Restricting physicians’ ability to view multiple patient records at the same time does not reduce wrong-patient orders, according to a large clinical trial published in JAMA.
Many accrediting agencies and patient safety experts suggest limiting the number of patient records displayed at once, ostensibly to avoid patient misidentification. Instead, the results of this study show that multiple open records may not actually cause wrong-patient orders, according to Bruce Lambert, PhD, director of the Center for Communication and Health in the School of Communication, in collaboration with the Institute for Public Health and Medicine (IPHAM), and a co-author of the study.
Wrong-patient orders are one of the most common medical errors. During 2016 alone, more than 600,000 U.S. patients had an order placed for them that was actually intended for another patient, according to previous studies.
Despite the recommendations of EHR providers, real-world open records practices vary widely. Among a survey of U.S. chief information officers at healthcare providers, just 38 percent reported restricting physicians to viewing a single patient’s record at a time.
In the current study, 3,300 physicians at a large health system in New York were randomly assigned to two groups: half to an EHR configuration that restricted physicians to one patient record at a time, and half to a configuration that allowed up to four patients’ records to be open at once.
The trial coincided with implementation of a new EHR system at all study sites — providing a blank slate for the trial, according to the authors. Near-miss wrong-patient orders were measured through an electronic “retract-and-reorder” query that searched for orders placed for a patient, retracted and then reordered shortly after by the same clinician, but for a different patient.
After two years and over 12 million total orders, investigators found that the proportion of wrong-patient order sessions was virtually unchanged between the restricted and unrestricted groups; 90.7 per 100,000 orders versus 88 per 100,000, respectively.
Notably, even clinicians in the unrestricted group completed 66 percent of orders with just one record open, somewhat limiting the inferences that can be drawn from the study. However, more granular analysis of the data suggest that having multiple open records is a symptom, rather than the cause of the problem, the authors said.
“The underlying problem is likely to be high clinician workload, which influences both the number of open records and the error rate,” said Lambert, who is also a professor of Medical Social Sciences and a member of the Robert H. Lurie Comprehensive Cancer Center of Northwestern University.
For example, the rate of wrong-patient orders was lowest in outpatient settings, where clinicians are likely to be caring for a single patient at a time. On the other hand, wrong-patient orders were highest in inpatient critical care and obstetrics units, where physicians are usually juggling multiple patients at once.
“A recent direct observation study demonstrated that multitasking and interruptions were associated with increased rates of prescribing errors, which may be unmeasured confounders in the post hoc analysis reported here,” the authors wrote. “These variations likely reflect differences in workflows and number of patients being cared for simultaneously, and highlight the need for targeted interventions to reduce wrong-patient errors in high-risk settings.”
This project was supported by grant R01HS023704 from the Agency for Healthcare Research and Quality.