不同类型的糖尿病流行病学研究 作者：赵南汉 翻译：Gwen256--DXY Dr. Nam-Han Cho. I am an associate professor of preventive medicine and clinical epidemiologist. Upon my graduation from University of Pittsburgh in 1989, I have been working on gestational diabetes research to identify the risk factors and possibly to prevent the disease. My research is funded from Korean government and we are following approximately 1000 subjects who experienced gestational diabetes during pregnancy. A number of epidemiologic studies are available and have been used for diabetes researches. In this lecture, I will discuss the type of epidemiologic studies that have been used in diabetes research and point out the pros and cons for each study. 赵南汉博士自我介绍： 我是一名预防医学副教授及临床流行病学家。自1989年毕业于匹兹堡大学以来，我一直致力于妊娠期糖尿病的研究，探寻其危险因素以期能够避免这一疾病的发生。我的研究资金由韩国政府提供，目前我们有约1000名在妊娠期间患上糖尿病的受试者。 目前的糖尿病研究有多种流行病学研究方法可供选择。本讲座将讨论不同类型的研究方法及其各自的优缺点。

研究类型取决于： 目的 时间 成本 可行性 Before conducting any diabetes research, the researcher must decide on the purpose, the duration, the cost, and finally the feasibility of the study. If the researcher plans to conduct a pilot study, he does not need to design a complex study but rather a simple one that is cheap, easily accessible to the subjects and can be completed in a short period of time. However, if the researchers are looking for a community based study. The research should expect a high cost, a longer research time and poorer feasibility. 在开始一个糖尿病研究以前，研究者必须明确研究的目的，时间，成本及可行性。如果要进行初步研究，就不必设计得太过复杂，而应该设计一个小成本投入、易被受试者接受并能在较短时间内完成的相对简单一些的研究方案。 但如果要开展的是社区研究，所耗的财力与时间就会相对多一些，而且开展起来难度也更大。

观察性研究（描述性，分析性） 横断面研究 回顾性研究 前瞻性研究 实验性研究 临床试验 社区干预试验 现场试验 研究类型 观察性研究（描述性，分析性） 横断面研究 回顾性研究 前瞻性研究 实验性研究 临床试验 社区干预试验 现场试验 Epidemiologic studies are divided into the two major areas: observational and experimental. Observational studies are usually called either descriptive or analytical. If one plans to conduct an observational study, one does not have control over the study subjects. The researcher simply observes the outcome and factors associated with the disease. On the other hand, an experimental study has full control over the study subjects. This type of study is usually conducted in a clinical, community intervention, and field trial setting. I will be discussing about each study in the latter part of my lecture. 流行病学研究分为两大类：观察性研究与实验性研究，前者通常又称为描述性或分析性研究。观察性研究不需要人为控制研究对象，研究者只需观察疾病的结局和与疾病相关的因素即可。而实验性研究则要求人为控制研究对象，这种研究通常应用于临床、社区或是现场试验中。接下来我将具体介绍各种研究方法。

关于糖尿病结局的研究 患病人群 发病率（并发症） 病死率 费用 治疗 教育 预防与控制 干预 These are the areas in diabetes research where epidemiologic studies are greatly in need. These areas including: Diabetes outcome researches in different type of diabetes. Research on the complications and the mortality. Economic research relates to diabetes care, treatment, diagnosis, and prevention. Research is related to patient education, and finally, research for the intervention and prevention of diabetes. More answers to the question of complication and mortality are still in need and open for answers. 糖尿病研究中有诸多方面需应用流行病学研究方法。这些研究方面包括：不同类型糖尿病的转归研究，并发症与病死率研究，糖尿病护理、治疗、诊断及预防的经济学研究，患者教育研究以及干预与预防的研究。关于并发症与病死率方面的研究还有待更进一步开展。

研究类型与时间的关系 过去 现在 将来 横断面 研究 回顾性研究 前瞻性研究 时间 The concept of time is very important in determining the type of epidemiologic study to be done. If you are looking for the relationship of cause and effect during present time, you will need a cross-sectional study. However, if you are looking at the cause and effect relationship of the past, you will need a retrospective study. If you looking at the relationships in the future, you will need a prospective study. I will be lecturing on each one of these studies more in details presenting with the aid of the next few slides. 时间的概念对于即将开始的流行病学研究类型的选择非常重要。如果要研究的是现在这段时间内疾病的病因因果关系，那么就需要进行横断面研究。但若是要研究过去的病因因果关系，则应进行回顾性研究。而对于将来的病因因果关系，则宜进行前瞻性研究。接下来几张幻灯片中我将更具体地谈谈这几种研究方法。 横断面 研究 回顾性研究 前瞻性研究 时间

横断面研究 现在 The concept of time is very important in determining the type of epidemiologic study to be done. If you are looking for the relationship of cause and effect during present time, you will need a cross-sectional study. However, if you are looking at the cause and effect relationship of the past, you will need a retrospective study. If you looking at the relationships in the future, you will need a prospective study. I will be lecturing on each one of these studies more in details presenting with the aid of the next few slides. 时间的概念对于即将开始的流行病学研究类型的选择非常重要。如果要研究的是现在这段时间内疾病的病因因果关系，那么就需要进行横断面研究。但若是要研究过去的病因因果关系，则应进行回顾性研究。而对于将来的病因因果关系，则宜进行前瞻性研究。接下来几张幻灯片中我将更具体地谈谈这几种研究方法。 横断面 研究 时间

分析性研究 横断面研究： 如同 一次快照 简单易行 可提供病因线索 应用最为广泛的初步研究 Cross-sectional study is considered an analytical study, because it helps to analyze the relationship between the disease and the potential risk factors. It act like a snap shot and is very simple to conduct in a short period of time. However, it only describes the associated factors of the outcome. Thus, this type of study should only be used as the first line of epidemiologic research 横断面研究被认为是一种分析性的研究，因为它有助于分析疾病与可能危险因素之间的关系。如同 一次快照，可以很容易地在短时间内完成。但是这种方法只是对结局相关因素进行描述，所以只能用于流行病学的初步研究。

横断面研究 何处不同 如何不同 疾病（糖尿病） 发病率（并发症） 病死率 是否应用雷帕霉素洗脱支架(SES) 发病年龄 季节 Cross-sectional study is considered an analytical study, because it helps to analyze the relationship between the disease and the potential risk factors. It act like a snap shot and is very simple to conduct in a short period of time. However, it only describes the associated factors of the outcome. Thus, this type of study should only be used as the first line of epidemiologic research 横断面研究被认为是一种分析性的研究，因为它有助于分析疾病与可能危险因素之间的关系。如同 一次快照，在短时间内很容易进行。但是这种方法只能描述相关因素，所以只能用于流行病学的初步研究。

横断面研究 One of the best results from the cross-sectional study has been the exposition of incidence rates. As shown on this slide that incidence rates of IDDM varied widely. The lowest rate was reported from Korea, and the highest from Finland. These differences are as much as 40 fold between the two countries. Based on this cross-sectional study, we immediately noticed that both genetic and environmental conditions should be regarded as risk factors for IDDM. In these cross-sectional studies, we also note that IDDM incidence is lower in Asia, moderate in North America, and higher in Scandinavian countries. 横断面研究用途之一在于其能说明发病率。如图所示，不同国家胰岛素依赖型糖尿病的发病率相差很大。有报道的发病率最低在韩国，而最高则在芬兰，二者之间差距竟高达40倍。基于横断面研究我们很快可以注意到遗传与环境条件都应被视为胰岛素依赖型糖尿病的危险因素。在此研究中，我们还能发现不同地区胰岛素依赖型糖尿病的发病率由低到高依次为亚洲、北美、北欧。

回顾性研究 现在 关键不同之处在于： 生活方式 饮食习惯 生育情况 药物使用 家族遗传 基因 过去 时间 The next type of epidemiologic study is the retrospective study. Based on the concept of the past. It is past, because we know the disease conditions of the subjects at the time of beginning of the study. In this type of study, we usually select both a group of diseased and normal healthy subjects, and try to find out what factors potentially cause people to get a disease. Usually, one interviews the subjects with regards to potential risk factors. Though outcomes is prone to a memory bias, retrospective study is cheaper, obtains results in a short period of time, and is the best method for studying rare diseases. 第二类研究方法是回顾性研究。这是一种以过去的资料为研究基础的方法。在研究开始之初我们已清楚研究对象的发病情况。我们通常选取一组患病的人群与一组正常的健康人群，通过比较试着从中找出潜在的致病原因。研究者通常都会从种种的可能危险因素入手调查。尽管回顾性研究的结果往往会受回忆偏倚的影响，但这种研究更为经济，并能在较短时间内完成，因此最适用于罕见疾病的研究。 过去 时间

回顾性研究的优缺点 优点 缺点 适用于研究罕见疾病 成本低 较少引起伦理争议 耗时短 所需样本量小 研究对象不需受人为控制 易受产生选择偏倚和回忆偏倚 不一致性：症状或疾病的定义有可能发生变化 不能计算发病率 资料有限 不能直接分析相对危险度 This slide further indicates the pros and cons for retrospective studies. 这张幻灯片分析的是回顾性研究的优点与缺点。  

前瞻性研究 现在 糖尿病 将来 危险因素?? 有视网膜病变 无视网膜病变 时间 The last analytical type of epidemiologic study in presented on this slide. It is called a prospective study and the goal is looking at the future outcome. In other words, at present time we know an individual’s potential risk factors. Therefore, the goal is to follow subjects who have been exposed and not been exposed to potential risk factors for a given period and observe the outcome of the disease. This type of study is openly used in diabetes research. DCCT and DPT is a few of the prospective diabetes research we have been hearing about for a long time.  这张幻灯片上显示的是我要介绍的最后一种分析性流行病学研究类型----前瞻性研究,这是一种追踪将来发病结局的研究方法。也就是说,现在我们知道某些因素可能是疾病的危险因素,因此我们的目的就是对暴露或未暴露于这些可能危险因素的研究对象进行随访追踪,观察他们发病的情况。这种研究方法在糖尿病研究领域的应用很广泛。如长期以来我们所知道的DCCT(糖尿病控制和并发症试验)和DPT(糖尿病预防试验)都属于前瞻性的糖尿病学研究。 无视网膜病变 时间

前瞻性研究的优缺点： 优点 缺点 不易受偏倚影响 不需回忆 可计算出发病率 可准确分析相对危险度 疾病与症状的定义可保持一致性 耗时长 仅适用于常见病研究 成本高 易引起伦理争议 失访率高 需要志愿者 样本需求量大 霍桑效应 This slide presents the pros and cons of the prospective study.  这张幻灯片分析的是前瞻性研究的优点与缺点。

研究类型与目的 横断面研究 了解患病情况 预测危险因素 回顾性研究 验证危险因素 前瞻性研究 预防-控制-干预-教育-治疗 I would like to summarize the three types of studies we have covered in this lecture. The goal of the cross-sectional study is a better understanding of the disease and risk factors that are associated with the disease. The retrospective study is good to get an estimation of the potential risk factors and usually these relationships are confirmed by the prospective studies. Therefore, the retrospective study is less valid than the prospective method. The results from these studies have been utilized for diabetes prevention, control, intervention, education and management.  我现在来总结一下刚才介绍的这三种研究方法。横断面研究的目的是为了更好地了解疾病和与疾病相关的危险因素。回顾性研究能够对可能危险因素与疾病进行估测（提出病因假设），而通常最后是由前瞻性研究来验证。因此，回顾性研究的真实性不及前瞻性研究。现在，在糖尿病疾病的预防、控制、干预、教育以及治疗等各个领域中，通过这些流行病学研究得到的结果已经被广为应用。 预防-控制-干预-教育-治疗

实验性研究 动物实验 药物实验 人体实验 Another area of the epidemiologic study is the experimental study. Unlike analytical studies, in experimental studies, the research has control over the study subject’s activity. These type of studies are usually applied in animals, drug trial and human subjects. 另一类流行病学研究是实验性研究。与分析性研究不同，实验性研究对研究对象的活动进行控制。这种研究方法通常可应用于动物实验，药物实验以及人体实验中。 

实验性研究 临床随机对照试验：DPP 结 果 评 随机选择 估 运动试验 饮食试验 药物试验 One of the best experimental studies is the diabetes prevention program which started two years ago. In this experimental study, the subjects were randomly selected from a community and divided to do one of the three trials, ie., exercise, diet or medication. Those high risk subjects assigned to each of the regiment were ask to followed the researcher instruction and the outcome was measured after the study period. The limitations of this time study was the high drop out rates, the high experimental cost, and the question of ethics. 最成功的实验性研究之一就是两年前开展的糖尿病预防项目（DPP）。在这个研究当中，研究对象是从一个社区中随机选出，然后分为三组进行不同方面的实验，即运动、饮食与药物。各组中的高危受试者按照研究者的要求进行实验，研究阶段结束后实验结果将被记录下来。这种研究的局限性在于耗资大以及可能会涉及到伦理问题。   药物试验

实验性研究 交叉设计 结 果 评 估 One of the best experimental studies is the diabetes prevention program which started two years ago. In this experimental study, the subjects were randomly selected from a community and divided to do one of the three trials, ie., exercise, diet or medication. Those high risk subjects assigned to each of the regiment were ask to followed the researcher instruction and the outcome was measured after the study period. The limitations of this time study was the high drop out rates, the high experimental cost, and the question of ethics. 最成功的实验性研究之一就是两年前开展的糖尿病预防项目（DDP）。在这个研究当中，研究对象是从一个社区中随机选出，然后分为三组进行不同方面的实验，即运动、饮食与药物。各组中的高危研究个体按照研究者的要求进行实验，研究阶段结束后实验结果将被记录下来。这种研究的局限性在于耗资大以及可能会涉及到伦理问题。