Sanz Vidorreta Fernando Javier, Dudley Michael T, Walling Anne M, Tseng Chi-Hong, Hogarth Michael, Wenger Neil S
Biomedical Informatics Program at UCLA Clinical and Translational Science Institute, University of California, Los Angeles, CA 90095, United States.
UCLA Office of Health Informatics and Analytics, University of California, Los Angeles, CA 90095, United States.
JAMIA Open. 2024 Oct 29;7(4):ooae121. doi: 10.1093/jamiaopen/ooae121. eCollection 2024 Dec.
Health systems are increasingly accountable for patients and require accurate electronic health record (EHR) vital status. We recently demonstrated that 19% of seriously ill primary care patients in one system were not marked dead in the EHR and 80% of these decedents had an encounter or appointment outstanding after death. Herein we describe the mechanism of identifying decedents whose death is not captured at the level of the EHR, characterize these decedents, and describe medications refilled after death.
Description of multistep process to identify deceased patients not marked dead in the EHR among a cohort of seriously ill primary care patients including public death file matching, utilization analysis, and chart abstraction. We compared decedents not marked dead in the EHR to known decedents and described pharmacy requests and refills.
Nearly 90% of encounters and appointments occurred because the health system EHR did not record the death although 11% of these encounters contained condolences or death notifications. Decedents not marked dead in the EHR were older and lived in more vulnerable areas than those marked dead. Of 146 refill requests after death, 88 medications were authorized.
Matching with a limited public death file is an inadequate solution to inaccurate vital status. Better workflows are needed to capture deaths about which clinicians and staff are aware, but will identify only a fraction of the decedents inaccurately listed as alive. Efforts are needed to connect EHRs with more specific sources of linkable decedent information.
卫生系统对患者的责任日益加重,需要准确的电子健康记录(EHR)生命状态信息。我们最近发现,在一个系统中,19%的重症初级保健患者在EHR中未被标记为死亡,且这些死者中有80%在死后仍有未完成的就诊或预约。在此,我们描述了识别EHR层面未记录死亡的死者的机制,对这些死者进行了特征描述,并介绍了死后重新配药的情况。
描述了多步骤流程,用于在重症初级保健患者队列中识别EHR中未标记为死亡的已故患者,包括与公共死亡档案匹配、利用情况分析和病历摘要。我们将EHR中未标记为死亡的死者与已知死者进行了比较,并描述了药房请求和重新配药情况。
近90%的就诊和预约发生是因为卫生系统EHR未记录死亡,尽管其中11%的就诊包含吊唁或死亡通知。EHR中未标记为死亡的死者比标记为死亡的死者年龄更大,居住在更脆弱的地区。在146份死后重新配药请求中,88种药物获得了批准。
与有限的公共死亡档案匹配对于不准确的生命状态信息来说是一个不充分的解决方案。需要更好的工作流程来记录临床医生和工作人员已知的死亡情况,但只能识别出一小部分被错误列为存活的死者。需要努力将EHR与更具体的可关联死者信息来源相连接。