关注今日：14 | 主题：145580
微信扫一扫
扫一扫，下载丁香园 App
即送15丁当
【求助】预后相关meta分析小修－－三个纠结的问题
页码直达：
丁香园准中级站友
这个帖子发布于7年零343天前，其中的信息可能已发生改变或有所发展。
开始接触循征医学是在07年，也是从丁香园循征版一点一滴，慢慢积累。写过两遍中文的meta分析，都顺利发表。去年完成一篇英文meta分析文章：关于一生物因子在肿瘤中的预后作用，两次投稿被拒。月初被一专科杂志（IF：1.4）小修，第一篇SCI文章，虽然还是小修改，但还是兴奋了很久。两个reviewers提了约20个问题，其中有3个问题比较纠结，都是统计学专业方面的问题（本人临床医生），最近几天被搞得晕头转向的。比如质量评估的方法，我参考的是很多文章提供的方法，也是比较权威的方法，只是年代比较久远了。可他给我提供了另外的一种比较新的方法，而且他本人也说对于预后研究的质量评价目前还没有定论。所以，我就不知道要不要根据他提供的质量评估重新评估了。担心回答的不恰当，被拒。会很遗憾。关于预后相关meta分析，讨论的比较少，方法学也不是很统一，此前我有专门开贴讨论过预后相关的问题。我感觉，此次有一reviewer为循征医学的专家（引用了很多专业文献给我参考）。此次，希望，借用我的一篇文章的小修机会，对于预后的meta分析能够有个比较深入的讨论。下面是一个reviewer提出的问题：质量评估的问题：&Data extraction and quality assessment& section: Steel et al devised their own criteria which are not reported in detail in this manuscript. There are no widely agreed quality criteria for assessing prognostic studies but I would recommend to use Hayden's (Ann Intern Med : 427-437) or Altman's (BMJ -228) criteria, which are more clear and comprehensive. Scores should be avoided.&Quality assessment& section: scores are not explicit. You should present comprehensible criteria (see Hayden or Altman's criteria) in the text and in table 2: for instance, the proportion of studies with adequate follow-up.异质性分析统计问题：
&Statistical method& section: the assessment of inconsistency was performed using Cochran's Q statistic - it should be specified-. This statistic is known to depend on the number of trials, and the associated test lacks power. You should also report the I? statistic with confidence interval (see the online Cochrane handbook for instance). The subgroup analyses must be mentioned in this section. You must specify how comparisons were performed (Asian vs non A IHC vs RT-PCR; disease stages)发表偏移的问题：
&Publication bias& section: concerning the overall analysis, Begg's and Egger's tests suggest asymmetry in the funnel plot (p=0.06 and 0.10, respectively). I would not exclude a possible publication bias. Your statement must be corrected. I would recommend using enhanced funnel plot to assess whether the asymmetry is associated to non-significant results (Peters J, Sutton AJ, Jones DR, Abrams KR, Rushton L: Contour-enhanced meta-analysis funnel plots help distinguish publication bias from other causes of asymmetry. J Clin Epidemiol 1-996).
不知道邀请谁？试试他们
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
giggle207 edited on
收起全部有料回复
质量评估，这个不需要担心，质量评价本来就没一个统一的标准，有文章用就可以异质性分析统计， 按照审稿人的意见做亚组分析，看亚组间的异质性就是了发表偏移的问题， 按照审稿人的意见做就是小修的文章，加油:)giggle207 wrote:质量评估的问题：
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
3466599 edited on
谢谢lz的经验分享。关于质量评估，我用的是2006年的REMARK，这个指南的意思和诊断实验的STARD一样，是提高生物标记物预后研究的方法学质量的。我看了很多的诊断准确性meta都将STARD当作质量评分，所以我采用了这个标准。而且记得handbook上有句话说并不规定用什么方法评价质量，甚至可自行设计量表。不知我说的对吗，请战友指正。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
关于预后相关meta分析，讨论的比较少，方法学也不是很统一那什么的meta分析讨论才多，方法才统一。。。。。。。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
回wuzhang2050 兄小弟愚见，可能是我表述有误，我指的是文献的质量评价方面还没有统一的标准。关于RCT的meta分析，拥有比较经典的方法学比如JADAD评分，还有Cochrane的推出的专门统计学软件RevM。只是对于预后的研究，大部分都是回顾性的非随机的对照研究此类研究的质量很难有一个统一的评分标准。对于肺癌研究领域，我比较熟悉的是2001年欧洲肺癌研究协作组推出的生物因子预后研究的评分标准：Steels’s method [9]. The overall score evaluated several dimensions of the methodology, grouped into four main categories: t the description of the methods used to identify an abnormality of the VEGF; the generalizab the analysis of the study data.9.Steels E, Paesmans M, Berghmans T, et al. Role of p53 as a prognostic factor for survival in lung cancer:a systematic review of the literature with a meta-analysis. Eur Respir J 5–719有兴趣的战友可以下载此文献看看关于预后研究的质量评估，06年以来有新的评价标准，也就是那个reviewer提出的Hayden's (Ann Intern Med : 427-437) or Altman's (BMJ -228) criteria可以PM我，我可以提供文献欢迎大家深入讨论
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
giggle207 edited on
质量评估，这个不需要担心，质量评价本来就没一个统一的标准，有文章用就可以异质性分析统计， 按照审稿人的意见做亚组分析，看亚组间的异质性就是了发表偏移的问题， 按照审稿人的意见做就是小修的文章，加油:)giggle207 wrote:质量评估的问题：
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
3466599 edited on
谢谢lz的经验分享。关于质量评估，我用的是2006年的REMARK，这个指南的意思和诊断实验的STARD一样，是提高生物标记物预后研究的方法学质量的。我看了很多的诊断准确性meta都将STARD当作质量评分，所以我采用了这个标准。而且记得handbook上有句话说并不规定用什么方法评价质量，甚至可自行设计量表。不知我说的对吗，请战友指正。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
Altman's (BMJ -228) 这篇文章刚开始做meta分析的时候看了好几遍，觉得他更多的是提出了prognostic test这类meta分析的困境。而这种困境并不是通过统计学方法能解决的，最好的办法其实是联系作者做IPD meta analysis. 说的不好,请指教.
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
就是没有固定方法才好。。。随便你怎么说。。。只要有理。。。另外，小修的怎么搞那么麻烦。。。我看到的小修都是一些简单的计算啊。。。似有似无的叫我改英文啊。。。还有就是回答一些明显已经不可能改变的问题。。。我想编辑无非是想要个答案忽悠一下其他人。。。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
哎，可能是我的第一篇SCI文章小修吧，自己总是战战兢兢的感觉。是啊，虽然是小修，但是两个审稿人一共提了20几个问题，有些还很复杂。比如，质量评价的问题，我就想引用已经发表的高质量文献，来驳审稿人了但是，对于publication bias的问题，他提出了contour-enhanced funnel plot的方法，我查了很多文献，都没有提供这种方法的统计软件，不知道怎样去做这样的funnel plot,唉，很纠结，很烦。。。。。有高手知道如何用stata软件做contour-enhanced funnel plot吗？？？因为，我看到Petesr JM的文章中最后一段话是这样说的：We believe the contour-enhanced funnel plot is a significantadvancement in methods to address publication bias in meta-analyses of comparative effects, as it can aid assessment of publication bias due to statistical significance.Because it can enhance interpretation of a funnel plot, it should be used routinely for assessing possible publication bias (and complimentary to existing methods). We encourage its implementation in common statistical softwareused to carry out meta-analysis, and a Stata macro is available from the corresponding author to construct suchplots.万分感谢了！！
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
giggle207 edited on
看过了没？
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
根据yinjl123战友的提示，我在我的stata里面search keywords：funnel plot发现我的stata里面已经安装了这个package可是，Contour enhanced funnel plots的命令是什么呢？？？？？----------------------------------------------------------------------------------------------------------------------------------------------package installation-----------------------------------------------------------------------------------------------------------------------------------------------package name:
gr0033.pkg
checking gr0033 consistency and verifying not already installed...all files already exist and are up to date.-----------------------------------------------------------------------------------------------------------------------------------------------(click here to return to the previous screen)--------------------------------------------------------------------------------package gr0033 from --------------------------------------------------------------------------------TITLE
SJ8-2 gr0033.
Contour enhanced funnel plots for...DESCRIPTION/AUTHOR(S)
Contour enhanced funnel plots for meta-analysis
by Tom M. Palmer, Department of Health Sciences,
University of Leicester, Leicester, UK
Jaime L. Peters, School of Mathematical Sciences,
Queensland University of Technology, Brisbane, Australia
Alex J. Sutton, Department of Health Sciences,
University of Leicester, Leicester, UK
Santiago G. Moreno, Department of Health Sciences,
University of Leicester, Leicester, UK
tmp8@le.ac.uk
After installation, type help confunnelINSTALLATION FILES
(click here to install)
gr0033/confunnel.ado
gr0033/confunnel.hlpANCILLARY FILES
(click here to get)
gr0033/confunnel_sjanalysis.do
gr0033/magnesium.dta
gr0033/passivesmoking.dta--------------------------------------------------------------------------------(click here to return to the previous screen)
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
Tom M. Palmer， Jaime L. Peters【题名】:
Contour–enhanced funnel plots for meta–analysis【期刊名，年份，卷（期），起止页码】: The Stata Journal. 2008，Volume 8 Number 2: pp. 242-254找到一篇可能有用的。查不到全文， 不知道谁可以帮忙，我尽力了啊
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
3466599 edited on
丁香园中级站友
3466599 【作者】:
Tom M. Palmer， Jaime L. Peters【题名】:
Contour–enhanced funnel plots for meta–analysis【期刊名，年份，卷（期），起止页码】: The Stata Journal. 2008，Volume 8 Number 2: pp. 242-254找到一篇可能有用的。查不到全文， 不知道谁可以帮忙，我尽力了啊
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
要付费的........郁闷中。。。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园中级站友
giggle207 要付费的........郁闷中。。。去绿色通道求助下。。。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
关于预后分析的质量评价我想argue一下因为，我做的是肺癌相关的，对于肺癌已经有比较权威的组织提出了专门的预后分析的质量评价。可以参考steel的文献。However, as you see, there are no widely agreed quality criteria for assessing prognostic studies. We appraised the relational prognostic studies according to steel et al ’ study which designed the methodological scale for biological prognostic factors specially for lung cancer on behalf of European Lung Cancer Working Party.Several systematic reviews with meta-analysis on prognostic studies adopted the steel’s criteria to assess the quality of studies.Thank you for your suggestions. We had read the Hayden or Altman's paper which you recommended carefully. Hayden or Altman's paper showed that adequate quality assessment should include judgments about 6 areas of potential study biases including study participation, study attrition, measurement of prognostic factors, measurement of and controlling for confounding variables, measurement of outcomes, and analysis approaches. Steels’s method used in our study included four main dimensions of the methodology related to the scientific design, the description of the methods used to identify an abnormality of the XXX, the generalizability of the results, the analysis of the study data. Meanwhile, we made subgroup analysis according to the method of XXX expression assessment (IHC, qRT-PCR and ELISA), histology (adenocarcinoma versus all types), XXX isoform (all versus XXX), and region (Asian vs. non-Asian studies) in order to reduce the bias of our study. In addition, we had considered the question of follow-up time at the beginning of selecting the studies, the eligible studies included the meta-analysis must meet a criteria that the follow up time must exceed 5 years, since the 5 year survival rate was the most commonly presentation of prognostic factor for NSCLC.Hayden JA, Cote P, Bombardier CE. Valuation of the Quality of Prognosis Studies in Systematic Reviews. Ann Intern Med. -437.Altman DG. Systematic reviews of evaluations of prognostic variables. BMJ 4–8Delmotte P, Martin B, Paesmans M,et al.(2002) [VEGF and survival of patients with lung cancer: a systematic literature review and metaanalysis]. Rev Mal Respir 19: 577–584Meert AP, Martin B, Delmotte P, et al.(2002a) The role of EGFR expression on patient survival in lung cancer: a systematic review with meta-analysis. Eur Respir J 20: 975–981Meert AP, Paesmans M, Martin B, et al.(2002b) The role of microvessel density on the survival of patients with lung cancer: a systematic review of the literature with meta-analysis. Br J Cancer 87: 694–701Meert AP, Martin B, Paesmans M, et al. (2003) The role of HER-2/neu expression on the survival of patients with lung cancer: a systematic review of the literature. Br J Cancer 89: 959–965Martin B, Paesmans M, Berghmans T, et al. (2003) Role of Bcl-2 as a prognostic factor for survival in lung cancer: a systematic review of the literature with meta-analysis. Br J Cancer 89: 55–64Martin B, Paesmans M, Mascaux C, et al. (2004) Ki-67 expression and patients survival in lung cancer: systematic review of the literature with meta-analysis. Br J Cancer 91: Mascaux C, Iannino N, Martin B, et al. (2005a) The role of RAS oncogene in survival of patients with lung cancer: a systematic review of the literature with meta-analysis. Br J Cancer 92:131–1399.Steels E, Paesmans M, Berghmans T, et al. Role of p53 as a prognostic factor for survival in lung cancer:a systematic review of the literature with a meta-analysis. Eur Respir J –719
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
Hayden's 教授确实是预后方面的大牛
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
楼主，恭喜！等待你的好消息！
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
TO:zhe992000正在努力中，有好消息，一定通知大家：）谢谢coss9989战友提醒，已经于绿色通道求到stata journal上面的那篇文献等我弄明白以后，再和大家交流:)
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
我也正在弄，希望比你快。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
呵呵，今天也等到我的小修的文章（1.3分），忙完再好好研究这个了
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
问题已基本解决，修改稿已经投出，焦急的等待中。。。对于，contour-enhanced funnel plot的运算。首先得保证stata软件中存在confunnel命令模块，可以去stata官方网站下载其运行的命令如下：gen logES=log(HR)generate selogES=(log(ul)-log(ll))/2*1.96------苦思冥想好多天，最后由热心站友提供此公式confunnel logES selogES就能得出contour-enhanced funnel plot参考文献见：【作者】: Tom M. Palmer， Jaime L. Peters【题名】: Contour–enhanced funnel plots for meta–analysis【期刊名，年份，卷（期），起止页码】: The Stata Journal. 2008，Volume 8 Number 2: pp. 242-254praying, amen.!!God bless me!!!!
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园荣誉版主
感谢你的有始有终。愿循证版记录每个人前进的点滴，汇聚成大家共同进步的大海。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
汇报一下修改稿投出去第二天直接接受Dear Dr.
,I am pleased to inform you that your work has now been accepted for publication in Journal of Thoracic Oncology. We note that the figures have been submitted as .doc files. The publisher requires digital artwork to be submitted as either .tif or .eps files with a resolution of at least 300 dpi. You may send these as email attachments to this email address.Upon receipt of the appropriate figures, all manuscript materials will be forwarded immediately to the production staff for placement in an upcoming issue.Thank you for submitting your interesting and important work to the journal.
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
这么老的帖子又别翻出来了，说明大家对预后分析的meta热情不减啊。
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
恭喜战友！
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
内牛满面…mark。。还在挣扎还在写…做出来的发表偏移是有的。。愁怎么处理中………
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
战友们好，小弟刚开始学习循证医学，想作关于肿瘤评分的系统评价，比较评分A和评分B对某肿瘤的生存期评价，评价哪个评分更准确，纳入了11篇的文章，一些文章中给出了预生存期和实际生存期方面的信息，现在疑惑的是要采用什么样的方法来评价，是用预后方面的系统评价还是用诊断性试验方面的系统评价法，感觉都不恰当，到底应该用哪种方法现在有点茫然，请版主指明方向，提供恰当的评价方法，还有需要提取文献中的哪些信息，不胜感激！
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
丁香园准中级站友
薛无涯 战友们好，小弟刚开始学习循证医学，想作关于肿瘤评分的系统评价，比较评分A和评分B对某肿瘤的生存期评价，评价哪个评分更准确，纳入了11篇的文章，一些文章中给出了预生存期和实际生存期方面的信息，现在疑惑的是要采用什么样的方法来评价，是用预后方面的系统评价还是用诊断性试验方面的系统评价法，感觉都不恰当，到底应该用哪种方法现在有点茫然，请版主指明方向，提供恰当的评价方法，还有需要提取文献中的哪些信息，不胜感激！抱歉这个我也不是太懂，我只做过干预和预后的。感觉你这个不适合做预后，因为预后是比较A和B哪个生存更好，而对于 你这个来说，A生存更好并不能说明A跟生存期的相关性更好。所以我更偏向于诊断性，不过我也不知道合不合适，如果原始文献将评分与生存好坏的相关性转换成类似精确性、敏感性这样的百分数，应该可以。譬如A跟生存的相关性为70%，B为80%这样的。 建议发帖请教园子中的其它战友或再看看诊断性SR，抱歉
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
giggle207 问题已基本解决，修改稿已经投出，焦急的等待中。。。对于，contour-enhanced funnel plot的运算。首先得保证stata软件中存在confunnel命令模块，可以去stata官方网站下载其运行的命令如下：gen logES=log(HR)generate selogES=(log(ul)-log(ll))/2*1.96------苦思冥想好多天，最后由热心站友提供此公式confunnel logES selogES就能得出contour-enhanced funnel plot参考文献见：【作者】: Tom M. Palmer， Jaime L. Peters【题名】: Contour–enhanced funnel plots for meta–analysis【期刊名，年份，卷（期），起止页码】: The Stata Journal. 2008，Volume 8 Number 2: pp. 242-254praying, amen.!!God bless me!!!!版主请教一下，我也碰见这个问题，然后做出来contour-enhanced funnel plot，能帮忙解读一下这个图形吗？
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
scsuzhou edited on
好帖子纠结了好久用哪个评分标准的问题，终于得到解答了
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
giggle207 开始接触循征医学是在07年，也是从丁香园循征版一点一滴，慢慢积累。写过两遍中文的meta分析，都顺利发表。去年完成一篇英文meta分析文章：关于一生物因子在肿瘤中的预后作用，两次投稿被拒。月初被一专科杂志（IF：1.4）小修，第一篇SCI文章，虽然还是小修改，但还是兴奋了很久。两个reviewers提了约20个问题，其中有3个问题比较纠结，都是统计学专业方面的问题（本人临床医生），最近几天被搞得晕头转向的。比如质量评估的方法，我参考的是很多文章提供的方法，也是比较权威的方法，只是年代比较久远了。可他给我提供了另外的一种比较新的方法，而且他本人也说对于预后研究的质量评价目前还没有定论。所以，我就不知道要不要根据他提供的质量评估重新评估了。担心回答的不恰当，被拒。会很遗憾。关于预后相关meta分析，讨论的比较少，方法学也不是很统一，此前我有专门开贴讨论过预后相关的问题。我感觉，此次有一reviewer为循征医学的专家（引用了很多专业文献给我参考）。此次，希望，借用我的一篇文章的小修机会，对于预后的meta分析能够有个比较深入的讨论。下面是一个reviewer提出的问题：质量评估的问题：&Data extraction and quality assessment& section: Steel et al devised their own criteria which are not reported in detail in this manuscript. There are no widely agreed quality criteria for assessing prognostic studies but I would recommend to use Hayden's (Ann Intern Med : 427-437) or Altman's (BMJ -228) criteria, which are more clear and comprehensive. Scores should be avoided.&Quality assessment& section: scores are not explicit. You should present comprehensible criteria (see Hayden or Altman's criteria) in the text and in table 2: for instance, the proportion of studies with adequate follow-up.异质性分析统计问题：
&Statistical method& section: the assessment of inconsistency was performed using Cochran's Q statistic - it should be specified-. This statistic is known to depend on the number of trials, and the associated test lacks power. You should also report the I? statistic with confidence interval (see the online Cochrane handbook for instance). The subgroup analyses must be mentioned in this section. You must specify how comparisons were performed (Asian vs non A IHC vs RT-PCR; disease stages)发表偏移的问题：
&Publication bias& section: concerning the overall analysis, Begg's and Egger's tests suggest asymmetry in the funnel plot (p=0.06 and 0.10, respectively). I would not exclude a possible publication bias. Your statement must be corrected. I would recommend using enhanced funnel plot to assess whether the asymmetry is associated to non-significant results (Peters J, Sutton AJ, Jones DR, Abrams KR, Rushton L: Contour-enhanced meta-analysis funnel plots help distinguish publication bias from other causes of asymmetry. J Clin Epidemiol 1-996). 我最近也在做肺癌预后的meta，有个问题一直没弄明白。Steels在文章&role a=&& as=&& cancer:a=&& factor=&& for=&& in=&& literature=&& lung=&& meta-analysis=&& of=&& p53=&& prognostic=&& review=&& survival=&& systematic=&& the=&& with=&&&中使用了“European LungCancer Working Party quality scale for biological prognostic factors for lung cancer”这个评价标准。文章中提到：总分为40分，每个部分总分10分。但是附录中该评价标准共包含了22个项目，每个项目为2分，总分应当为44分。我看到好多meta分析都是引用的这个评价标准，而且也说明了总分为40分。我不太明白应该怎么计算得分，麻烦giggle兄指点一下。&/role&
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
受益匪浅！
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
肥猪无排骨 giggle207 开始接触循征医学是在07年，也是从丁香园循征版一点一滴，慢慢积累。写过两遍中文的meta分析，都顺利发表。去年完成一篇英文meta分析文章：关于一生物因子在肿瘤中的预后作用，两次投稿被拒。月初被一专科杂志（IF：1.4）小修，第一篇SCI文章，虽然还是小修改，但还是兴奋了很久。两个reviewers提了约20个问题，其中有3个问题比较纠结，都是统计学专业方面的问题（本人临床医生），最近几天被搞得晕头转向的。比如质量评估的方法，我参考的是很多文章提供的方法，也是比较权威的方法，只是年代比较久远了。可他给我提供了另外的一种比较新的方法，而且他本人也说对于预后研究的质量评价目前还没有定论。所以，我就不知道要不要根据他提供的质量评估重新评估了。担心回答的不恰当，被拒。会很遗憾。关于预后相关meta分析，讨论的比较少，方法学也不是很统一，此前我有专门开贴讨论过预后相关的问题。我感觉，此次有一reviewer为循征医学的专家（引用了很多专业文献给我参考）。此次，希望，借用我的一篇文章的小修机会，对于预后的meta分析能够有个比较深入的讨论。下面是一个reviewer提出的问题：质量评估的问题：&Data extraction and quality assessment& section: Steel et al devised their own criteria which are not reported in detail in this manuscript. There are no widely agreed quality criteria for assessing prognostic studies but I would recommend to use Hayden's (Ann Intern Med : 427-437) or Altman's (BMJ -228) criteria, which are more clear and comprehensive. Scores should be avoided.&Quality assessment& section: scores are not explicit. You should present comprehensible criteria (see Hayden or Altman's criteria) in the text and in table 2: for instance, the proportion of studies with adequate follow-up.异质性分析统计问题：
&Statistical method& section: the assessment of inconsistency was performed using Cochran's Q statistic - it should be specified-. This statistic is known to depend on the number of trials, and the associated test lacks power. You should also report the I? statistic with confidence interval (see the online Cochrane handbook for instance). The subgroup analyses must be mentioned in this section. You must specify how comparisons were performed (Asian vs non A IHC vs RT-PCR; disease stages)发表偏移的问题：
&Publication bias& section: concerning the overall analysis, Begg's and Egger's tests suggest asymmetry in the funnel plot (p=0.06 and 0.10, respectively). I would not exclude a possible publication bias. Your statement must be corrected. I would recommend using enhanced funnel plot to assess whether the asymmetry is associated to non-significant results (Peters J, Sutton AJ, Jones DR, Abrams KR, Rushton L: Contour-enhanced meta-analysis funnel plots help distinguish publication bias from other causes of asymmetry. J Clin Epidemiol 1-996). 我最近也在做肺癌预后的meta，有个问题一直没弄明白。Steels在文章&role a=&& as=&& cancer:a=&& factor=&& for=&& in=&& literature=&& lung=&& meta-analysis=&& of=&& p53=&& prognostic=&& review=&& survival=&& systematic=&& the=&& with=&&&中使用了“European LungCancer Working Party quality scale for biological prognostic factors for lung cancer”这个评价标准。文章中提到：总分为40分，每个部分总分10分。但是附录中该评价标准共包含了22个项目，每个项目为2分，总分应当为44分。我看到好多meta分析都是引用的这个评价标准，而且也说明了总分为40分。我不太明白应该怎么计算得分，麻烦giggle兄指点一下。&/role&总分是44h分，但是他实际给出的应该是44分，想问一下你是怎么解决的？
微信扫一扫
广告宣传推广
政治敏感、违法虚假信息
恶意灌水、重复发帖
违规侵权、站友争执
附件异常、链接失效
关于丁香园