Johns Hopkins University, Baltimore, MD.
Johns Hopkins University, Baltimore, MD.
Ann Epidemiol. 2019 May;33:54-63. doi: 10.1016/j.annepidem.2019.01.015. Epub 2019 Mar 4.
Use of electronic health records (EHRs) in health research may lead to the false assumption of complete event ascertainment. We estimated "observation windows" (OWs), defined as periods within which the assumption of complete ascertainment of events is more likely to hold, as a quality control approach to reducing the likelihood of this false assumption. We demonstrated the impact of OWs on estimating the rates of type II diabetes mellitus (diabetes) from HIV clinical cohorts.
Data contributed by 16 HIV clinical cohorts to the NA-ACCORD were used to identify and evaluate OWs for an operationalized definition of diabetes occurrence as a case study. Procedures included (1) gathering cohort-level data; (2) visualizing and summarizing gaps in observations; (3) systematically establishing start and stop dates during which the assumption of complete ascertainment of diabetes events was reasonable; and (4) visualizing the diabetes OWs relative to the cohort open and close dates to identify immortal person-time. We estimated diabetes occurrence event rates and 95% confidence intervals in the most recent decade that data were available (January 1, 2007, to December 31, 2016).
The number of diabetes events decreased by 17% with the use of the diabetes OWs; immortal person-time was removed decreasing total person-years by 23%. Consequently, the diabetes rate increased from 1.23 (95% confidence interval [1.20, 1.25]) per 100 person-years to 1.32 [1.29, 1.35] per 100 person-years with the use of diabetes OWs.
As the use of EHR-curated data for event-driven health research continues to expand, OWs have utility as a quality control approach to complete event ascertainment, helping to improve accuracy of estimates by removing immortal person-time when ascertainment is incomplete.
在健康研究中使用电子健康记录（EHR）可能导致对完整事件确定的错误假设。我们估计了“观察窗口”（OWs），即假设完整确定事件更有可能成立的时间段，作为一种质量控制方法，以降低这种错误假设的可能性。我们展示了 OW 在从 HIV 临床队列估计 II 型糖尿病（糖尿病）发生率方面的影响。
NA-ACCORD 中由 16 个 HIV 临床队列提供的数据用于识别和评估用于操作化糖尿病发生定义的 OW，作为案例研究。程序包括：（1）收集队列级数据；（2）可视化和总结观察中的差距；（3）系统地确定开始和停止日期，在此期间，假设完整确定糖尿病事件是合理的；（4）可视化糖尿病 OW 与队列开放和关闭日期的关系，以确定不朽的人时。我们估计了最近十年（2007 年 1 月 1 日至 2016 年 12 月 31 日）数据可用时的糖尿病发生事件发生率和 95%置信区间。
使用糖尿病 OW 后，糖尿病事件数量减少了 17%；去除不朽人时，总人年数减少了 23%。因此，糖尿病发生率从每 100 人年 1.23（95%置信区间[1.20，1.25]）增加到使用糖尿病 OW 后的每 100 人年 1.32 [1.29，1.35]。
随着 EHR 管理数据在事件驱动的健康研究中的使用继续扩大，OW 作为一种完整事件确定的质量控制方法具有实用性，通过在确定不完整时去除不朽人时，有助于提高估计的准确性。
