文献速递神经创伤

创伤性脑损伤文献中的曲解:患病率及相关因素的系统综述

Spin in traumatic brain injury literature: prevalence and associated factors. A systematic review.
PubMed 文献 · J Neurosurg · 2024-05-10

该系统综述检索MEDLINE/PubMed数据库1960年1月至2020年7月发表的英文随机对照试验,纳入150项仅针对创伤性脑损伤患者的高影响力期刊研究。结果显示22%(n=33)的研究存在曲解,最常见的为类型1和类型3。总体偏倚风险(p<0.001)、第一作者为神经外科成员(p=0.009)、作者中无统计学家(p=0.042)和样本量较小(p=0.033)与曲解相关。

要点

  • 创伤性脑损伤文献中曲解患病率为22%(n=33/150),最常见类型1和类型3。
  • 总体偏倚风险高(p<0.001)、第一作者为神经外科成员(p=0.009)、无统计学家(p=0.042)、样本量小(p=0.033)与曲解相关。
  • 作者认为无论研究设计和发表期刊如何,都建议对结果和作者结论进行批判性解读。

核心结论

  • 创伤性脑损伤文献中曲解患病率高,即使在领先医学期刊中也是如此。
  • 偏倚风险较高的研究更常与曲解相关。
  • 作者建议无论研究设计和发表期刊如何,都应对结果和作者结论进行批判性解读。

研究设计与样本

  • 遵循PRISMA推荐进行系统综述,检索MEDLINE/PubMed数据库1960年1月至2020年7月发表的英文文献。
  • 纳入仅招募创伤性脑损伤患者、研究手术或非手术干预、发表于高影响力期刊的随机对照试验。
  • 曲解定义为6种类型:1)关注非主要结局的统计学显著结果;2)对主要结局优效性分析中统计学非显著结果进行解释;3)尽管统计学非显著仍声称或强调治疗获益;4)结论聚焦于符合方案或按治疗分析而非意向性治疗结果;5)错误的统计分析;6)重新发表显著的次要分析而未适当说明主要结局分析结果。
  • 主要结局为文章中明确报告的主要结局,无明确主要结局的研究被排除。
  • 使用传统描述性统计描述研究特征,并进行探索性推断分析以识别与曲解相关的因素。

主要发现

  • 共纳入150项随机对照试验,22%(n=33)存在曲解,最常见为类型1和类型3。
  • 总体偏倚风险与曲解相关(p<0.001)。
  • 第一作者为神经外科部门成员与曲解相关(p=0.009)。
  • 作者中无统计学家与曲解相关(p=0.042)。
  • 样本量较小与曲解相关(p=0.033)。

临床怎么用

  • 作者认为无论研究设计和发表期刊如何,都建议对结果和作者结论进行批判性解读。

局限性

  • 原文未报告局限性

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

OBJECTIVE: Spin is characterized as a misinterpretation of results that, whether deliberate or unintentional, culminates in misleading conclusions and steers readers toward an excessively optimistic perspective of the data. The primary objective of this systematic review was to estimate the prevalence and nature of spin within the traumatic brain injury (TBI) literature. Additionally, the identification of associated factors is intended to provide guidance for future research practices. METHODS: The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations were followed. A search of the MEDLINE/PubMed database was conducted to identify English-language articles published between January 1960 and July 2020. Inclusion criteria encompassed randomized controlled trials (RCTs) that exclusively enrolled TBI patients, investigating various interventions, whether surgical or nonsurgical, and that were published in high-impact journals. Spin was defined as 1) a focus on statistically significant results not based on the primary outcome; 2) interpreting statistically nonsignificant results for a superiority analysis of the primary outcome; 3) claiming or emphasizing the beneficial effect of the treatment despite statistically nonsignificant results; 4) conclusion focused in the per-protocol or as-treated analysis instead of the intention-to-treat (ITT) results; 5) incorrect statistical analysis; or 6) republication of a significant secondary analysis without proper acknowledgment of the primary outcome analysis result. Primary outcomes were those explicitly reported as such in the published article. Studies without a clear primary outcome were excluded. The study characteristics were described using traditional descriptive statistics and an exploratory inferential analysis was performed to identify those associated with spin. The studies' risk of bias was evaluated by the Cochrane Risk of Bias Tool. RESULTS: A total of 150 RCTs were included and 22% (n = 33) had spin, most commonly spin types 1 and 3. The overall risk of bias (p < 0.001), a neurosurgery department member as the first author (p = 0.009), absence of a statistician among authors (p = 0.042), and smaller sample sizes (p = 0.033) were associated with spin. CONCLUSIONS: The prevalence of spin in the TBI literature is high, even at leading medical journals. Studies with higher risks of bias are more frequently associated with spin. Critical interpretation of results and authors' conclusions is advisable regardless of the study design and published journal.

来源:J Neurosurg,查看原文。版权归原作者所有。

引用

  1. Porto de Oliveira JVM, de Oliveira Júnior ALF, de Freitas Martins LP, et al. Spin in traumatic brain injury literature: prevalence and associated factors. A systematic review. J Neurosurg. 2024. PMID: 38728757. doi:10.3171/2023.11.JNS231822

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。