析：即额外获得的质量调整生命年以理论状态即常规的被动发现策略作为对照,社区主动发现策略下10,000名糖尿病患者循环20年后能够额外获得的质量调整生命年数为152年,门诊主动发现策略能够额外获得的质量调整生命年数为61年。(3)成本-效果与成本-效用分析：社区主动发现策略每发现一名结核病患者所需花费的成本为3,749.53元,门诊主动发现策略每发现一名结核病患者所需花费的成本为1,770.35元,增量分析结果显示社区主动发现策略与门诊主动发现策略相比,增量成本效果比即每多发现一名结核病患者所花费的成本为5,728.52元；社区主动发现策略每预防一名结核病患者所需花费的成本为5,130.80元,门诊主动发现策略每发现一名结核病患者所需花费的成本为3,835.76元,增量成本效果比即每多预防一名结核病患者所花费的成本为5,728.52元。社区主动发现策略下10,000名糖尿病患者循环20年后每避免一名结核病患者所需花费的成本为88,036.83元,门诊主动发现策略每避免一名结核病患者所需花费的成本为239,779.66元,增量成本效果比即社区主动发现策略下每多避免一名结核病患者所花费的成本为15,788.34元；社区主动发现策略下10,000名糖尿病患者循环20年后每降低1/10万的结核病发病率所需花费的成本为104,165.72元,门诊主动发现策略每降低1/10万的结核病发病率所需花费的成本为282,093.72元,增量成本效果比即社区主动发现策略下即每多降低班0万的结核病发病率所多花费的成本为18,720.07元;社区主动发现策略下10,000名糖尿病患者循环20年后每避免一名糖尿病人群死亡患者所需花费的成本为32,881.23元,门诊主动发现策略每避免一名糖尿病人群死亡患者所需花费的成本为79,926.55元,增量成本效果比即社区主动发现策略下每多避免一名糖尿病人群死亡患者所花费的成本为6,251.80元；社区主动发现策略下10,000名糖尿病患者循环20年后每降低1‰的人群死亡率所需花费的成本为3,898,773.99元,门诊主动发现策略对模拟的10,000名糖尿病患者每降低1‰的人群死亡率所需花费的成本为11,988,983.15元,增量成本效果比即社区主动发现策略下每多降低1‰的人群死亡率所花费的成本为157,783.41元；社区主动发现策略下所模拟的10,000名糖尿病患者循环20个周期每挽救1个生命年所需花费的成本为16,342.17,元,门诊主动发现策略每挽救1个生命年所需花费的成本为36,889.18元,增量成本效果比即社区主动发现策略每多挽救一个生命年所需花费的成本为3,248.48元。社区主动发现策略下所模拟的10,000名糖尿病患者循环20个周期每额外获得1个QALY所需花费的成本为17,954.88元,门诊主动发现策略每额外获得1个QALY所需花费的成本为38,060.26元,增量成本效果比即社区主动发现策略每增加一个额外的QALY所需花费的成本为3,722.97元。结论与政策建议结论：(1)本研究发现,社区主动发现策略的中间结果即短期效果优于门诊主动发现策略,即能在糖尿病患者中发现更多的结核病患者和预防更多的结核病患者,从长远来看,本研究所构建的决策树-马尔可夫模型结果提示,社区主动发现策略的最终结果即长期效果优于门诊主动发现策略；(2)本研究发现,以中间结果为指标进行的成本效果分析显示,门诊主动发现策略优于社区主动发现策略,以最终结果和效用为结果指标所进行的成本效果与成本效用分析显示,社区主动发现策略优于门诊主动发现策略；(3)增量分析结果显示,以2012年我国人均国内生产总值38,459.47元作为第三方的付费标准,与门诊主动发现策略相比,除降低的人群死亡率这一指标,社区主动发现策略每多获得一个健康结果所需付出的成本均低于人均国内生产总值,按照世界卫生组织对经济学评价的推荐意见,社区主动发现策略的实施符合经济学原则。糖尿病患者是结核病的高发人群,加强糖尿病与结核病的协同管理与控制在我国具有深远的意义,结合效果分析与成本效果分析结果,基于结核病的外部效应特点,在我国采取社区主动发现策略会取得更加良好的效果。利用现有的工具在我国加强糖尿病与结核病两种疾病的管理与控制最核心的是建立两种疾病早期发现的协同管理机制,主要包括：(1)加强对社区居民健康档案的完善,加强对社区糖尿病患者的规范管理,利于糖尿病和结核病两种疾病的有效控制；(2)同时,建议将结核病的疾病管理纳入基本公共卫生服务项目,社区主动发现策略的经济学评价结果提示对社区内糖尿病患者进行结核病病例的主动发现会产生良好的效果同时符合经济学原则；(3)通过强化社区功能、强化社区医生和综合医院专科医师职责,对易感人群定期普查等规范糖尿病患者中结核病的早期发现工作,通过机制建设规范筛查程序；(4)在糖尿病与结核病监测过程中,门诊主动发现策略可以作为补充方式纳入常规监测项目,加强结核病的早期发现。更多还原

【Abstract】 BackgroundTuberculosis (TB) and diabetes mellitus (DM) are major public health challenge in China. In2009, the national sampling survey results show that DM prevalence has been reached to9.7%, and will keep rising in future. According to WHO2013global TB control report, the burden of TB in China takes the second part in the world, and the number of Multi Drug Resistant TB (MDR-TB) patients comes to the first. In2012, the number of patients with TB is90,000in China. Both of the two diseases have been caused heavy economic burden to patients, families, and society. A series of studies from different countries have confirmed that people with DM are at high risk of TB. More seriously, patients with comorbidity of DM and TB lead to the difficulty of treatment, extended sputum culture turn, increased treatment failure, drug resistant rate and mortality rate.In China, the most effective in TB control strategy is case finding. Early finding of TB cases will lead to the early treatment and better outcome of TB, reducing the economic burden of TB, controlling the spread of TB, and moving forward the control gate of TB. Delivering TB case finding among people with risk of TB will diagnose more cases, and achieve better effect. Therefore, based on huge DM population in China and the connection between the two diseases, conducting TB case finding in patients with DM is significant meaningful and important for TB control.In China, TB cases finding in patients with DM include passive case finding and active case finding. Passive case finding is patient oriented, patients with suspected TB seek for service t in TB prevention and control institutions due to their symptoms. Active case finding is project or institutions oriented, it strengthen TB case detection in particular patients. In China, active case finding includes community-based active case finding and clinic-based active case finding. Community-based active case finding is conducted by health institutions through covering the community patients with DM to maximum TB case detection. Clinic-based active case finding is hold by health provider from clinic of endocrinology in the hospital for examination of DM in TB case finding. Active case finding is effective for early finding for TB cases. From the point of government policy-making, it is essential to focus on how to evaluate the economic effect of cases finding in patients with DM using appropriate quantitative methods. Due to the recent researches, study of TB case finding in patients with DM is limited, and lack of evidence on economics evaluation of TB case finding in patients with DM. Previous studies focus on economics evaluation of TB case finding tool.The domestic and foreign researches have shown that decision-tree model is intuitionistic, multivariate and quantized, which is used for optimization of TB controlling strategies and is easy to be understood and accepted assisted with relevant health economics indicators. Markov model can evaluate the effectiveness of life years (LY) and quality adjusted life years (QALYs) of the research objects under different controlling strategies through simulating the complex process of disease chronic evolution. As TB is a chronic infectious disease, decision tree-Markov model can combine the merits of two models and apply in the evaluation research of TB ultimate effectiveness. Although applying decision tree-Markov model for economic evaluation in TB controlling area is common in foreign researches, decision-tree model is applied more often in domestic researches while the combination of decision tree-Markov model is lack of application.This research carries out effectiveness evaluation and economic evaluation by building decision tree-Markov model for TB case finding strategies among patients with DM in China and optimizes the existing case finding strategies in order to provide evidence for health decision makers to formulate TB controlling strategies.ObjectivesGeneral objective:To build decision tree-Markov model for the existing main case finding strategies and carry out effectiveness evaluation and economic evaluation for existing case finding strategies quantitatively to provide evidence-based evidence for China’s TB controlling strategies. This is based on data of three parts:the comprehensive analysis of the characters of China’s existing main TB case finding strategies in patients with DM, the program survey data for community-based active case finding and clinic-based active case finding, the regular data and literature data analysis of passive case finding in China.Specific objectives:(1) To analyze the characters of China’s existing main TB case finding strategies in patients with DM and the survey results of program.(2) To provide important parameters through program survey data and literature data, and build decision tree-Markov model for TB case finding strategies in patients with DM in China.(3) To carry out effectiveness evaluation and economic evaluation for the existing main case finding strategies and optimize the case finding strategies using the decision tree-Markov model in order to provide scientific evidence for health decision making.Methods(1) Economic Evaluation Methods. The economic evaluation of this research is based on social perspective, which is most common and most rational. Cost analysis is based on social perspective and the total expected cost produced by different case finding strategies includes screening cost and infection cost. Screening cost is cost of active programs of different case finding strategies, and what we measured in this research is direct cost including direct health service cost and direct non-health service cost. Infection cost is social economic burden caused by TB infection. The result indicators selected in this research include effectiveness and utility. Effectiveness includes intermediate effectiveness and ultimate effectiveness. TB cases found and TB cases prevented are the intermediate effectiveness indicators. Ultimate effectiveness indicators include TB cases averted, TB morbidity reduced, Death averted, Mortality reduced and life year saved. Utility indicator is quality adjusted life years (QALYs). This research discounts the health results using the same rate as cost. This research builds decision-tree models of different case finding strategies for intermediate effectiveness and decision tree-Markov models for ultimate effectiveness. Then simulate the expected cost and expected results in a cohort of patients with DM after20years TB incubation period under different case finding strategies. After that, apply cost-effectiveness analysis and cost-utility analysis to carry out the economic evaluation. Finally, analyze the uncertainty of this research, which is called sensitivity analysis. It analyzes the influence of change of important parameters on cost-effectiveness ratio and cost-utility ratio of different case finding strategies.(2) Data source and methods. The passive TB case finding for patients with DM in China is the routine work of TB case finding. Community-based active case finding relies on the program of Operational Research in Intensified Case Finding for Active TB in a High-Risk Population with DM, which is funded by Office of WHO Western Pacific Region TB control program. The research group carried out the TB case finding research from May to July,2012in two sample areas of Shandong Province, China.4,085patients with DM were included and14TB cases were newly found. Clinic-based active case finding relies on the program of Perspective Research of TB case finding in patients with DM in China funded by World DM Foundation, which was carried out in5cities in China. The5cities were selected according to the economic developing level and geographical location. The field investigation was carried out in one hospital of every city, which detects TB cases in patients with DM who seek for service in endocrinology clinics. The field investigation was carried out from September2011to May2012and11,330outpatient patients with DM were included, among which48TB cases were newly found.The cost parameters in the decision-tree models and decision tree-Markov models built include screening cost and infection cost parameters. Screening cost parameters mainly came from field investigation of the program while some of them came from the literature analysis. Infection cost parameters mainly referred to the literature analysis data and experts consultation results. Screening procedure parameters include participation rate of patients with DM and TB case finding rate in patients with DM. Screening procedure parameters are the core parameters in the models. Screening procedure parameters in community-based and clinic-based active case finding strategies came from field investigation results while the parameters in passive case finding strategies came from the national level data published by WHO. Markov model simulates the complex process of disease chronic evolution, which is the interconverting process among different disease statuses of TB. TB infection parameters are the interconverting probabilities among different disease statuses of TB, which mainly come from WHO reports, TB morbidity predicting models in China, relevant literature data and experts consultation results.Main ResultsCost analysis:It includes the analysis of screening cost and infection cost of different case finding strategies.The total screening cost of community-based active case finding strategy is￥57,970,4,085patients with DM were screened. The screening cost of community-based active case finding strategy is￥14.19per case. The total screening cost of clinic-based active case finding strategy is￥129,704,11,330patients with DM were screened. The screening cost of clinic-based active case finding strategy is￥11.45per case. The infection cost of non-drug resistant TB is￥2,908.32, the infection cost of MDR-TB infection cost is￥5,561.67. A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model, the expected cost of community-based active case finding strategy is￥2,729,141.79; the expected cost of clinic-based active case finding strategy is￥2,397,796.63.(2) Effectiveness and utility analysis.Intermediate effectiveness analysis:1) TB cases found:According to the research findings of the project, there were14TB cases under the community-based active case finding strategy and48TB cases were found in the clinic-based active case finding strategy. To analysis the effect of different case finding in societal perspective, the participation rate was considered and was put into the decision tree model. A cohort of10,000patients with DM was simulated in the decision tree model. The results showed that26TB cases can be found under community-based active case finding strategy,13TB patients can be found under clinic-based active case finding strategy and7TB patients can be found under passive case finding strategy.2) TB cases prevented.19TB cases can be prevented under community-based active case finding strategy,6TB patients can be prevented under clinic-based active case finding strategy.Ultimate effectiveness analysis:1) TB cases averted. A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model,31TB cases can be averted under community-based active case finding strategy,10TB cases can be averted under clinic-based active case finding strategy.2) TB morbidity reduced. A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model, community-based active case finding strategy could reduce26.2/100,000TB morbidity, clinic-based active case finding strategy could reduce8.5/100,000TB morbidity.3) Deaths averted. A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model,83deaths can be averted under community-based active case finding strategy,30deaths can be averted under clinic-based active case finding strategy.4) Mortality reduced. A cohort of 10,000patients with DM was simulated for20years in the decision tree-Markov model, community-based active case finding strategy could reduce0.7‰TB mortality, clinic-based active case finding strategy could reduce0.2‰TB mortality.5) Life year saved. A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model,167life years can be saved under community-based active case finding strategy,65life years can be saved under clinic-based active case finding strategy.Utility analysis:Quality adjusted life years (QALYs). A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model,152QALYs can be gained additionally under community-based active case finding strategy,61QALYs can be gained additionally under clinic-based active case finding strategy.(3) Cost-effectiveness and cost-utility analysis:The cost of finding a TB case under community-based active case finding strategy was￥3,749.53, cost of finding a TB case under clinic-based active case finding strategy is■1,770.35, incremental analysis showed that￥5,728.52was needed to be paid when finding an additional TB case under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group. The cost of preventing a TB case under community-based active case finding strategy was￥5,130.80, cost of preventing a TB case under clinic-based active case finding strategy is￥3,856.76, incremental analysis showed that￥5,728.52was needed to be paid when preventing an additional TB case under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group.A cohort of10,000patients with DM was simulated for20years in the decision tree-Markov model. The cost of averting a TB case under community-based active case finding strategy was￥88,036.83, cost of averting a TB case under clinic-based active case finding strategy is￥239,779.66, incremental analysis showed that￥15,783.34was needed to be paid when averting an additional TB case under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group. The cost of reducing1/100,000TB morbidity under community-based active case finding strategy was￥104,165.72, cost of reducing1/100,000TB morbidity under clinic-based active case finding strategy is￥282,093.72, incremental analysis showed that￥18,720.07was needed to be paid when reducing additional1/100,000TB morbidity under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group. The cost of averting a death case under community-based active case finding strategy was￥32,881.23, cost of averting a TB death under clinic-based active case finding strategy is￥79,926.55, incremental analysis showed that￥6,251.80was needed to be paid when averting an additional TB death under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group. The cost of reducing1‰mortality under community-based active case finding strategy was￥3,898,773.99, cost of reducing1‰mortality under clinic-based active case finding strategy is￥11,988,983.15, incremental analysis showed that￥157,783.41was needed to be paid when reducing additional1‰mortality under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group. The cost of saving one life year under community-based active case finding strategy was￥16,342.17, cost of saving one life year under clinic-based active case finding strategy is￥36,889.18, incremental analysis showed that￥3,248.48was needed to be paid when additionally saving one life year under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group.The cost of additionally gaining one QALY under community-based active case finding strategy was￥17,954.88, cost of additionally gaining one QALY under clinic-based active case finding strategy is￥38,060.26, incremental analysis showed that￥3,722.97was needed to be paid when having one additional QALY under community-based active case finding strategy when using clinic-based active case finding strategy as the reference group. Conclusions and Policy ImplicationsConclusions:(1) The research found that, as for the intermediate effectiveness and ultimate effectiveness, community-based active case finding strategy is better than clinic-based case finding strategy;(2) As for intermediate effectiveness, clinic-based active case finding strategy is more cost-effective; as for ultimate effectiveness and utility(QALYs), community-based active case finding strategy is more cost-effective;(3) As for all effectiveness indicators except for mortality reduced, the incremental analysis showed that community-based active case finding strategy is consistent with the principles of economics when using China’s per capita gross domestic product (￥38,459.47) in2012as a third party payment standard (ICER<per capita gross domestic product).Patients with DM is a high-risk population of developing TB, collaborative management and control of DM and TB would have far-reaching significance in China. And according to the conclusion of the research, there would be suitable strategies for strengthening the management of the two diseases. Combined with effectiveness analysis and cost-effectiveness analysis, community-based active case finding strategy may be the best way when considering the external effects of TB. The core of using existing tools to strengthen the management and control of DM and TB disease is to establish collaborative management mechanisms to find the two diseases early, specific strategies include:(1) Strengthening the improvement of the community health records and the community-level management of patients with DM, which will help to effectively control DM and TB;(2) At the same time, it is recommended that management of patients with TB bringing into basic public health services. The result of the economic evaluation suggests the community-based active case finding strategy may be the best way;(3) Strengthening community functions and duties of community physicians and general hospital specialist, taking measures such as carrying regular census in vulnerable populations to find TB cases early and constructing the mechanism of standard screening programs;(4) In the course of monitoring DM and TB, clinic-based active case finding strategy could be a supplement ways of routine monitoring programs to enhance early detection of TB.更多还原