Twins, Triplets at Higher Risk for Mislabeling in NICU

'Multiples with very similar names can be easily confused with each other,' study says

This article was first published on Monday, August 26, 2019 in MedPage Today.

By Elizabeth Hlavinka, Staff Writer, MedPage Today.

Multiple-birth infants had a higher risk of wrong-patient order errors compared with singleton babies in the neonatal intensive care unit (NICU), according to an analysis of New York hospitals.

Twins, triplets, and higher-order multiples had a significantly higher risk of wrong-patient order errors compared with singleton births (adjusted odds ratio 1.75, 95% CI 1.39-2.20, P<0.001), reported Jason Adelman, MD, MS, of Columbia University/NewYork-Presbyterian Hospital in New York City.

And the risk increased for each additional multiple birth, so that an error occurred in 1 of 7 sets of twins and in 1 of 3 sets of higher-order multiple births, the authors wrote in JAMA Pediatrics.

The rate of these errors, identified through the Wrong-Patient Retract-and-Reorder measure in electronic health systems, was similar between multiple-birth babies and singleton-birth babies (36.1 per 100,000 orders vs 41.7 per 100,000), indicating the excess risk among multiple-births was largely caused by mix-ups made within the family, they noted.

"We wanted to show what a lot of doctors and nurses and clinicians that work in the NICU know anecdotally: multiples with very similar names can easily be confused with each other, and we believe it's largely due to the naming convention," Adelman told MedPage Today.

These mix-ups can lead to the improper administration of medication or breast milk in the NICU, Adelman cautioned.

Because of their small size, infants in the NICU may be more prone to such errors, he and his team reported. Also, some families have not selected a name upon their child's birth and clinicians may be required to use a temporary pseudonym, adding to the confusion.

In January 2019, the Joint Commission updated its National Patient Safety Goals in an attempt to reduce the number of NICU infant mix-ups. Instead of identifying a baby by sex and last name, for example, clinicians can additionally label babies with the mother's first name, such that a girl born to a woman named "Mary Smith," would be labeled "Smith, Marysgirl" instead of "babygirl Smith."

But for multiple births, this new convention would still be subject to misidentification errors, since infants would only differ by a single character in some cases, like Marysgirl1 and Marysgirl2 for twins.

This study confirms this hypothesis, and appears to be a case in which policy changes "seem so obvious to either regulatory bodies or a health system leadership team that actual data are not used to validate or justify the modification," noted Gary Freed, MD, MPH, of the University of Michigan in Ann Arbor, in an accompanying editorial.

For example, Adelman and his team found that certain medical record systems truncate names on patients' wristbands, which may remove the distinguishing characters from infants' names.

In an era when many physicians are overwhelmed by electronic health records, it becomes even more essential to ensure that quality improvement measures are actually working towards the clinician and patient advantage, Freed noted.

"Importantly, data on the outcome of any change are necessary for both patients and health care professionals, so that no party believes changes are only cosmetic and do not constitute an additional unnecessary burden," Freed wrote. "Yet, in the current environment, it is unclear who, if anyone, has the responsibility for such assessments, especially when they are promulgated by national accreditation authorities."

The study by Adelman and colleagues involved four NICUs at New York-Presbyterian Hospital, which used the Sunrise Clinical Manager electronic health system, and two NICUs at Montefiore Health System, which used Centricity EMR.

Overall, 1,536,160 orders were placed for 10,819 infants, of which 55.8% were boys and 85.5% were singleton births. Compared with singleton-birth babies, those included in sets of two, three, or more were more likely to be girls, white, and commercially insured.

Babies born at New York-Presbyterian Hospital were more likely to be white and commercially insured, whereas infants at Montefiore were more likely to be black or Hispanic and have Medicaid health insurance. Both the overall rate of errors and the increased risk of multiple versus single births were similar across the two study sites, the authors noted.

They acknowledged the study may not be generalizable to different settings since it was conducted in one state. Singleton and multiple-birth infants also differed significantly in terms of demographic characteristics, although this was adjusted for in their analysis, they noted. Lastly, the Wrong-Patient Retract-and-Reorder measure also captures near-miss errors that were caught by the ordering clinicians, as well as errors that reached the patient.

Adelman received support from the National Institutes of Health. The study was funded by the Agency for Healthcare Research and Quality and the National Institutes of Health.