Biases in electronic health record data due to processes within the healthcare system: retrospective observational study

TLDR

The study evaluates how healthcare processes affect the predictive value of EHR laboratory test data across 272 common tests. A retrospective observational study of 669,452 patients at two Boston hospitals over 2005–2006 examined the relative predictive accuracy of each lab test for three‑year survival, comparing test result values to the time of day, day of week, and ordering frequency. The study found that test ordering alone predicted survival in 86% of tests, and ordering timing was more predictive than results in 68% of tests, indicating that healthcare processes must be considered in EHR analyses, though they can be leveraged to gain patient insight.

Abstract

<h3>Abstract</h3> <h3>Objective</h3> To evaluate on a large scale, across 272 common types of laboratory tests, the impact of healthcare processes on the predictive value of electronic health record (EHR) data. <h3>Design</h3> Retrospective observational study. <h3>Setting</h3> Two large hospitals in Boston, Massachusetts, with inpatient, emergency, and ambulatory care. <h3>Participants</h3> All 669 452 patients treated at the two hospitals over one year between 2005 and 2006. <h3>Main outcome measures</h3> The relative predictive accuracy of each laboratory test for three year survival, using the time of the day, day of the week, and ordering frequency of the test, compared to the value of the test result. <h3>Results</h3> The presence of a laboratory test order, regardless of any other information about the test result, has a significant association (P&lt;0.001) with the odds of survival in 233 of 272 (86%) tests. Data about the timing of when laboratory tests were ordered were more accurate than the test results in predicting survival in 118 of 174 tests (68%). <h3>Conclusions</h3> Healthcare processes must be addressed and accounted for in analysis of observational health data. Without careful consideration to context, EHR data are unsuitable for many research questions. However, if explicitly modeled, the same processes that make EHR data complex can be leveraged to gain insight into patients' state of health.