2023 HSR&D/QUERI National Conference

1006 — Trends in substance use disorder diagnoses among Veterans, 2009-2019

Lead/Presenter: Katherine Hoggatt,  San Francisco VA Health Care System
All Authors: Hoggatt KJ (San Francisco VA Health Care Center; University of California, San Francisco), Chawla N (VA Health Services Research & Development (HSR&D) Center for the Study of Healthcare Innovation, Implementation, & Policy, VA Greater Los Angeles Healthcare System, Los Angeles); Washington DL (VA Health Services Research & Development (HSR&D) Center for the Study of Healthcare Innovation, Implementation, & Policy, VA Greater Los Angeles Healthcare System, Los Angeles; Division of General Internal Medicine and Health Services Research, Department of Medicine, UCLA Geffen School of Medicine, Los Angeles); Yano EM (VA Health Services Research & Development (HSR&D) Center for the Study of Healthcare Innovation, Implementation, & Policy, VA Greater Los Angeles Healthcare System, Los Angeles; Department of Health Policy and Management, UCLA Fielding School of Public Health, Los Angeles)

Objectives:
Substance use disorder (SUD) represents a substantial health burden to US Veterans. Prior research has presented conflicting information on the direction of time trends in recent years, with few examining trends in substance-specific disorders or among demographic subgroups. Our goal was to quantify recent time trends in Veterans’ substance-specific use disorders using data from the Veterans Health Administration (VA).

Methods:
We identified Veteran patients with a documented inpatient or outpatient VA encounter in fiscal years (FY) 2010-2019 (10/1/09-9/30/19). For each FY cohort (~6 million patients per FY), we extracted patient demographics and diagnoses from VA CDW. We defined substance-specific disorders for alcohol, cannabis, cocaine, opioids, sedatives, and stimulants using ICD-9 (FY10-FY15) or ICD-10 (FY16-FY19) codes. We created summary variables for polysubstance use disorder (2+ substance-specific disorders, including alcohol), drug use disorder (any non-alcohol, non-nicotine use disorder), and SUD (alcohol or drug use disorder). To summarize the smoothed time-trend, we fit log-binomial models with separate intercepts and slopes to account for the change from ICD-9 to ICD-10. We used the fitted model to estimate the relative change in diagnoses (percent change in diagnoses per year) for two periods: FY10-FY15 and FY16-FY19.

Results:
In FY10, 7.96% of Veteran patients had a SUD diagnosis, rising to 9.04% in FY15. There was a decrease in the SUD diagnosis rate after the change to the ICD-10 coding system (7.60% in FY16) and an increasing trend through FY19 (8.58%). For FY10-FY15, diagnoses for most substance-specific disorders, polysubstance use disorder, drug use disorder, and SUD increased 2%-13% annually, whereas cocaine use disorder decreased by 1.24% per year. For FY16-FY19, alcohol, cannabis, and stimulant use disorders increased 4%-18% annually, while cocaine, opioid, and sedative use disorders changed by < = 1%. The sharpest increases were for stimulant use disorder (13.01% FY10-FY15; 18.01% FY16-FY19) and cannabis use disorder (7.58% FY10-FY15; 8.01% FY16-FY19). Among both women and men, the oldest Veterans had the largest annual increases in diagnoses. For women ages 65+, there were annual increases of 13%-43% in FY10-FY15 and 9%-35% in FY16-FY19. For men ages 65+, there were annual increases of 11%-39% in FY10-FY15 and 7%-32% in FY16-FY19.

Implications:
Rates of diagnosed SUD among VA patients have increased overall in recent years, with notably strong trends for older Veterans. The relatively large trends for both cannabis and stimulant use disorders highlight a need to identify and intervene early on substance misuse before misuse leads to a more serious disorder, which may be difficult in systems like VA that do not have universal drug use screening.

Impacts:
As VA continues to evolve as a learning health care system, information on time trends is critical for care planning and resource allocation. Rapid increases in cannabis and stimulant use disorders present a treatment challenge, with fewer evidence-based options available than for alcohol or opioid use disorders. Efforts to ensure prevention and care for SUD may require expanded screening, particularly for subgroups (e.g., patients prescribed stimulants, older adults) who may not be identified through existing screening and identification programs.