Read more

September 05, 2024
2 min read
Save

Up to 1 in 5 children have the wrong race listed in their EMR

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Key takeaways:

  • Across three health systems, 79% to 88% of patients’ electronic medical records had the correct racial designation.
  • Systems with more racial and ethnic options to choose from had higher error rates.

As many as one in five children had the wrong race listed in their electronic medical records in a study of pediatric health systems in Michigan.

“Without knowledge of the degree of misattribution in racial and ethnic designations, studies run the risk of missing existing inequities and disparities and identifying some that do not exist,” Gary L. Freed, MD, MPH, Percy and Mary Murphy Professor of Pediatrics at University of Michigan School of Medicine and professor of health management at the University of Michigan School of Public Health, and colleagues wrote in JAMA Network Open.

tablet_stock
As many as one in five children’s electronic medical records misidentified their race in a study. Image: Adobe Stock.

The cross-sectional study surveyed a convenience sample of 4,333 parents and guardians of patients at three pediatric health systems from Sept. 1, 2023, to Jan. 31, 2024. The researchers compared the race and ethnicity of children as reported by their parent or guardian to the race and ethnicity in their EMR.

The researchers matched the patients’ races in three stages with varying specificity. First, they searched for exact matches between parent-reported race and EMR-identified race. Second, if parents reported more than one race, the researchers narrowed it to one, prioritizing minoritized groups. Third, they narrowed down race designators to five or six broad groups because some health systems had more options than the others.

Two health systems had only Hispanic or non-Hispanic ethnic designators, whereas one had 10 ethnicities to choose from. The researchers started by looking for exact matches and then narrowed the 10 ethnic groups into Hispanic or non-Hispanic for consistency across all three health systems.

In the first stage, looking for exact matches yielded the highest number of errors, with matching rates of 41% to 78%. When the researchers narrowed racial designators in the third stage, they found matching rates of 79% to 88%.

The proportion of patients whose reported ethnicity matched their EMR ranged from 65% to 95%.

Overall, 2% to 10% of patients were missing racial data in their EMR, and 1% to 11% were missing ethnic data.

The researchers noted that health systems tended to undercount racial designations. Almost all patients (97%) whose EMR listed them as Black were identified as Black on the survey as well. However, only 71% to 91% of patients whose survey identified them as Black were also listed as Black in their EMR.

The health systems with more race and ethnicity options appeared to have a higher error rate, according to the researchers.

“Most previous studies of disparities and inequities using medical records, claims or national datasets have not reported any assessments of either the accuracy of racial and ethnic designations or the rates of missing data within those data sources,” Freed and colleagues wrote. “Thus, the validity of such studies may be called into question. Depending on the degree of error in these variables, some of these studies may have missed disparities and inequities that exist and/or found some that do not.”