Clinicians in obstetric units place nearly twice as many wrong-patient orders as their medical-surgical counterparts, based on a retrospective look at more than 1.3 million orders.
These findings suggest that obstetric patients are at particular risk for this type of medical error, and that steps are needed to address obstetric clinical culture, work flow, and electronic medical record interfaces, reported lead author Adina R. Kern-Goldberger, MD, of the department of obstetrics and gynecology at the University of Pennsylvania, Philadelphia, and colleagues.
The root of the issue may come from the very nature of obstetrics, and the homogeneity of the patient population, they wrote in Obstetrics & Gynecology.
“Obstetrics is a unique clinical environment because all patients are admitted with a common diagnosis – pregnancy – and have much more overlap in demographic characteristics than a typical inpatient unit given that they are all females of reproductive age,” the investigators wrote. “The labor and delivery environment also is distinct in the hospital given its dynamic tempo and unpredictable work flow. There also is the added risk of neonates typically being registered in the hospital record under the mother’s name after birth. This generates abundant opportunity for errors in order placement, both between obstetric patients and between postpartum patients and their newborns.”
To determine the relative magnitude of this risk, Dr. Kern-Goldberger and colleagues analyzed EMRs from 45,436 obstetric patients and 12,915 medical-surgical patients at “a large, urban, integrated health system in New York City,” including 1,329,463 order sessions placed between 2016 and 2018.
The primary outcome was near-miss wrong-patient orders, which were identified by the Wrong-Patient Retract-and-Reorder measure.
“The measure uses an electronic query to detect retract-and-reorder events, defined as one or more orders placed for patient A, canceled by the same clinician within 10 minutes, and reordered by the same clinician for patient B within the next 10 minutes,” the investigators wrote.In obstetric units, 79.5 wrong-patient orders were placed per 100,000 order sessions, which was 98% higher than the rate of 42.3 wrong-patient orders per 100,000 order sessions in medical-surgical units (odds ratio, 1.98; 95% confidence interval, 1.64-2.39), a disparity that was observed across clinician types and times of day.Advanced practice clinicians in obstetrics placed 47.3 wrong-patient orders per 100,000 order sessions, which was significantly lower than that of their colleagues: attending physicians (127.0 per 100,000) and house staff (119.9 per 100,000).
Wrong-patient orders in obstetrics most often involved medication (73.2 per 100,000), particularly nifedipine, antibiotics, tocolytics, and nonoxytocin uterotonics. The “other” category, including but not limited to lab studies and nursing orders, was associated with 51.0 wrong-patient orders per 100,000 order sessions, while errors in diagnostic imaging orders followed distantly behind, at a rate of 5.7 per 1000,000.
“Although the obstetric clinical environment – particularly labor and delivery – is vibrant and frequently chaotic, it is critical to establish a calm, orderly, and safe culture around order entry,” the investigators wrote. “This, combined with efforts to improve house staff work flow and to optimize EMR interfaces, is likely to help mitigate the threat of wrong order errors to patient care and ultimately improve maternal health and safety.”
According to Catherine D. Cansino, MD, associate clinical professor of obstetrics and gynecology at UC Davis (Calif.) Health, the findings highlight the value of medical informatics while revealing a need to improve EMR interfaces.
“Medical informatics is a growing field and expertise among ob.gyns. is very important,” Dr. Cansino said in an interview. “This study by Kern-Goldberger and colleagues highlights the vulnerability of our EMR systems (and our patients, indirectly) when medical informatics systems are not optimized. The investigators present a study that advocates for greater emphasis on optimizing such systems in obstetrics units, especially in the context of high acuity settings such as obstetrics, compared to medical-surgical units. Appropriately, the study highlights the avoided harm when correcting medical errors for obstetric patients since such errors potentially affect both the delivering patient and the newborn.”
The study was funded by AHRQ. One coauthor disclosed funding from the Icahn School of Medicine at Mount Sinai, Georgetown University, the National Institutes of Health – Office of Scientific Review, and the Social Science Research Council. Another reported funding from Roche.