UK, VA investigate racial, ethnic bias when diagnosing alcohol use disorder among veterans

LEXINGTON, Ky. (May 25, 2023) — University of Kentucky College of Public Health faculty member Rachel Vickers-Smith, Ph.D., and a team of investigators recently examined whether there is racial or ethnic bias in who receives an alcohol use disorder (AUD) diagnosis.

Vickers-Smith, an assistant professor in the Department of Epidemiology and Environmental Health in the College of Public Health, and her team analyzed survey and electronic health record data from 700,000 Black, white and Hispanic veterans enrolled in Veterans Affairs' Million Veteran Program, a national research program that looks at how genes, lifestyle, military experiences and exposures affect health and wellness in veterans.

The research team compared measures of how much and how often a person drank alcohol in the past year to whether they had a diagnosis of AUD, and examined this comparison by race and ethnicity.

“If two groups drink similar amounts of alcohol, but one group has a label of ‘alcohol use disorder’ more often, then that suggests the presence of bias,” said Vickers-Smith.

Findings  

“We found that Black and Hispanic veterans are more likely to receive a diagnosis of alcohol use disorder than white veterans, even when the level of alcohol consumption is similar for all groups,” said Vickers-Smith.

The difference was greatest between Black and white men, Vickers-Smith and team wrote in the recent article published in the American Journal of Psychiatry. Black men, at all but the lowest and highest levels of alcohol consumption, had 23-109% greater odds of an AUD diagnosis compared to white men. These findings remain even after considering alcohol consumption, alcohol-related disorders such as cirrhosis and other potential confounders like age.   

While the researchers say the bias is likely a combination of an over-diagnosis of AUD among Black and Hispanic veterans, and an under-diagnosis among white veterans, the biggest discrepancy was seen at the positive screening threshold for harmful alcohol consumption. The research team writes that this finding suggests that clinicians may be quicker to assign an AUD diagnosis to Black and Hispanic veterans than white veterans when their drinking borders on a harmful level.

Conclusions

“Stronger efforts are needed to reduce bias in the diagnostic process to address racialized differences in AUD diagnosis,” said Vickers-Smith. “With that said, these public health studies make it possible to uncover important health-related findings, including disparities in diagnosis, and provides a foundation for making changes to improve the services delivered to a diverse veteran population.”

Vickers-Smith acknowledges that although similar disparities likely exist in health systems besides the VA, that data is less accessible which limits the opportunities to conduct parallel analyses.

“It remains to be seen whether the higher rate of an AUD diagnosis in Black and Hispanic individuals actually translates into a greater likelihood of receiving alcohol disorder treatment and experiencing better alcohol-related health outcomes,” said Vickers-Smith. “Regardless, any of these potential benefits should not overshadow the central issue that racial bias in alcohol assessment, particularly of Black patients, appears to exist. It is critical that we identify and correct the multilevel bias and racism-related contributors to racialized differences in diagnosis.”

Vickers-Smith is a substance use epidemiologist whose research focuses on identifying and preventing adverse outcomes and harms related to changing patterns in the drug use landscape, particularly among veterans and people in rural communities who use drugs. Vickers-Smith’s work was among the first to identify gabapentin as an emerging drug of misuse in the U.S. and is cited frequently in studies of gabapentinoids.

This material is based upon work supported by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Million Veteran Program. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.