Addiction Center, University of Michigan, North Campus Research Complex Building 16, 2800 Plymouth Rd., Room 222W, Ann Arbor, MI, 48109, USA.
North Jersey Community Research Initiative, 393 Central Ave, Newark, NJ, 07103, USA.
BMC Med Res Methodol. 2023 Jul 6;23(1):162. doi: 10.1186/s12874-023-01986-6.
Adaptive interventions are often used in individualized health care to meet the unique needs of clients. Recently, more researchers have adopted the Sequential Multiple Assignment Randomized Trial (SMART), a type of research design, to build optimal adaptive interventions. SMART requires research participants to be randomized multiple times over time, depending upon their response to earlier interventions. Despite the increasing popularity of SMART designs, conducting a successful SMART study poses unique technological and logistical challenges (e.g., effectively concealing and masking allocation sequence to investigators, involved health care providers, and subjects) in addition to other challenges common to all study designs (e.g., study invitations, eligibility screening, consenting procedures, and data confidentiality protocols). Research Electronic Data Capture (REDCap) is a secure, browser-based web application widely used by researchers for data collection. REDCap offers unique features that support researchers' ability to conduct rigorous SMARTs. This manuscript provides an effective strategy for performing automatic double randomization for SMARTs using REDCap.
Between January and March 2022, we conducted a SMART using a sample of adult (age 18 and older) New Jersey residents to optimize an adaptive intervention to increase COVID-19 testing uptake. In the current report, we discuss how we used REDCap for our SMART, which required double randomization. Further, we share our REDCap project XML file for future investigators to use when designing and conducting SMARTs.
We report on the randomization feature that REDCap offers and describe how the study team automated an additional randomization that was required for our SMART. An application programming interface was used to automate the double randomizations in conjunction with the randomization feature provided by REDCap.
REDCap offers powerful tools to facilitate the implementation of longitudinal data collection and SMARTs. Investigators can make use of this electronic data capturing system to reduce errors and bias in the implementation of their SMARTs by automating double randomization.
The SMART study was prospectively registered at Clinicaltrials.gov; registration number: NCT04757298, date of registration: 17/02/2021.
适应性干预措施常用于个性化医疗保健，以满足客户的独特需求。最近，越来越多的研究人员采用序贯多项分配随机试验（SMART），这是一种研究设计，以构建最佳的适应性干预措施。SMART 要求研究参与者根据他们对早期干预的反应，在时间上多次随机化。尽管 SMART 设计越来越受欢迎，但与所有研究设计共有的其他挑战（例如，研究邀请、资格筛选、同意程序和数据保密协议）相比，成功进行 SMART 研究还需要应对独特的技术和后勤挑战（例如，有效地对研究人员、参与的医疗保健提供者和研究对象隐藏和屏蔽分配序列）。Research Electronic Data Capture（REDCap）是一种安全的、基于浏览器的网络应用程序，广泛用于数据收集。REDCap 具有独特的功能，支持研究人员进行严格的 SMART。本文提供了一种使用 REDCap 对 SMART 进行自动双重随机化的有效策略。
2022 年 1 月至 3 月，我们使用新泽西州成年（18 岁及以上）居民的样本进行了一项 SMART，以优化一项提高 COVID-19 检测率的适应性干预措施。在本报告中，我们讨论了如何在我们的 SMART 中使用 REDCap，该 SMART 需要双重随机化。此外，我们分享了我们的 REDCap 项目 XML 文件，以供未来的研究人员在设计和进行 SMART 时使用。
我们报告了 REDCap 提供的随机化功能，并描述了研究团队如何自动化我们的 SMART 所需的额外随机化。使用应用程序编程接口与 REDCap 提供的随机化功能结合，实现了双重随机化的自动化。
REDCap 提供了强大的工具，有助于实施纵向数据收集和 SMART。研究人员可以利用这个电子数据采集系统，通过自动化双重随机化来减少 SMART 实施中的错误和偏差。
SMART 研究在 Clinicaltrials.gov 上进行了前瞻性注册；注册号：NCT04757298，注册日期：2021 年 2 月 17 日。
