Randomize Everyone: Creating Valid Instrumental Variables for Learning Health Care Systems [Methods Study], New Hampshire, 2016-2022

Comparative effectiveness research, or CER, compares two or more treatments. In some CER studies, researchers use patient data from electronic health records, or EHRs, to compare treatments. But patient traits like age may affect doctors' and patients' choice of treatments, which can bias results. Using EHR systems to identify eligible patients and assign them to treatments by chance could improve results of CER studies that use EHR data.
In this study, the research team explored the views of patients, clinic staff, and clinicians, such as doctors or nurses, on doing CER studies in clinics. The team also tested software with a widely used EHR system. The software finds patients who qualify for a study. During a clinic visit, the software prompts doctors to invite patients to take part in the study. If patients agree, the software assigns patients by chance to a treatment.

比较效果研究（Comparative effectiveness research, CER）旨在对两种及以上治疗手段开展对比评估。在部分CER研究中，研究者会利用电子健康记录（electronic health records, EHRs）中的患者数据开展治疗方案对比。然而，患者年龄等个体特征可能会影响医患双方对治疗方案的选择，进而对研究结果引入偏倚。若通过电子健康记录系统筛选符合入组标准的患者，并通过随机方式为其分配治疗方案，则可优化基于电子健康记录数据的CER研究结果。
本研究中，研究团队调研了患者、门诊工作人员以及临床从业者（如医师、护士）对于在门诊场景中开展CER研究的态度。同时，团队还针对一款广泛应用的电子健康记录系统开发并测试了配套软件：该软件可自动识别符合研究入组标准的患者，并在患者门诊就诊期间提示医师邀请其参与本研究。若患者同意参与，软件将通过随机方式为其分配对应治疗方案。