Politics
Increase In Marijuana Misuse Diagnoses Among Veterans Is More Complicated Than It Looks, VA Records Reveal (Op-Ed)
“What emerges is not simply a story of rising diagnoses, but a deeper set of concerns about how those diagnoses are defined, applied and ultimately used to shape both policy and perception.”
By Nate Landau, Veterans Action Council
The number of cannabis use disorder (CUD) diagnoses among veterans has risen substantially within the Department of Veterans Affairs (VA), but newly released VA records obtained via the Freedom of Information Act (FOIA) raise questions about what that increase actually means.
The data spans a period in which the diagnostic definition of CUD changed significantly, including VA’s transition from the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV to DSM-5, making it difficult to determine how much of the recorded increase reflects actual changes in cannabis use among veterans and how much merely reflects changes in how CUD is defined, identified and coded.
The records that the Veterans Action Council (VAC) obtained through FOIA include more than 17 years of data and underpin this part 5 of an ongoing series. They offer a rare look into how VA’s Veterans Health Administration (VHA) tracks CUD, revealing how the department’s transition from DSM-IV to DSM-5 fundamentally changed how cannabis misuse is defined and diagnosed.
The article explores whether shifting diagnostic criteria, regional inconsistencies, limited internal cannabis research and reports from veterans of CUD diagnoses following positive THC tests—often without a documented behavioral assessment—raise broader questions about the consistency, transparency and clinical application of CUD within the VA healthcare system.
Ultimately, it asks whether current diagnostic practices reflect rigorous patient-centered care or administrative convenience, and what those practices mean for the trust veterans place in their healthcare system.
What emerges is not simply a story of rising diagnoses, but a deeper set of concerns about how those diagnoses are defined, applied and ultimately used to shape both policy and perception.
An important distinction is needed at the outset. Much of the information examined here consists of administrative health data, diagnostic codes, screening results, prevalence counts and other records generated through healthcare delivery. This data can reveal patterns and associations, but it does not, by itself, establish why those patterns exist. Correlation means that two variables occur or change in relation to one another; it does not demonstrate that one caused the other.
Throughout this analysis, increases or geographic differences in recorded CUD diagnoses should therefore not automatically be interpreted as evidence of corresponding increases in cannabis misuse among veterans.
At first glance, the numbers suggest a dramatic increase. In 2005, approximately 37,000 veterans were diagnosed with cannabis-related conditions, representing about 8.5 per 1,000 patients. By 2014, that number had climbed to roughly 140,000 and, by 2022, it remained elevated at over 139,000, or nearly 24 per 1,000 veterans.
On the surface, this appears to reflect a sharp rise in CUD. However, that interpretation unravels in proper context.
Medical diagnoses recorded in VA health records are assigned standardized codes from the International Classification of Diseases (ICD), the system healthcare providers use to classify diagnoses and conditions. When that coding system changes, a condition’s appearance in large healthcare datasets can change as well, even when the underlying patient population does not.
A critical inflection point occurred in October 2015, when VA transitioned from the ICD, Ninth Revision (ICD-9), to the Tenth Revision (ICD-10). This change aligned VA diagnostic coding with the framework established in the DSM-5, which had, two years earlier, consolidated cannabis abuse and cannabis dependence into a single diagnosis: CUD.
While this shift was intended to modernize and standardize diagnostic criteria, it introduced a fundamental complication for longitudinal analysis. Simply put, the data before and after 2015 are not necessarily measuring the same thing.
Prior to DSM-5, cannabis abuse and cannabis dependence were treated as distinct clinical conditions under the DSM-IV. Abuse generally captured patterns of harmful use, while dependence indicated a more entrenched physiological and behavioral condition.
After 2015, these distinctions disappeared into a single, broader category. Any analysis comparing pre- and post-transition data without carefully adjusting for this shift risks drawing conclusions based on incompatible definitions. In effect, it becomes a comparison of fundamentally different diagnostic constructs.
From the VAC perspective, this is not a minor technical issue but a central flaw. VA research has been inappropriately treating these datasets as continuous, sometimes referencing CUD in periods before the diagnosis even formally existed. This methodological inconsistency calls into question the reliability of any conclusions drawn from the data.
Equally concerning are veterans’ reports of how diagnoses are applied in practice. Within the VA system, there are recurring accounts from veterans’ firsthand experience that a positive THC result on a routine urinalysis can trigger a CUD diagnosis.
In many of these cases, veterans report that they were neither evaluated against DSM-5 criteria nor informed that such a diagnosis had been added to their medical record. If as widespread as VAC believes it is, this represents a significant departure from accepted clinical standards, which require behavioral assessment and patient engagement, not simply evidence of substance use.
The issue becomes even more complex when examining regional variability. VA is organized into 18 Veterans Integrated Service Networks (VISNs), and FOIA data reveal striking differences in CUD diagnosis rates across regions. Some report fewer than 5 cases per 1,000 veterans, while others report more than 50 per 1,000.
Higher recorded rates appear in some western states and urban areas, including jurisdictions where cannabis has been legalized. But geographic overlap alone does not establish that legalization caused the differences. VA research examining cannabis laws has found relatively modest associations between legalization and increases in CUD prevalence, suggesting that other factors account for much of the overall increase.
Regional differences may therefore reflect a combination of factors, including patterns of cannabis use, patient demographics, state cannabis policies, differences in screening and documentation and clinical or administrative practices.
Despite the dataset’s scale, the VA’s research portfolio on cannabis remains relatively limited. Much of the existing work is observational in nature and relies heavily on external reviews, such as those conducted by the National Academies of Sciences, Engineering and Medicine.
While observational studies can offer valuable insights, they cannot establish causality as randomized controlled trials can. To date, VA has not conducted large-scale, internally driven clinical trials on cannabis use, leaving a significant gap between real-world usage and scientific understanding.
This gap is particularly notable given the evolving legal and medical landscape. Across most of the United States, medical cannabis is now legal at the state level and widely used for conditions such as chronic pain, post-traumatic stress and sleep disorders, conditions that are prevalent among veterans.
Yet VA’s policy framework has remained largely static. Under current rules, providers cannot recommend cannabis, even though they are prohibited from denying care to veterans who use it, as outlined in VHA Directive 1315.
The institutional stance continues to align with the American Psychiatric Association, which maintains that evidence for cannabis in psychiatric treatment remains insufficient.
This creates a difficult-to-ignore paradox. On one hand, VA applies the diagnosis of CUD on a significant scale. On the other hand, it lacks robust internal research to fully understand the substance use it is diagnosing. Federal restrictions, including cannabis’s historical classification as a Schedule I substance, have undoubtedly limited research opportunities. The result is a system that diagnoses broadly while studying narrowly.
For VAC, these issues converge into a broader concern about systemic integrity. When diagnostic categories shift without clear continuity, diagnoses are applied without a full clinical evaluation, and patients are neither informed nor consulted, the line between medical care and administrative labeling blurs. In such an environment, CUD risks becoming less a precise clinical diagnosis and more a default classification.
For veterans navigating the VA health system, this has tangible implications. Medical records influence treatment decisions, provider perceptions and even access to certain services. A diagnosis entered without proper evaluation or patient awareness can carry consequences far beyond the initial encounter.
As a result, advocates are encouraging veterans to take an active role in their healthcare documentation by reviewing their records, questioning how diagnoses were determined and requesting an explanation or reassessment when appropriate.
At its core, this is an issue of trust. Veterans depend on VA for care that is fair, accurate and respectful, and FOIA data underscore the need for greater transparency and accountability. Numbers alone are not enough; their meaning depends on how they are generated and applied.
A diagnosis must reflect careful clinical judgment grounded in consistent methodology and patient context. When it does not, it undermines confidence in the system. The path forward is clear: CUD should be applied with rigor, not convenience, and every diagnosis should reflect a genuine commitment to patient care rather than administrative expediency.
With over 11 years in the cannabis industry, Nate Landau brings expertise in cultivation, veteran advocacy, and cannabis policy. A former sergeant in the Israeli Defense Forces special forces paratrooper reconnaissance unit. An award-winning no-till living soil cultivator, Nate has earned consecutive California State Fair honors. He also works alongside the Veterans Action Council, advocating for cannabis as a potential tool to address PTSD and veteran suicide, while contributing practical policy experience from local, state, national, & International cannabis initiatives.
