Study reveals bias in substance use screening for traumatized teens

Injured teens from marginalized groups treated at pediatric trauma centers are more likely to be tested for drugs and alcohol than white teens, even when injury severity is taken into account, according to a study by researchers at UCLA and at Children’s Hospital Los Angeles.

The findings, to be published on October 4 in the peer-reviewed journal JAMA Network Opensuggest that clinician biases could influence the selection of adolescents for biochemical screening at pediatric trauma centers, said Dr. Jordan Rook, a general surgery resident at the David Geffen School of Medicine at UCLA and lead author of the study. These unfair screening standards can lead to stigmatization and perhaps even legal implications for some traumatized adolescents.

While screening can positively impact patients if followed by counseling and treatment, it can also lead to negative consequences. We believe that existing substance use screening guidelines may be insufficient to achieve equitable high-quality screening in adolescent trauma care. Tighter guidance and supervision and/or implementation of universal screening protocols and equitable use of support services may be required.’

Dr. Jordan Rook, general surgery specialist, David Geffen School of Medicine at UCLA

The researchers used data from the 2017-2021 ACS Trauma Quality Programs on 85,400 adolescent trauma patients aged 12 to 17 years from 121 pediatric trauma centers. Of these adolescents, 67% were white, 82% were non-Hispanic, 72% were male, and 51% had private insurance.

Of all teenagers, 25% were tested for alcohol and 22% for drugs. Overall, American Indian, black, Hispanic, female, Medicaid-insured and uninsured teens were more likely to be screened for both alcohol and drugs, the researchers found.

Among the findings:

• For black adolescents, the odds of alcohol and drug testing were 8% and 13% higher, respectively, than for white adolescents.
• For American Indian adolescents, the odds of being screened for alcohol and drugs were 117% and 75% higher, respectively, than for white adolescents.
• For Hispanic adolescents, the odds of alcohol and drug testing were 20% and 12% higher, respectively, than for white adolescents.
• For adolescent females, the odds of testing alcohol and drugs were 32% and 28% higher, respectively, than males.
• For adolescents insured by Medicaid, the odds of alcohol and drug testing were 15% and 28% higher, respectively, than for adolescents with private insurance.

The authors note that there are some limitations to the study. The data the authors used did not describe whether the trials resulted in treatment or intervention, so it was unclear whether the screenings’ benefits outweighed the potential harms. In addition, the data only include biochemical screening tests and not interview-based screenings, thereby underestimating overall screening rates.

Researchers are conducting more studies expanding on these findings to identify potential solutions to the disparities, Rook said. Using national data, they are studying whether individual hospital practices reduce screening disparities and will also examine the accuracy and effectiveness of interview-based screening versus biochemical screening.

“All of these efforts seek to equitably increase substance use screening and support services for all adolescents,” Rook said.

The study’s senior author is Dr. Lorraine Kelley-Quon of Children’s Hospital Los Angeles and the University of Southern California. Additional co-authors are Drs. Catherine Juillard of UCLA. Dr. Ryan Spurrier, Dr. Cathy Shin of Children’s Hospital Los Angeles and the University of Southern California. Dr. Christopher Russell of Stanford University. and Dr. Stephen Lee of Seattle Children’s Hospital.

The study was funded by the VA Office of Academic Affiliations through a National Clinician Scholars Program Fellowship, an Association for Academic Surgery Clinical Outcomes and Health Services Research Award, and a National Center for Advancing Translational Sciences UCLA CTSI Grant (UL1TR001881).