Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Vic, Australia.
Faculty of Health, York University & University Health Network, University of Toronto, Toronto, Canada.
Heart Lung Circ. 2020 Feb;29(2):224-232. doi: 10.1016/j.hlc.2018.12.012. Epub 2019 Jan 28.
Clinical registries are effective for monitoring clinical practice, yet manual data collection can limit their implementation and sustainability. The objective of this study was to assess the feasibility of using a data capture tool to collect cardiac rehabilitation (CR) minimum variables from electronic hospital administration databases to populate a new CR registry in Australia.
Two CR facilities located in Melbourne, Australia participated, providing data on 42 variables including: patient socio-demographics, risk factors and co-morbidities, CR program information (e.g. number of CR sessions), process indicators (e.g. wait time) and patient outcomes (e.g. change in exercise capacity). A pre-programmed, automated data capture tool (GeneRic Health Network Information for the Enterprise [20]: https://www.grhanite.com/) (GRHANITE™) was installed at the sites to extract data available in an electronic format from hospital sites. Additionally, clinicians entered data on CR patients into a purpose-built web-based tool (Research Electronic Data Capture: https://www.project-redcap.org/) (REDCap). Formative evaluation including staff feedback was collected.
The GRHANITE™ tool was successfully installed at the two CR sites and data from 176 patients (median age = 67 years, 76% male) were securely extracted between September-December 2017. Data pulled electronically from hospital databases was limited to seven of the 42 requested variables. This is due to CR sites only capturing basic patient information (e.g. socio-demographics, CR appointment bookings) in hospital administrative databases. The remaining clinical information required for the CR registry was collected in formats (e.g. paper-based, scanned or Excel spreadsheet) deemed unusable for electronic data capture. Manually entered data into the web-tool enabled data collection on all remaining variables. Compared to historical methods of data collection, CR staff reported that the REDCap tool reduced data entry time.
The key benefits of a scalable, automated data capture tool like GRHANITE™ cannot be fully realised in settings with under-developed electronic health infrastructure. While this approach remains promising for creating and maintaining a registry that monitors the quality of CR provided to patients, further investment is required in the digital platforms underpinning this approach.
临床注册是监测临床实践的有效方法，但手动数据收集可能会限制其实施和可持续性。本研究的目的是评估使用数据采集工具从电子医院管理数据库中收集心脏康复（CR）最小变量以填充澳大利亚新的 CR 注册表的可行性。
澳大利亚墨尔本的两家 CR 机构参与了这项研究，提供了 42 个变量的数据，包括：患者社会人口统计学、风险因素和合并症、CR 计划信息（例如 CR 疗程数）、过程指标（例如等待时间）和患者结局（例如运动能力的变化）。在现场安装了一个预编程的自动化数据采集工具（GeneRic Health Network Information for the Enterprise [20]：https://www.grhanite.com/）（GRHANITE™），从医院站点提取可用的电子格式的数据。此外，临床医生将 CR 患者的数据输入到一个专门构建的基于网络的工具（Research Electronic Data Capture：https://www.project-redcap.org/）（REDCap）中。收集了形成性评估，包括员工反馈。
GRHANITE™工具成功安装在两个 CR 站点，2017 年 9 月至 12 月期间安全地从 176 名患者（中位数年龄=67 岁，76%为男性）中提取了数据。从医院数据库中电子提取的数据仅限于 42 个请求变量中的 7 个。这是因为 CR 站点仅在医院管理数据库中捕获基本的患者信息（例如社会人口统计学、CR 预约）。CR 注册表所需的其余临床信息以电子数据采集不可用的格式（例如基于纸张、扫描或 Excel 电子表格）收集。在网络工具中手动输入的数据使所有剩余变量的数据收集成为可能。与历史数据收集方法相比，CR 工作人员报告说，REDCap 工具减少了数据录入时间。
在电子健康基础设施不完善的环境中，像 GRHANITE™这样可扩展的自动化数据采集工具的主要优势无法完全实现。虽然这种方法对于创建和维护监测向患者提供的 CR 质量的注册表仍然很有希望，但需要在支持这种方法的数字平台上进一步投资。
