部分公共卫生类期刊调查研究文献中抽样设计的常见问题分析（一）

目的 分析总结公共卫生类期刊涉及调查研究文献中抽样设计的常见问题及特点,为建立公共卫生类期刊文献中抽样设计的规范体系奠定基础。方法 设计公共卫生类期刊文献调查研究抽样设计常见问题调查表,随机抽取10种公共卫生类核心及以上级别期刊,查阅其2014年发表的所有文献,采用定性研究方法总结分析其中常见抽样方法的问题类型与特点。结果 公共卫生类期刊文献中抽样设计的常见问题主要表现为:（1）分层抽样:未对“层”定义;未说明分层标准;未说明所分层数及每层单位数具体大小;未提供上述各点判定依据;（2）整群抽样:未对“群”定义;未说明分群标准;未说明所分群数及每群单位数具体大小;未提供上述各点判定依据;（3）系统抽样问题:未明确抽样间隔;未明确欲抽样样本组数;（4）多阶段抽样问题:各阶段表达过于简单;单阶段抽样表述不完整;多阶段抽样命名问题。结论 选定公共卫生类期刊文献中,抽样设计问题表现较为集中、特点明显,目前文献中抽样设计的表述尚不规范,有待进一步研究以规范化。     

部分公共卫生类期刊文献中实验性研究对照设计常见问题探讨

目的分析部分公共卫生类期刊文献中实验性研究对照设计常见问题并提出规范建议,为今后的规范运用提供一定参考。方法随机抽取6本公共卫生类期刊杂志,查阅其2014年刊发的所有文献,将其中涉及对照设计的实验性研究分为流行病学实验、动物实验和其他实验研究,总结其对照设计问题类型及特点并提出规范建议。结果流行病学实验研究中对照设计常见问题主要有:对照可比性欠缺、对照样本量不当、对照设置不当、误用自身前后对照和对照信息不完整问题;动物实验研究对照设计常见问题主要有:对照不足、对照不当、误用空白对照和对照组表述问题;其他实验研究对照设计常见问题主要有:对照不足、对照可比性欠缺和对照组表述问题。实验性研究对照设计规范建议为:对照应满足均衡原则、对照组例数与实验组应均衡、对照形式与研究目的应相符、慎用自身前后对照和空白对照、对照组信息应规范完整。结论公共卫生类期刊文献中的对照设计仍存在较多问题,其运用与表述均有待进一步规范和完善。     

部分公共卫生类期刊文献中观察性研究对照设计常见问题探讨

目的分析部分公共卫生类期刊文献中观察性研究对照设计常见问题,为今后的规范运用提供一定参考。方法随机抽取6本公共卫生类期刊杂志,查阅其2014年刊发的所有文献,将涉及对照设计的观察性研究分为病例对照和队列研究,系统总结其对照设计问题类型及特点。结果观察性研究文献中共同存在的对照设计常见问题主要表现为:对照代表性或可比性不足、对照纳入排除标准不完整和对照样本量不当。此外,病例对照研究还存在匹配不当和对照设计无法回答研究问题,而队列研究还存在对照类型不明确问题。观察性研究中对照设计规范建议为:对照应满足代表性和可比性原则、对照组样本量应适当、纳入排除标准应描述完整、匹配应适当、对照设计应能回答研究问题、对照类型应指明。结论公共卫生类期刊文献中观察性研究对照设计仍存在较多问题,其运用与表述均有待进一步规范和完善。     

部分公共卫生类期刊论文中抽样错误的影响因素初探

目的 了解目前公共卫生类期刊抽样方法误用现况,初步探讨导致抽样设计错误的影响因素。 方法 收集整理4种公共卫生期刊杂志2014年01期至12期的全部学术文献,针对常见抽样设计问题,并利用EpiData3.1软件建立数据库,录入所收集文献的基本信息及抽样设计错误。采用SPSS 18.0软件进行统计描述与决策树分析,初步探讨导致抽样设计错误的影响因素。 结果 收集整理文献共1 349篇,经过筛选后得到有效文献260篇。其中抽样设计错误个数的均值为3.40个,错误个数＞2的文献有172篇,占66.5%。单因素得到2个有统计学意义的变量:概率抽样类型、杂志类别（P ＜0.05）,其χ2值分别为11.457、5.403。按概率抽样类型分层后,多因素分析得到2个决策树模型,经过10 层交叉验证后,模型识别总正确率分别为67.4%、73.7%。结论 公共卫生类期刊论文所存在的抽样方法误用问题不容乐观,针对不同类型抽样方法及其误用影响因素应采用不同手段加以改进。     

国内公共卫生类期刊论文中的调查研究抽样设计质量影响因素探讨

目的了解国内公共卫生期刊论文中调查研究的抽样设计质量情况,探讨其可能影响因素。方法随机抽取公共卫生类期刊8种,纳入2014年发表的全部涉及抽样调查的文献。采用抽样调查质量评价量表对每篇文章进行抽样调查质量评分,分别采用Epi Data3.1和SPSS22.0进行数据收集和单因素及分类回归树模型分析。结果共纳入有效文献646篇,总体抽样调查质量评分均值为4.251±2.157。单因素分析结果显示抽样类型、杂志类别和作者所在城市类别的内部分层之间的质量评分差异有统计学意义(P0.001);分类回归树模型分析表明对质量评分影响的因素按贡献性大小依次为杂志类别、抽样类型、作者机构类别和城市类别。结论公共卫生类期刊论文中调查研究抽样设计质量现况总体欠佳,核心期刊类杂志可进一步加强抽样设计方面的质量把关,研究者尤其是非研究机构类研究者应加强自身对各类抽样设计的掌握水平。     

公共卫生类核心期刊论文常见单因素统计推断问题调查与分析

目的 了解公共卫生领域核心期刊论文的单因素统计推断问题的发生情况，提出对策建议，为降低文章单因素统计推断错误发生率，提高文章质量提供参考。方法 采用目的抽样选取四种公共卫生领域期刊，检索其2019年度刊登的全部文章，对文章逐篇阅读并记录、分析其单因素统计推断应用情况。结果 本次研究共调查2023篇文章，最终纳入1139篇文章，出现单因素统计推断错误的文章共53篇，单因素统计推断错误率达4.65%，主要错误类型为资料不满足正态性和（或）方差齐性误用t检验或方差分析、理论频数太小误用x2检验、重复测量资料误用成组t检验等。结论 公共卫生类期刊论文单因素统计推断应用正确率有待提高，研究者、编辑、审稿专家及期刊编辑部应加强统计学知识学习，把握论文发表各环节，以促进公共卫生领域期刊论文质量的提高。     

公共卫生与预防医学类期刊热点和前沿研究

目的 通过Citespace软件分析国内公共卫生与预防医学类期刊的研究热点、前沿与知识基础研究,洞悉国内预防医学领域研究动向。方法 运用Citespace对从CNKI、万方、维普等数据库中收集的近五年（2013 - 2017年）中文核心期刊中预防医学、卫生学类核心期刊所载文献进行分析。按照中国学术期刊影响因子年报中公共卫生与预防医学领域期刊分类方式,将上述中文核心期刊分为两类,即医药卫生事业管理和预防医学与卫生学。结果 医药卫生事业管理类近五年的研究热点及前沿主要集中在基层医疗机构、医疗保险、医疗相关政策中,且研究更注重卫生服务效果和评价等方面。预防医学与卫生学类近五年的研究热点及前沿主要集中在医院感染、影响因素、艾滋病等各类传染病的流行病学特征及耐药性方面。结论 在公共卫生与预防医学领域中文核心期刊中,预防医学与卫生学类更注重艾滋病、医院感染等方面,医药卫生事业类主要集中在公立医院及满意度方面。     

基于共词分析的公共卫生领域研究热点分析

目的探索全球公共卫生领域的研究热点,并基于公共卫生专业知识解读聚类结果。方法对公共卫生领域影响因子前10%期刊2014-2015年所发表文献中的高频主题词进行共词分析和聚类分析。结果共检索到文献5 806篇,获得高频主题词95个,基于聚类分析结果得出7个研究热点,包括空气污染与环境监测、健康状况的社会人口学影响因素、慢性病的流行病学调查与健康管理、健康行为相关研究、卫生政策与管理研究、流行病学调查设计与统计分析模型、以及控烟相关研究。结论研究结果较为客观地反映了当前公共卫生领域的研究热点,为科研人员进行公共卫生研究提供一些思路。     

复杂抽样数据统计分析方法回顾

<正>当今社会科学与健康科学调查研究,尤其是大规模调查,往往涉及多地区或多中心的抽样问题,采取单纯随机抽样选择样本因调查对象过于分散,成本高,可行性低[1],调查设计者更倾向于可行性较高的复杂抽样,但其通常使样本结构复杂化。若采用忽略抽样特征的传统统计学方法分析此类数据,会导致标准误的低估,进而低估可信区间,且增大犯I类错误的可能性,最终导致偏倚甚至得到错误的统计推断[2]。目前,对于复杂抽样数据的统计分析主要分为基于设计     

2013 - 2017年广州市突发公共卫生事件相关文献分析

目的 分析广州市突发公共卫生事件相关文献的特点,为有效应对突发公共卫生事件提供依据。方法 通过CNKI等数据库系统检索2013 - 2017年广州市突发公共卫生事件的文献,并应用归纳法、比较分析法对所检索文献进行整理分析。结果 2013 - 2017年发表的广州市各类突发公共卫生事件文献共223篇,位于前三位的是传染病(71.75%)、食物中毒(8.52%)、环境卫生（8.07%）。广州市突发公共卫生事件文献以流行病学分析研究最多,占文献总数的37.67%（84/223）。发表在北大中文核心期刊上的文献占总篇数的30.94%（69/223）。结论 2013 - 2017年广州市突发公共卫生事件种类多,以传染病为主。研究的时效性强,涉及的学科范围较广,对广州市突发公共卫生事件的防控有所裨益。     

国家人体生物监测项目抽样方法及误差估计

目的 探讨国家人体生物监测项目抽样方法,并通过计算抽样误差来评估抽样设计。方法 该监测采用多阶段复杂抽样方法,并以贵州省抽样结果为例,应用多阶段不等概率抽样误差计算方法计算贵州省抽样误差及变异系数,探讨多阶段抽样设计下误差估计方法。结果 该监测覆盖全国31个省（自治区、直辖市）的152个监测点,抽取21 888名3~79岁城乡居民,三个阶段置换率分别为5.26%、6.35%及40.6%。经计算贵州省抽样误差为3 207 594人,变异系数为0.097。结论 根据多阶段不等概率抽样方法计算,贵州省抽样变异系数较小,精度较高,但该计算方法未考虑人群缺失率、应答率等非抽样误差的权重调整,大型公共卫生监测项目多阶段抽样误差计算方法有待继续研究。     

Survey research methods in evaluation and case-control studies

Survey research methods are widely used in two types of analytic studies: evaluation studies that measure the effects of interventions; and population-based case-control studies that investigate the effects of various risk factors on the presence of disease. This paper provides a broad overview of some design and analysis issues related to such studies, illustrated with examples.The lack of random assignment to treatment and control groups in many evaluation studies makes controlling for confounders critically important. Confounder control can be achieved by matching in the design and by various alternative methods in the analysis. One popular analytic method of controlling for confounders is propensity scoring, which bears a close resemblance to survey weighting.The use of population-based controls has become common in case-control studies. For reasons of cost, population-based controls are often identified by telephone surveys using random digit dialling (RDD) sampling methods. However, RDD surveys are now experiencing serious problems with response rates. A recent alternative approach is to select controls from frames such as driver license lists that contain valuable demographic information for use in matching.Methods of analysis developed in the survey sampling literature are applicable, at least to some degree, in the analyses of evaluation and population-based case-control studies. In particular, the effects of complex sample designs can be taken into account using survey sampling variance estimation methods. Several survey analysis software packages are available for carrying out the computations.     

Comparison of tests for categorical data from a stratified cluster randomized trial

Two features commonly exhibited by randomized trials of health promotion interventions are cluster randomization and stratification. Ignoring correlations between individuals within clusters can lead to an inflated type I error rate and hence a P-value which overstates the significance of the result. This paper compares several methods for analysing categorical data from a stratified cluster randomized trial. We propose an extension of a method from survey sampling that uses the design effect to reduce the effective sample size. We compare this with three methods from Zhang and Boos that extend the standard Cochran-Mantel-Haenszel (CMH) statistic by using appropriate covariance matrices, and with a bootstrap method. The comparison is based on empirical type I error rates from a simulation study, in which the number of clusters randomized is small, as in most public health intervention studies. The method that performs consistently well is one of the Zhang and Boos extensions of the standard CMH statistic.     

Identifying research priorities for public health research to address health inequalities: use of Delphi-like survey methods

Background

In the funding of health research and public health research it is vital that research questions posed are important and that funded research meets a research need or a gap in evidence. Many methods are used in the identification of research priorities, however, these can be resource intensive, costly and logistically challenging. Identifying such research priorities can be particularly challenging for complex public health problems as there is a need to consult a number of experts across disciplines and with a range of expertise. This study investigated the use of Delphi-like survey methods in identifying important research priorities relating to health inequalities and framing tractable research questions for topic areas identified.

Methods

The study was conducted in two phases, both using Delphi-like survey methods. Firstly, public health professionals with an interest in health inequalities were asked to identify research priorities. Secondly academic researchers were asked to frame tractable research questions relating to the priorities identified. These research priorities identified using Delphi-like survey methods were subsequently compared to those identified using different methods.

Results

A total of 52 public health professionals and 21 academics across the United Kingdom agreed to take part. The response rates were high, from public health professionals across three survey rounds (69%, 50% and 40%) and from academics across one round (52%), indicating that participants were receptive to the method and motivated to respond. The themes identified as encompassing the most important research priorities were mental health, healthy environment and health behaviours. Within these themes, the topic areas that emerged most strongly included community interventions for prevention of mental health problems and the food and alcohol environment. Some responses received from academic researchers were (as requested) in the form of tractable research questions, whereas others contributed further potential topic areas instead.

Conclusions

Delphi-like survey methods are practical and productive as a means of obtaining opinions from a wide number of relevant experts identifying potential priority topic areas for research; however, this method is less appropriate for framing tractable research questions.

     

华北5城市民众SARS疫情中心理行为研究

目的 考察华北5城市民众在SARS疫情中的风险认知特征。探讨风险认知对应对行为、心理健康水平的影响作用。建立社会心理预警模型，为政府决策和舆论导向提供心理学依据和对策建议。方法 采用分层抽样和方便抽样相结合的方法，对北京、天津、石家庄、呼和浩特和太原市市民1500人进行问卷调查。结果 (1)空间距离近的疫情信息更能引起民众的关注，北京市民对SARS患病信息和治愈信息敏感性显高于邻近城市。(2)女性表现出更高的风险感，政府公众信息更能引起中、老年市民的关注，20岁以下的青少年更重视与自己学习、生活相关的信息。50岁以上民众对于“愈后对于身体有无影响”等信息的警觉性显高于其他年龄组。(3)与北京邻近的城市市民对SARS疫情的风险评估明显低于北京市民。某些地区大学生的心理问题值得关注。结论 在突发事件中，通过风险认知特征调查。可以了解人们的风险认知差别及其对不同人群的应对行为、心理健康水平的影响；今后应加强公共卫生突发事件中民众社会心理行为的预测研究。     

Comparing sexual minority status across sampling methods and populations

The health of sexual minority women (SMW) has recently received research attention. Previous research into the health of SMW (e.g., lesbians, bisexuals, transgendered women) used a mixture of sampling methods, many of which were poorly documented and difficult to understand. The results of these previous studies do not present a consistent pattern of findings, possibly due to differences in sampling methods. The present study compared the characteristics ofSMWacross four survey sampling methods, three in the same geographic area. Differences were found among groups of SMW by sampling method, including in demographic data (e.g., level of education) and personal health data (e.g., rates of regular mammography screening). These findings provided a possible explanation for the variety of findings in the published literature and identified rigorous sampling methods that can be used in future research.     

A 10-Year Content Analysis of Original Research Articles Published in Health Communication and Journal of Health Communication (2000–2009)

This study presents data from a content analysis of original research articles published in Health Communication and Journal of Health Communication from 2000 to 2009. The authors coded 776 articles using categories that identified health topics, theory, population characteristics, and methods used in each study. Distinctions between the published research in Health Communication and Journal of Health Communication are highlighted. Across both journals, findings demonstrated articles sometimes lack racial demographic information, primarily perform research in the United States, rely heavily on survey data, and often lack a theoretical framework. The top physical health topic addressed across both journals was cancer, and the top non–physical health topic addressed was the role of media in health. Journals displayed differences in several areas and those differences often mirrored each journal's stated objectives. Results are discussed in terms of their implications for expanding health communication research to be reflective of issues salient to public health within the United States and around the world.     

Does citizen science have the capacity to transform population health science?

Citizen science engages members of the public in research design, data collection, and analysis – in asking and answering questions about the world around them. The United States, European Union, and Australia have placed citizen science at the forefront of national science policy. Journals such as Science, Nature and Bioscience regularly feature projects conducted by citizens. Citizen science engages millions of people worldwide. However, to date, population health science has not relied heavily on citizen contributions. Although community-based participatory action research remains a strong foundational method to engage those affected by public health problems, there is additional potential to mainstream population health through wider, less intensive opportunities to be involved in our science. If we are to tackle the complex challenges that face population health then new avenues are needed to capture the energy and attention of citizens who may not feel affected by public health problems, i.e. to engage the ‘by-standers’ in population health science. Particular types of citizen science methods have the potential to do this. But simply increasing the breadth and volume of scientific evidence will not be enough. Complex, intractable, macro-level problems in population health require change in how our journals and funding bodies respond to data generated by the public. Of course, democratisation of science and the potential decentralisation of scientific authority will bring deep challenges. But potentially it brings a future where population health science is better known, understood and respected, with benefits for the types of public policies that derive from this science.