By Hugo Francisco de SouzaReviewed by Susha Cheriyedath, M.Sc.May 19 2025

New research reveals that more than one in ten older US veterans use cannabis, and over a third of these users meet the threshold for cannabis use disorder, highlighting urgent needs for screening and prevention in an aging population.

Study: Cannabis Use Among Older Adults. Image Credit: BenLJohnson / Shutterstock

I

n a recent study published in JAMA Network Open, researchers leveraged a Veterans Health Administration (VHA) cohort (ages 65 to 84 years; 4,503 participants) to investigate cannabis use patterns and cannabis use disorder (CUD) among older adults. The cross-sectional study, conducted via interviews between February 2020 and August 2023, revealed that 10.3% of participants reported cannabis use in the past 30 days, with 36.3% of these users meeting criteria for CUD. Within this older cohort, participants at the younger end of the age range (65–75 years), those using inhaled cannabis, and individuals with anxiety or functional impairments faced the highest risk.

Background

State-directed legalization of cannabis has contributed to rising use across all age groups in the U.S. Among adults aged 65 and older, prevalence increased from 0.4–0.7% in 2005–2006 to 8.4% in 2022 in the general population, but the VHA cohort showed even higher rates (10.3%). Older adults are particularly vulnerable to the adverse effects of cannabis, such as impaired mobility or exacerbation of chronic conditions, yet prior research has largely overlooked this age group. This study is the first to detail CUD prevalence and risk factors in older veterans.

Study Design

The VA Cannabis and Aging Study analyzed medical records and telephone interviews from 4,503 community-dwelling VHA patients. Participants were identified using a text-processing algorithm to screen medical records for cannabis-related terms. Exclusions included severe cognitive impairment or terminal illness. Interviews assessed lifetime and past 30-day cannabis use, forms of consumption (e.g., smoking, edibles), and DSM-5 CUD criteria. Analyses accounted for sociodemographic, behavioral, and health-related factors.

Key Findings

The study found that 57.4% of participants reported lifetime cannabis use, with 28.9% using it for medical reasons such as pain, mental health, or sleep management. Among recent users, 10.3% had consumed cannabis in the past 30 days, and more than half of these individuals (52.4%) were frequent users, defined as consuming cannabis on 20 or more days per month. Smoking was the most common form of consumption (72.4% of recent users), followed by edibles (36.9%).

Notably, 36.3% of past 30-day users met the criteria for cannabis use disorder (CUD), including 22.9% with mild, 10.9% with moderate, and 2.5% with severe cases. The risk of CUD was 3.56 times higher among those who inhaled cannabis (e.g., smoking or vaping) compared to those who exclusively used edibles. Additional risk factors included younger age within the older cohort (65–75 years vs. 76–84 years), anxiety symptoms, illicit drug use, and deficits in activities of daily living. Participants who used cannabis exclusively for medical purposes had lower odds of CUD than recreational users.

The cohort was predominantly male (85.4%), White (78.4%), and retired (73.3%), with unmarried individuals and men more likely to report recent cannabis use.

Conclusions and Implications

The study underscores that cannabis use rates among older veterans are now comparable to tobacco use within this population, highlighting a critical need for routine clinical screening. While the findings are specific to VHA patients and may not generalize to non-veterans, they emphasize emerging public health challenges as cannabis becomes more accessible. Frequent use (≥20 days/month) and inhaled cannabis were strongly linked to CUD, whereas edible-only use appeared less harmful, suggesting potential avenues for harm reduction.

Study limitations include its cross-sectional design, which prevents causal conclusions, and reliance on self-reported data, which may underestimate true prevalence. The overrepresentation of men (85.4%) also limits insights into gender-specific patterns. Researchers advocate for integrating validated screening tools, such as DSM-5 criteria or the Cannabis Use Disorder Identification Test–Revised (CUDIT-R), into primary care for older adults.

The authors also note that residing in a state with recreational cannabis legalization was associated with higher odds of use, and that most CUD cases were mild in severity.