Schwarze Juan-Enrique, Tennant Peter W G, Barnhart Kurt, Platt Robert W, Gupta Shiv, Venetis Christos, D'Hooghe Thomas, Schisterman Enrique F
Merck Healthcare KGaA, Darmstadt, Germany.
School of Medicine, University of Leeds, Leeds, United Kingdom; Leeds Institute for Data Analytics, University of Leeds, Leeds, United Kingdom.
Fertil Steril. 2025 Jun 2. doi: 10.1016/j.fertnstert.2025.05.168.
To provide a framework for conducting rigorous nonrandomized studies of interventions in fertility treatment research, addressing their role as complements to randomized controlled trials (RCTs) in evaluating treatment outcomes.
Multidisciplinary expert consensus on best practices for nonrandomized studies of interventions, informed by advancements in novel methodologies, including causal inference.
Patients undergoing assisted reproductive technologies (ARTs) procedures, such as ovarian stimulation, laboratory techniques, and embryo transfer.
None.
Guidance on methodological rigor, transparency, and relevance in nonrandomized studies of interventions study design and analysis.
Randomized controlled trials are the gold standard for determining the efficacy and safety of fertility treatment/ART interventions but can face logistical, practical, and sometimes ethical challenges. Nonrandomized studies of interventions, when conducted with high methodological rigor, complement RCTs by offering insights into real-world clinical practices and diverse patient populations. Key limitations of nonrandomized studies of interventions include susceptibility to confounding and selection bias, which require meticulous study design and advanced analytical techniques to address. Recent innovations, such as target trial emulation studies, have enhanced the validity of causal inferences based on nonrandomized studies of interventions. This article outlines 7 recommendations to improve the credibility of nonrandomized studies of interventions in ART research: clearly define research questions with precise estimands; design nonrandomized studies of interventions as emulated trials; use directed acyclic graphs to clarify causal assumptions; preregister study protocols; separate data analysis from study planning; incorporate negative controls to detect biases; and use appropriate analytical methods to account for confounding and selection bias.
Integrating evidence from RCTs and well-conducted nonrandomized studies of interventions enhances clinical decision making in fertility treatment research. By adhering to these recommendations, researchers can improve the quality, transparency, and impact of nonrandomized studies of interventions, ultimately fostering robust, evidence-based clinical practices in fertility treatment/ART.
为生育治疗研究中的干预措施进行严格的非随机研究提供一个框架，探讨其在评估治疗结果方面作为随机对照试验（RCT）补充的作用。
基于包括因果推断在内的新方法的进展，对干预措施的非随机研究的最佳实践达成多学科专家共识。
接受辅助生殖技术（ART）程序的患者，如卵巢刺激、实验室技术和胚胎移植。
无。
关于干预措施研究设计和分析中方法严谨性、透明度和相关性的指导。
随机对照试验是确定生育治疗/ART干预措施有效性和安全性的金标准，但可能面临后勤、实际和有时的伦理挑战。当以高度的方法严谨性进行干预措施的非随机研究时，通过提供对现实世界临床实践和不同患者群体的见解来补充RCT。干预措施的非随机研究的主要局限性包括易受混杂和选择偏倚的影响，这需要精心的研究设计和先进的分析技术来解决。最近的创新，如目标试验模拟研究，提高了基于干预措施非随机研究的因果推断的有效性。本文概述了7项建议，以提高ART研究中干预措施非随机研究的可信度：用精确的估计量明确界定研究问题；将干预措施的非随机研究设计为模拟试验；使用有向无环图来阐明因果假设；预先注册研究方案；将数据分析与研究计划分开；纳入阴性对照以检测偏倚；并使用适当的分析方法来处理混杂和选择偏倚。
整合来自RCT和精心进行的干预措施非随机研究的证据可增强生育治疗研究中的临床决策。通过遵循这些建议，研究人员可以提高干预措施非随机研究的质量、透明度和影响力，最终促进生育治疗/ART中稳健的、基于证据的临床实践。
