文献速递流行病学与生物统计基础与转化研究

系统评价发现卵巢癌生物标志物评估中夸大解读普遍存在

A systematic review finds that spin or interpretation bias is abundant in evaluations of ovarian cancer biomarkers.
PubMed 文献 · J Clin Epidemiol · 2019-07-19

该研究系统检索2015年发表的卵巢癌生物标志物临床评估研究,纳入前200项符合标准的研究,评估标题、摘要或正文结论中的夸大解读(spin)。结果显示140项(70%)存在至少一种形式的夸大解读,最常见为声称未研究的标志物用途、目的与结论不匹配及结果呈现不正确,提示该领域报告存在过度解读问题。

要点

  • 70%的卵巢癌生物标志物评估研究在标题、摘要或结论中存在夸大解读。
  • 最常见夸大形式为声称未研究的标志物用途(32.5%)和目的与结论不匹配(28.5%)。
  • 阅读此类研究时应警惕结论对标志物性能的过度乐观表述。

核心结论

  • 近期卵巢癌生物标志物临床评估研究中普遍存在对结果的夸大解读和过度解释。

研究设计与样本

  • 系统检索PubMed上2015年发表的所有卵巢癌生物标志物评估研究,纳入报告预后、预测或诊断性能的临床研究,共筛选1026项,326项符合标准,评估其中前200项。

主要发现

  • 200项研究中140项(70%)在标题、摘要或正文结论中包含至少一种形式的夸大解读,夸大标志物性能。
  • 最常见的夸大形式为声称未研究的标志物其他用途(65项,32.5%)。
  • 其次为目的与结论不匹配(57项,28.5%)和结果呈现不正确(40项,20%)。

临床怎么用

  • 临床医生在解读卵巢癌生物标志物研究时,应关注结论是否超出实际研究目的和结果范围。
  • 科研人员在撰写标志物评估报告时应确保结论与预设目的和实际结果一致,避免夸大表述。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: In the scientific literature, "spin" refers to reporting practices that make the study findings appear more favorable than results justify. The practice of "spin" or misrepresentation and overinterpretation may lead to an imbalanced and unjustified optimism in the interpretation of study results about performance of putative biomarkers. We aimed to classify spin (i.e., misrepresentation and overinterpretation of study findings) in recent clinical studies evaluating the performance of biomarkers in ovarian cancer. METHODS: We searched PubMed systematically for all evaluations of ovarian cancer biomarkers published in 2015. Studies eligible for inclusion reported the clinical performance of prognostic, predictive, or diagnostic biomarkers. RESULTS: Our search identified 1,026 studies; 326 studies met all eligibility criteria, of which we evaluated the first 200 studies. Of these, 140 (70%) contained one or more form of spin in the title, abstract, or main-text conclusion, exaggerating the performance of the biomarker. The most frequent forms of spin identified were (1) other purposes of biomarker claimed not investigated (65; 32.5%); (2) mismatch between intended aim and conclusion (57; 28.5%); and (3) incorrect presentation of results (40; 20%). CONCLUSION: Our study provides evidence of misrepresentation and overinterpretation of finding in recent clinical evaluations of ovarian cancer biomarkers.

来源:J Clin Epidemiol,查看原文。版权归原作者所有。

引用

  1. Ghannad M, Olsen M, Boutron I, et al. A systematic review finds that spin or interpretation bias is abundant in evaluations of ovarian cancer biomarkers. J Clin Epidemiol. 2019. PMID: 31330251. doi:10.1016/j.jclinepi.2019.07.011

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。