Documented Clinical Diagnoses Underestimate the Prevalence of Substance Use Disorders in Younger, Hispanic, and Women Veterans

BACKGROUND:
Though effective treatments for substance use disorder (SUD) exist, SUD has one of the largest treatment gaps for any medical condition. Ensuring equitable identification and diagnosis of SUD in clinical settings requires understanding the true underlying prevalence of SUD diagnoses, particularly across subgroups (i.e., race and ethnicity, gender, and age). This study compared clinical diagnosis rates of alcohol use disorder (AUD), drug use disorder (DUD), and total SUD (AUD and/or DUD) to survey-based prevalence among a random sample of VA patients from 30 VA healthcare facilities. Investigators recruited 5,995 Veterans between January 2018 and April 2019. Surveyed patients completed the Mini-International Neuropsychiatric Interview (MINI 7.0) by phone, which assessed DSM-5 criteria for AUD and substance-specific DUD over the 12 months prior to the survey. Responses were compared to clinically documented diagnoses in the VA electronic health record for each participant.

FINDINGS:

• The survey-based prevalence of AUD, DUD, and SUD was generally higher than clinical diagnosis rates among all Veterans: 10% vs 6% for AUD, 4.7% vs 4.6% for DUD, and 13% vs 9% for SUD.
• The survey-based prevalence of AUD and SUD exceeded clinical diagnosis rates in every demographic subgroup.
• For DUD, the greatest levels of clinical underdiagnosis/under-recognition were seen in the youngest age group and among those reporting White race/ethnicity. For SUD overall, the greatest levels of under-recognition were for women, youngest and oldest age groups, and those reporting Hispanic ethnicity. For AUD, the greatest levels of under-recognition were among women, youngest and oldest age groups, persons of “other” race/ethnicity, and White persons.

IMPLICATIONS:

• Documented clinical diagnoses are insufficient to capture the prevalence of SUD, particularly for women, younger, and Hispanic/Latinx patients, the latter of whom may often experience the greatest consequences of SUD.
• Implementation of brief screens for drug use and structured clinical assessment for AUD may support increased equity in diagnosis and treatment and optimize identification of target populations used in performance monitoring.

LIMITATIONS:

• Differences between survey-based and clinically-documented diagnoses may be due partially to the diagnostic criteria used, given that DSM-5 criteria may capture a higher prevalence than DSM-IV, and study instrumentation did not include assessment of DSM-IV.

AUTHOR/FUNDING INFORMATION:
This study was funded by HSR&D (IIR 15-436), and Dr. Harris is supported by an HSR&D Research Career Scientist Award. Dr. Williams and Mss. Fletcher and Frost are part of HSR&D’s Center of Innovation for Veteran-Centered and Value-Driven Care. Dr. Harris is with HSR&D’s Center for Innovation to Implementation (Ci2i); Dr. Washington is with HSR&D’s Center for the Study of Healthcare Innovation, Implementation & Policy (CSHIIP); and Dr. Hoggatt is with the VA San Francisco Health Care System.

Williams EC, Fletcher O, Frost MC, Harris AHS, Washington D, and Hoggatt KJ. Comparison of Substance Use Disorder Diagnosis Rates from Electronic Health Record Data with Substance Use Disorder Prevalence Rates Reported in Surveys across Sociodemographic Groups in the Veterans Health Administration. JAMA Network Open. June 30, 2022;5(6):e2219651.