【作者】 张震巍；

【导师】 陈洁； 胡仁明； 陈英耀； 唐智柳；

【作者基本信息】 复旦大学 ， 社会医学与卫生事业管理， 2007， 博士

【摘要】 糖尿病及其并发症是影响我国城乡居民的三大类慢性疾病之一，已经成为严重威胁我国人民健康的重要公共卫生问题。研究它的疾病负担，衡量其对人群健康、社会经济和患者生命质量的影响是制定和评价糖尿病防治措施的重要前提。本研究的开展以我国最近的糖尿病流行病学数据、国际糖尿病联盟(International Diabetes Federation，IDF)推荐的糖尿病疾病经济负担计算公式以及欧洲五维度生命质量量表(EQ-5D)为基础，旨在更新社会对糖尿病疾病负担的认识，为减少糖尿病对人群的健康损害、控制糖尿病疾病经济负担提供基础信息。第一部分我国糖尿病流行病学负担研究【目的】分析我国糖尿病患病率、门诊就诊率、住院率的变化趋势。计算我国2005年的糖尿病死亡损失寿命年(Years of Life Lost，YLL)以及相应的间接经济损失。【方法】对糖尿病患病率、门诊就诊率、住院率变化情况进行趋势描述，采用世界卫生组织的疾病负担计算公式计算YLL和相应的间接经济损失。分析YLL的城乡分布、年龄分布和性别分布。【结果】1993—2003年，城乡人群的糖尿病患病率分别上升194.73％和375％。城乡糖尿病病人的卫生服务利用率都呈现快速上升趋势。城乡之间卫生服务利用率的相对差距不断缩小。与2000年相比，2005年卫生部门城市医院出院病人中糖尿病人的比例上升34.75％，卫生部门县医院出院病人中糖尿病人的比例上升13.43％。2005年，我国城乡每10万人口因糖尿病损失的YLL值分别为154.82和87.96，全国平均为105.75／10万。无论是城市还是农村，女性损失的YLL都高于男性。城市男性损失的YLL是农村男性的2.01倍，城市女性损失的YLL是农村女性的1.60倍。随着年龄上升，损失的YLL增加，从全国来看，在70岁时达到峰值，其中50岁—75岁年龄段损失的YLLs占73.41％。女性的峰值出现在50岁，城市和农村男性的峰值分别在70岁和55岁。2005年我国因糖尿病共损失139.95万YLLs，造成的间接经济损失为80.68亿元。【结论】糖尿病对我国居民健康造成的危害不断加重，并造成巨大的生产力损失，应进一步重视开展糖尿病的防治。第二部分2型糖尿病疾病经济负担研究【目的】估算我国2005年糖尿病疾病经济负担，评价国际糖尿病联盟(IDF)推荐的糖尿病疾病经济负担计算公式在我国使用的可行性。【方法】运用1∶1配比的病例对照研究，对上海市某中心城区2个毗邻街道筛选出的618名确诊2型糖尿病人以及按年龄、性别、医疗保障状况进行逐一配比的618名非糖尿病人进行调查。重点通过糖尿病人与非糖尿病人的人均直接卫生费用比值(根据年龄权重调整后)，运用公式计算我国的糖尿病疾病经济负担。并对公式计算的可靠性进行评价。对2型糖尿病人的血糖、血压、血脂控制情况和体重指数进行分析，采用多元线性回归模型对2型糖尿病人卫生费用的影响因素进行分析。【结果】1．我国糖尿病疾病经济负担糖尿病人的人均直接卫生费用是非糖尿病人的2.47倍。2005年我国糖尿病人的直接卫生费用为655.68亿元，占当年卫生总费用的7.57％。经过对IDF推荐的糖尿病疾病经济负担计算公式的验证，说明该公式具有较高的可靠性。2．糖尿病人卫生费用影响因素多元线性回归分析表明，年龄、工作状况、并发症、受教育程度、医疗保障状况、治疗方式(仅口服药、仅胰岛素、口服药+胰岛素)、吸烟、饮酒、高血脂与糖尿病人的直接费用有统计学关联。在其他因素一致时，年龄上升1岁，直接费用上升15.7％；有并发症(或合并症)者的直接费用比无并发症(或合并症)者高12.1％；有基本医疗保险的糖尿病人的直接费用比自费病人高13.5％；仅口服药治疗、仅胰岛素治疗、口服药+胰岛素治疗分别使直接费用上升21.0％、32.0％、36.5％；吸烟者的直接费用为不吸烟者的1.110倍；饮酒者的直接费用是不饮酒者的86.1％；高血脂者的直接费用是血脂正常者的92.4％。3．糖尿病人血糖、血压、血脂等糖尿病慢性并发症主要危险因素控制情况对618名2型糖尿病人进行检查显示，空腹血糖在理想范围内(＜6.0mmol／L)的比例为13.92％，糖化血红蛋白(HbA1c)在理想范围内(＜6.5％)的比例为38.19％。血压＜130／80mmHg(血压控制目标)的比例仅为14.08％。总胆固醇在理想范围内的比例为22.01％(＜4.5mmol／L)。糖尿病人中血糖、血压、血脂全面达标率仅为2.10％。体重指数达到目标值(男性＜25kg／m~2，女性＜24kg／m~2，)的比例，男性为49.36％，女性为40.78％。【结论】我国糖尿病直接卫生费用在全国卫生总费用中的比重已经接近甚至超过某些发达国家。初步认为IDF推荐的糖尿病经济负担计算公式适用于我国。改善糖尿病慢性并发症危险因素的控制。第三部分糖尿病及其并发症的住院费用研究【目的】比较糖尿病及其主要并发症的住院费用。【方法】选取上海市某三级甲等医院2000年1月1日—2005年12月31日期间的出院病人中，以糖尿病及其9类主要并发症(糖尿病合并高血压、糖尿病合并高脂血症、糖尿病眼病、糖尿病肾病、脑血管疾病、心血管疾病、糖尿病神经疾病、糖尿病酮症、糖尿病足部病变)为出院主要诊断或第一其他诊断的病人。对糖尿病及其并发症的住院费用进行方差分析，对住院费用的影响因素进行灰色关联分析。【结果】合并脑血管疾病的病人的例均住院费用低于合并心血管疾病病人，与糖尿病足部病变病人例均住院费用无显著性差别，高于其他各类病人的例均住院费用。所有病人平均住院日为21天，伴发脑血管疾病的病人的平均住院日最长(27天)，伴发高脂血症的病人的平均住院日最短(12天)。从日均住院费用均数来看，单纯糖尿病病人最低，合并心血管疾病病人最高。从住院费用的构成来说，对糖尿病及其并发症住院费用影响从大到小的顺序依次是：药费＞床位费＞化验检查费＞治疗费＞手术费。【结论】糖尿病并发症住院费用高昂，尤其是心脑血管并发症和糖尿病足部病变。采取多种措施控制糖尿病及其并发症的发生是降低住院费用的重要环节。第四部分2型糖尿病人健康相关生命质量病例对照研究【目的】比较2型糖尿病人与非糖尿病人的健康相关生命质量(health-related quality of life，HRQOL)，分析糖尿病人HRQoL的影响因素，评价EQ-5D在测量2型糖尿病人HRQoL中的信度和效度。【方法】人群的选择见第二部分。采用欧洲生命质量组织开发的EQ-5D量表(中文版)做为HRQoL测量工具。采用两分类logistic回归模型分析糖尿病人EQ-5D五维度描述系统的影响因素，采用多元线性回归的方法分析糖尿病人EQ VAS(Visual Analogue Scale，VAS)分值的影响因素。【结果】2型糖尿病人的HRQoL显著低于非糖尿病人。糖尿病人的EQ VAS分值显著低于非糖尿病人。在EQ 5D描述系统的五个维度中，报告“有一些问题”或者“有严重问题”的应答者比例，糖尿病组均高于对照组。2型糖尿病人HRQoL的影响因素包括：并发症、年龄、性别、家庭月均收入、住房面积、家庭规模、住房面积、吸烟、治疗措施、体育锻炼、饮食控制、高血压、血脂紊乱。【结论】控制糖尿病并发症，提倡戒烟和体育锻炼，纠正血脂紊乱，关注较低龄(在本研究中指40岁—50岁)的糖尿病人和独居者的的心理状况，改善糖尿病人HRQoL。EQ-5D具有较高的信度，能够有效区分出有并发症(或合并症)的糖尿病人与非糖尿病人、有并发症(或合并症)的糖尿病人和无并发症(及合并症)的糖尿病人的HRQoL差异。但是具有明显的天花板效应，即无论是糖尿病人还是非糖尿病人的EQ VAS分值都较高。全文总结与建议1、全文总结(1)糖尿病流行病学负担、疾病经济负担和生命质量研究结果显示，我国的糖尿病疾病负担上升较快，对患者的健康水平、社会经济产生较大危害。(2) IDF推荐的糖尿病疾病经济负担计算公式可靠性较高。(3)糖尿病并发症引起糖尿病人直接费用上升，损害病人的生命质量。(4)糖尿病人的健康意识偏低，糖尿病并发症危险因素控制效果不佳。2、建议(1)拟定预防、治疗和护理糖尿病的国家政策，为糖尿病的防治提供必要的社会支持环境和条件。(2)积极开展糖尿病的三级预防，控制糖尿病及其并发症的发生，降低糖尿病疾病经济负担，减少糖尿病及其并发症对病人生命质量的损害。(3)加强糖尿病疾病负担研究，为制定和评价糖尿病防治措施提供及时、准确的信息。更多还原

【Abstract】 As a consequence of increasing urbanization and associated lifestyle changes, diabetes mellitus has emerged as a major and growing public health problem in China. It is of great importance to measure its impact on population health and social development. This study is based on late data of epidemics of diabetes in China and the formulae proposed by International Diabetes Federation (IDF) to estimate the direct economic burden of diabetes. Health-related quality of life of patients with diabetes is the other important content in this study. The study aimed to provide timely imformation on the burden of diabetes in China for health policy makers and researchers to take effective action to battle the disease.Part 1 The Study on Epidemiological Burden of Diabetes in ChinaOBJECTIVE To analyze the changes in prevalence of diabetes from 1993 to 2003, as well as the health sevice utiliaztion of diabetic patients from 2000 to 2005. To measure the years of life lost (YLL) and the indirect economic burden due to diabetes in 2005. METHODS Trend analysis and the formula used by WHO in estimating global burden of diseases were employed. The calucation of indirect economic burden was based on the formula used by WHO and World Bank. RESULTS From 1993—2003, the prevalence of diabetes in urban and rural area increased by 194.73% and 375%. The health service utilization of diabetic patients rose rapidly. The proportion of inpatients with diabetes in all hospitalized patients raised by 34.75% in city hospitals and 13.43% in county hospitals. The YLLs due to diabetes was 154.82 per 100000 in the urban and 87.96 per 100000 in rural area and 105.75 per 100000 in whole country in 2005. The YLLs in females was more than those in males wherever in urban and rural area. The YLLs in the urban area was about 2 times that in rural area. The indirect economic burden incurred by YLLs was 8.068 billion Yuan. CONCLUSIONS The prevalence of diabetes imposes heavy burden on population health and the indirect economic burden is substantial. More efforts should be taken to prevent and control diabetes. Part2 The Study on Direct Economic Burden of Diabetes in ChinaOBJECTIVE To estimate the direct health expenditure of diabetes in 2005 in China. To evaluate the feasibility of appling the formula proposed by IDF in China. METHODS Matched case-control design was used to calculate the ratio of direct health expenditure between 618 diabetic patients and 618 individuals without diabetes. The matched factors were age, sex, and medical insurance status. The socioeconomic and biological factors were explored by multiple liner regression model and those factors significantly influence the direct health expenditure of diabetic patients were identified. RESULTS Direct health expenditure for diabetic patients per person was 2.47 times that for individuals without diabetes. The direct health expenditure for diabetic patients was 65.568 billion Yuan in 2005, which accounted for 7.57% national health expenditure. The direct health expenditure due to diabetes was 39.022 billion Yuan. The formulae were proved to be reliable. Multiple liner regression analysis showed age, retiring, complications and comorbities, educational level, medical insurance, treatment, smoking, drinking, dyslipidemia, have significant influence on direct health expenditure for diabetic patients.The medical examination revealed only 13.92% patients reached target level of fasting blood glucose concentrations, 38.19% reached target level of HbA1c, 14.08% controlled their blood pressure well, 22.01% controlled their blood cholesterol concentrations well, 49.36% males and 40.78% females were overweight. CONCLUSIONS Diabetes produces considerable direct economic burden. The proportion of direct health expenditure for diabetic patients in national health expenditure reached a high level, which was even larger than in some developed countries. The formulae proposed by IDF could be used in estimating the direct health expenditure due to diabetes although more evidences should be collected to evaluate it throughly. More efforts should be taken to reduce the risk factors of diabetic chronic complications.Part3 The Study on Medical Expense of Inpatients with DiabetesOBJECTIVE To compare the medical expenses for inpatients with diabetes and its 9 main chronic complications. METHODS 2381 inpatients with diabetes and its 9 main chronic complications were extracted from a tertiary hospital in Shanghai. One-way ANOVA was employed to compare the medical expenses for these diseases. Grey correlations analysis was used to indentify those factors influencing medical expense. RESULTS The medical expenses of inpatients with cardiovascular complications (29946.27 yuan) and cerebrovascular complications (17289.30 yuan) and low limbs complications (13918.42 yuan) were significant higher than other inpatients. From 2000-2005, the average length of day of the inpatients was 21 days and declined from 24 days to 15 days. The medical expense per day for inpatients with cardiovascular complications was highest (1225.04 yuan) and the patients without complitations were the lowest (360.37). Expense on pharmaceuticals was the largest component in medical expense (36.50%) and was the most important expense driver. CONCLUSIONS The medical expense on diabetic complications is huge. More efforts should be given to prevent and control these diseases in their early stages; especially the cardiovascular complications; cerebrovascular complications; and low limbs complications.Part4 Case Control Study on the Health Related Quality of Life of Type 2 Diabetic Patients and People without DiabetesOBJECTIVE To compare the health related quality of life (HRQoL) between type 2 diabetic patients and nondiabetic individuals. To explore the factors associated with HRQoL of diabetic patients. To examine the reliability and validity of EQ-5D instrument in measuring the HRQoL of diabetic patients. METHODS Multiple liner regression analysis and binary logistic analysis were used to explore the relationship between HRQoL and characters of patients. RESULTS The results revealed type 2 diabetes had significant adverse effect on HRQoL. Factors including age, sex, number of family members, family income level, size of housing, smoking, diet control, complications, taking exercise and abnormal blood cholesterol concentrations had statistical significance correlations with HRQoL of diabetic patients. CONCLUSIONS Efforts to improve the HRQoL of diabetic patients include managing the complications and blood cholesterol concentrations, promoting smoking cessation and taking exercise, caring for the psychology of solitary and youngers. This suggests EQ-5D is a reliable and valid instrument in measuring the HRQoL of diabetic patients. Howerer, EQ-5D has significant ceiling effect in measuring HRQoL of diabetic patients and nondiabetic individuals.更多还原