乳腺癌患者焦虑和抑郁的患病率：一项荟萃分析。
完整摘要 · 中文译文
引言：乳腺癌女性中抑郁和焦虑的报告较为常见，但不同研究环境和评估工具下的患病率估计差异很大。本荟萃分析综合了近期国际证据，并考察了总体、按严重程度分层以及按评估工具分类的患病率估计。 方法：我们检索了PubMed、Web of Science、Embase、Cochrane Library、APA PsycINFO和Scopus，纳入2015年1月至2026年9月10日发表的观察性研究。经Freeman-Tukey双反正弦转换后，采用随机效应模型合并比例。预设亚组分析考察地区、评估工具和研究设计；还进行了发表年份荟萃回归和小样本效应正式检验。质量阈值敏感性分析重新纳入了所有其他方面符合条件、经质量评价且可提取事件数和分母的报告。 结果：69篇报告接受了质量评价；52篇进入主要荟萃分析，17篇未超过预设质量阈值。52篇报告代表16,850名独立参与者。抑郁数据包括49篇报告和16,504名参与者，焦虑数据包括39篇报告和11,536名参与者。合并患病率抑郁为44.3%（95% CI：38.3-50.4%；I 2 = 98.3%），焦虑为47.9%（95% CI：40.9-54.9%；I 2 = 98.2%）。重新纳入所有经质量评价的报告后，合并估计为抑郁44.6%（95% CI：39.6-49.6%；66篇报告）和焦虑48.1%（95% CI：41.0-55.3%；53篇重新 端口），绝对变化分别为0.27和0.24个百分点。发表年份对任一结局均不是显著的调节因素。 结论：抑郁和焦虑症状被频繁报告，但估计值在不同人群和测量工具之间差异显著。合并估计值应被解释为描述性总结，而非精确的普遍患病率数值。研究定义的严重程度估计值和特定测量工具的结果可为筛查提供参考，但直接进行跨工具比较需谨慎。
完整摘要 · 英文原文
INTRODUCTION: Depression and anxiety are frequently reported among women with breast cancer, but prevalence estimates vary substantially across settings and assessment instruments. This meta-analysis synthesized recent international evidence and examined overall, severity-stratified, and instrument-specific prevalence estimates. METHODS: We searched PubMed, Web of Science, Embase, the Cochrane Library, APA PsycINFO, and Scopus for observational studies published from January 2015 through September 10, 2026. Proportions were pooled with a random-effects model after Freeman-Tukey double-arcsine transformation. Prespecified subgroup analyses examined region, assessment instrument, and study design; publication-year meta-regression and formal tests for small-study effects were also conducted. A quality-threshold sensitivity analysis re-included all otherwise eligible quality-appraised reports with extractable event counts and denominators. RESULTS: Sixty-nine reports underwent quality appraisal; 52 entered the primary meta-analysis and 17 did not exceed the prespecified quality thresholds. The 52 reports represented 16,850 independent participants. Depression data comprised 49 reports and 16,504 participants, and anxiety data comprised 39 reports and 11,536 participants. The pooled prevalence was 44.3% for depression (95% CI: 38.3-50.4%; I 2 = 98.3%) and 47.9% for anxiety (95% CI: 40.9-54.9%; I 2 = 98.2%). Re-including all quality-appraised reports produced pooled estimates of 44.6% for depression (95% CI: 39.6-49.6%; 66 reports) and 48.1% for anxiety (95% CI: 41.0-55.3%; 53 reports), absolute changes of 0.27 and 0.24 percentage points, respectively. Publication year was not a significant moderator for either outcome. CONCLUSION: Depression and anxiety symptoms were frequently reported, but estimates varied markedly across populations and instruments. The pooled estimates should be interpreted as descriptive summaries rather than precise universal prevalence values. Study-defined severity estimates and instrument-specific results may inform screening, but direct cross-instrument comparisons require caution.