【作者】 克瓦库；

【导师】 谭红专；

【作者基本信息】 中南大学 ， 流行病与卫生统计学， 2010， 博士

【摘要】 背景自古以来,结核病就是全球一个主要的公共卫生问题。据估计,全球90%以上的结核病例和死亡发生在发展中国家,而且主要影响具有经济产出能力的年龄群体(15-54岁)。几个世纪以来,结核病的控制逐步发展,经历了以下几个阶段：结核菌的发现；疗养时代；结核分支杆菌的识别；人工气胸主动治疗；x射线的应用；卡介苗的研制；结核菌素实验的发明；链霉素、对氨基水杨酸、异烟肼、吡嗪酰胺、环丝氨酸、乙胺丁醇、利福平等抗结核药物的问世；国家垂直管理的结核病控制项目的发展；结核病直接观察疗法(DOT)综合控制计划的实施；1991年世界卫生大会制订结核病控制目标：到2000年传染性结核病的治愈率达到85%和估计病例的发现率达到70%；WHO建立遏制结核部门,负责制定控制结核的专门管理方案；随后将世界卫生大会制定的2000年全球结核控制目标延迟到2005年,以及明确了在2015年要逆转全球结核发病率的千年发展目标。由于结核菌与人类免疫缺陷病毒(HIV)双重感染,加上耐多药结核菌的产生和蔓延,使得全球结核病形势急剧恶化。中国是世界上22个结核病高负担的国家之一,在2005年末全国结核病治愈率达到94%(高于全球85%的目标)。而加纳,并非结核病高负担国家,但其结核病的治愈率较低,仅为72%。目的和目标本次研究旨在利用2005年-2006年的全国监测数据综合分析湖南省结核病治疗结果,并与加纳东部地区进行治疗成功率的比较,为两国的结核病防治计划提供有用的信息。具体目标是阐明湖南省结核病治疗结果的影响因素,建立治疗结果的预测模型,进一步评价湖南和加纳东部在结核病成功治疗方面的差异。方法对湖南省的数据,用于分析的治疗结果包括成功治疗,失败,结核死亡和一般死亡。分析这些指标与结核患者的人口学特征(如性别、年龄、城市居民的人均GDP)、患者分类(包括新发病例或老病例)、治疗方案(Ⅰ类,Ⅱ类,Ⅲ类)；治疗付款方式(免费或自费)及治疗方式(直接观察下治疗,改良的直接观察下治疗,病例管理,自行治疗)的关系。单因素分析采用卡方检验和Fisher’s确切概率法来检验治疗结果与解释变量间的相关,运用多变量Logistic回归分析来检验所有有统计学显著性的解释变量对每种治疗结局的独立效应,并建立治疗结果的预测模型。为分析湖南和加纳东部在结核病成功治疗方面的差异,从湖南省和加纳东部各选取2个城市,即省会城市/区首府和随机抽取的一个经济不发达城市。在两个国家的监测数据中同时存在的变量被用于评价两国成功治疗差异的解释变量,这些变量包括性别、年龄组、患者类别、城市居民类型以及治疗方案。不同治疗方案(Ⅰ类,Ⅱ类,Ⅲ类)之间的治疗结果进行互相比较。运用多变量Logistic回归分析法进一步分析各解释变量对成功治疗的独立效应。为了探讨国家在结核病成功治疗上的作用,在将两研究地区资料合并进行多元Logistic回归分析时,将国家这个变量也归为解释变量。结果对湖南省登记的68430名结核病人进行了评价。研究对象中,大多数为男性,占72.9%；接近一半的人(48.8%)的年龄在15-44岁之间；新发结核病占93.3%；50.7%的城市居民人均GDP为1000-1999 US$；接受Ⅰ类治疗的占93.4%。几乎所有的患者(97.3%)都是免费治疗,DOT是主要的治疗方式,占85.9%。大多数患者(67%)在治疗前痰检阳性,其中96.2%的人得到了成功治疗。男性成功治疗的优势比低于女性(OR=0.87; 95%CI=0.79,0.95),相反,治疗失败的风险更高(OR=1.25; 95%CI=1.08,1.44)。与65岁及以上年龄的病人相比,0-14岁病人成功治疗的优势比最高(OR=6.46；95%CI=2.40,17.35),治疗失败的风险最低(OR=0.67; 95%CI=0.57,0.78),死亡(包括因TB所致的死亡和总死亡)风险也是最低的,0-14岁组病人没有死亡报告。随着年龄的增加,成功治疗的比例逐步下降,而治疗失败和死亡的风险逐步增加。与旧病例相比,新发结核病人具有较高的成功治疗优势比(OR=1.40; 95%CI=1.03,1.89),较低的死亡风险(OR=0.50；95%CI=0.33,0.76)。居住在人均GDP少于1000 US$地区的病人与居住在人均GDP2,000 US$及以上地区的病人相比,治疗失败的风险最低(OR=0.60；95%CI=0.49,0.72),而结核相关死亡的风险最高(OR=2.35；95%CI=1.81,3.05)。接受Ⅱ类治疗方案的病人比接受Ⅰ类治疗方案的病人治疗失败的风险高(OR=1.99; 95%CI=1.22,3.23).免费治疗病人的总死亡率比自费治疗病人要高(OR=12.54; 95%CI=3.92,40.15)。在所有的治疗方式中,采取病例管理的病人成功治疗率最高,与其相比,采用DOT和MDOT治疗方式的病人的成功治疗比较低(OR=0.78,95%CI=0.65-0.94, OR=0.65,95%CI=0.51-0.82),而采取自行治疗的病人与病例管理组相比,其死亡风险最高(OR=3.47; 95%CI=1.27-9.46)。在湖南以及加纳东部(2个省会城市/区首府和2个经济不发达城市),分别选取了7384个病例和483个病例,总体上来说,湖南结核病治疗的成功率(93.1%)显著高于加纳东部(60.7%)。分层分析在各层都显示了相同的结果。在湖南,与成功治疗相关的关键因素包括性别,年龄和居住城市的类型；在加纳东部,与成功治疗相关的关键因素只有年龄。在湖南,男性成功治疗的优势比明显低于女性(OR=0.78,95%CI=0.63-0.97)；居住在省会城市的病人,成功治疗的优势比较高(OR=1.35,95%CI=1.12-1.62)；在湖南和加纳东部,治疗成功率都随着年龄增加而下降。两研究地区合并的资料分析结果提示,湖南的治疗成功率明显高于加纳东部(OR=9.84,95%CI=7.89-12.18)。结论本次研究显示湖南省结核病治疗结果与国家监测资料库中的几个关键变量均有关,这些变量包括性别,年龄,患者类别,居住城市居民的人均GDP,治疗方案,治疗付款方式和治疗方式。而且,研究还表明中国的结核病控制与加纳相比较好。在中国的卫生融资,网络直报系统,医疗从业者的培训和巩固及其他方面吸取有益的教训,将有助于改善加纳的结核病防治形势。更多还原

【Abstract】 BACKGROUNDTuberculosis (TB) has remained a major global public health problem since antiquity. It is estimated that over 90% of global TB cases and TB deaths occur in developing countries, and affects mainly people within the economically productive age group (15-54 years). Tuberculosis control has evolved over the centuries. The TB control evolution includes the identification of tubercles; the sanitoria era; identification of Mycobacterium tuberculosis; artificial pneumothorax as an active therapy; the discovery of x-ray; development of BCG vaccine; tuberculin testing; the discovery of streptomycin, p-aminosalicylic acid, isoniazid, pyrazinamide, cycloserine, ethambutol, and rifampicin; the development of vertical national control programmes; development of integrated control programmes using directly observed therapy (DOT); the setting of World Health Assembly (WHA) targets in 1991 to achieve a cure rate of 85% of infectious TB cases and detecting 70% of estimated cases by the year 2000; creation of the stop TB department within WHO to provide specialized managerial approach to TB control; deferral of the 2000 WHA target to 2005; and the millennium development goal of reversing the global incidence of TB by 2015. The problem of TB has been aggravated by co-infections with the human immunodeficiency virus (HIV) and the emergence and spread of multi-drug resistance. China is listed among the 22 high-burden countries in the world. However, the nation by the end of 2005 had attained a treatment success rate of 94%(above the global target of 85%) whilst Ghana, not among the high-burden countries, had attained a lower treatment success rate of 72%.AIMS AND OBJECTIVESThis study aimed at using the 2005-2006 national surveillance data to comprehensively analyze TB treatment outcomes within the Hunan province, and compare treatment success between the Province and Eastern Ghana with the view of providing useful information for TB control. The specific objectives of the study were to elucidate factors associated with TB treatment outcomes in Hunan; to assess prediction models for TB treatment outcomes in Hunan; and to assess TB treatment success differences between Hunan and Eastern Ghana.METHODSTreatment outcomes selected for analyzing the Hunan data were successfully treated, defaulted, TB deaths, and general deaths. These were analyzed by demographic characteristics of patient (gender, age group, and per capita GDP of city of residence); classification of patient (new or old case); treatment regimen (categoryⅠ, categoryⅡ, and categoryⅢ); mode of payment for treatment (free or payment made); and mode of treatment (DOT, modified DOT, Case management, and self-treatment). Univariate analyses (Chi-square and fisher’s exact tests) were used to determine significant associations between treatment outcome and the explanatory variables. Multivariate logistic regression analyses were used to determine the independent effects of all the significant explanatory variables on each selected treatment outcome, and to establish prediction models for the outcomes.For the comparative analyses of treatment success differences between Hunan and Eastern Ghana,2 cities were selected from both areas. These were the Provincial/Regional capital and a randomly selected city of lower economic status. Explanatory variables common to the 2 national registers were used to assess treatment success differences. The variables were gender, age group, patient classification, type of city of residence, and treatment regimen. Treatment outcomes for the different treatment regimens (categoryⅠ,Ⅱ, andⅢ) were compared with each other. Furthermore, multivariate logistic regression was used to analyze the independent effect of explanatory variables on treatment success. To confirm the role of country in determining treatment success, country was added to the explanatory variables in a combined data from the 2 study areas and used in the multivariate logistic regression.RESULTSA total of 68430 registered patients in Hunan were evaluated. The majority of these were males (72.9%); aged 15-44 years (48.8%); new TB patients (93.3%); residents of cities with per capita GDP of 1,000-1,999 US$ (50.7%); and received categoryⅠtreatment (93.4%). Nearly all patients (97.3%) received free treatment, and DOT was the main mode of treatment used (85.9%). The majorities of patients were smear positive (67%) prior to treatment, and were successfully treated (96.2%). Males showed a lower odds ratio for treatment success compared with females (OR=0.87; 95% CI=0.79,0.95), and were at a higher risk of treatment default (OR=1.25; 95% CI=1.08,1.44). The odds ratios for treatment success were highest among patients aged 0-14 years compared with those aged 65 years or above (OR=6.46; 95% CI=2.40,17.35), and decreased with increasing age. The risk of default was lowest among patients aged 15-44 years (OR=0.67; 95% CI=0.57,0.78), and increased with increasing age. The risk of death (both TB-related and general) increased with increasing age, with no reported deaths among patients aged 0-14 years. New TB cases showed higher odds ratios for treatment success (OR=1.40; 95% CI=1.03,1.89) and lower risk of death (OR=0.50; 95% CI=0.33,0.76), compared with old cases. Patients living in cities with per capita GDP of less than 1,000 US$ showed the lowest risk of default (OR=0.60; 95% CI=0.49,0.72) and the highest risk of TB-related death (OR=2.35; 95% CI=1.81, 3.05), compared with patients living in cities with per capita GDP of 2,000 US$ and above. Patients receiving categoryⅡtreatment regimen were at a higher risk of default compared with those receiving categoryⅠtreatment regimen (OR=1.99; 95% CI=1.22,3.23). The odds ratios for general deaths were higher among patients who received free treatment compared with those who paid for their treatment (OR=12.54; 95% CI=3.92,40.15). Patients treated under DOT and MDOT showed lower odds ratios for treatment success compared with patients treated under case management (OR=0.78; 95% CI=0.65,0.94 and OR=0.65,95% CI=0.51,0.82, respectively) whilst patients who treated themselves showed the highest risk of death compared with those treated under case management (OR=3.47; 95% CI=1.27,9.46).Selection of the Provincial/Regional capital and another city of lower economic status yielded 7384 cases for Hunan and 483 cases for Eastern Ghana. Generally, TB treatment success rate was significantly higher in Hunan than Eastern Ghana (93.1% vs.60.7%). Stratified analyses also showed a similar pattern for all the group-specific rates. Key factors associated with treatment success in Hunan were gender, age group, and type of city of residence whilst in Eastern Ghana treatment success associated with only age group. The odds ratio for treatment success in Hunan was significantly lower for males (OR=0.78; 95% CI=0.63,0.97). The odds ratio of treatment success were also higher for patients resident in the provincial capital (OR=1.35; 95% CI=1.12,1.62). In both Hunan and Eastern Ghana treatment success rates decreased with increasing age. Analysis of combined data from the 2 study areas further showed that Hunan had a significantly higher treatment success rate than Eastern Ghana (OR=9.84,95%CI=7.89-12.18).CONCLUSIONIn conclusion, the study has shown that TB treatment outcomes in the Hunan province are associated with key variables in the national surveillance dataset. These variables are gender, age group, patient classification, per capita GDP of city of residence, treatment regimen, mode of payment of treatment, and mode of treatment. Furthermore, the study suggests that TB control in China is comparatively better than that of Ghana. Useful lessons from China, in the area of health financing, internet-based reporting system, and training and retaining of health personnel, among others, would help improve the situation in Ghana.更多还原