Error Rates in Race and Ethnicity Designation Detected in Pediatric Records

Gary L. Freed, MD, MPH

Credit: University of Michigan, School of Public Health

A new study detects misattribution of race and ethnicity in electronic medical records across 3 of Michigan's largest pediatric health systems. The research suggests these significant errors may weaken strategies to improve care.

Studies often use electronic medical records to collect demographic data of their participants. With inaccurate racial and ethnic designations, investigators cannot use the research to discover ways to improve clinical care for underrecognized demographic groups.

“…if the data regarding race and ethnicity are wrong, there’s a real risk of missing some inequities and even trying to correct inequities that may not exist,” said lead author and Mi-CHEC founder Gary L. Freed, MD, MPH, professor in the department of pediatrics at the U-M Medical School, in a press release.

Hospitals rarely verify the accuracy of racial and ethnic designation in health records. In recent years, many institutions have put in effort to give patients the option to designate their own race or ethnicity.

“However, we found that a greater number of choices may lead to greater error in the electronic medical record,” Freed said.

Errors in racial and ethnic designations may impact improvement for children with various conditions, such as asthma, obesity, diabetes, and other health issues. Without knowing a child’s race, it can be difficult to know if efforts to improve a condition are successful.

Investigators conducted a cross-sectional study of 4333 participants to assess how accurate racial and ethnic designations were for children in electronic medical records, particularly studying the 3 largest pediatric health systems in Michigan: U-M Health C.S. Mott Children’s Hospital, Children’s Hospital of Michigan, and Corewell Health Helen DeVos Children’s Hospital. In total, 1595 parents or guardians participated at Health System A, 1537 at Health System B, and 1202 at Health System C from September 1, 2023, to January 31, 2024; the health systems included outpatient clinics, emergency departments, and inpatient units.

The team used parent or guardian reports of race and ethnicity for a child as a gold standard which was then compared to the race and ethnicity designation in the electronic medical records. Race designation options across healthcare systems ranged from 6 – 49 and ethnic designation options ranged from 2 – 10.

The matching process occurred 3 stages:

1. Parents’ racial and ethnic designations for their children were compared with those in the electronic medical record.
2. Children with multiple racial categories had the designation of the minorized racial group for matching purposes.
3. The designations were combined or collapsed into 5 or 6 categories depending on the health system.

Compared to the health system A which included 25 American Indian or Pacific Islander, 122 Asian, 345 Black, 51 Middle Eastern, 989 White, 48 Other, and 14 with no data, the electronic medical record showed there were 8 American Indian or Pacific Islander, 96 Asian, 312 Black, 8 Middle Eastern, 949 White, 26 Other, and 1 with no data. Investigators also saw health systems B and C also had variations in race rates compared with electronic medical records. A similar finding was found with ethnic designation rates.

Investigators discovered the rates of exact match of parental report of race with the racial designation ranged from 41% to 78% across the 3 health systems. The racial matching rate improved for each health system with the consolidation of race options, ranging from 89% to 88%.

Rates of ethnicity had a closer match, with exact ethnicity matching rates ranging from 65% to 95% between parental reports and electronic medical records.

“This is the first step to better understanding the misattribution in racial and ethnic designations in medical records,” Freed said. “There are no perfect data; however, if we don’t know the degree of accuracy in the race and ethnicity of our patient populations, we are truly flying blind when we and others are assessing equity and disparity.”

References

1. ^{Freed GL, Bogan B, Nicholson A, Niedbala D, Woolford S. Error Rates in Race and Ethnicity Designation Across Large Pediatric Health Systems. JAMA Netw Open. 2024;7(9):e2431073. doi:10.1001/jamanetworkopen.2024.31073}
2. ^{Study: Racial And Ethnic Designation Inaccuracies in Children’s Medical Records May Impede Equity Efforts. EurekAlert! September 3, 2024. https://www.eurekalert.org/news-releases/1056141. Accessed September 5, 2024.}