A Geriatrics Cannabis Clinic Shows Feasibility, But Not Yet Effectiveness

Preliminary data from 144 visits at a physician-led medical cannabis clinic embedded in geriatric primary care describe a medically complex population with high drug-interaction burden, but the study offers no evidence on whether the clinic improves safety or health outcomes for older adults.

Why This Matters

Cannabis use among older adults is rising sharply, yet most of this use occurs outside any structured medical supervision, in a population uniquely vulnerable to drug interactions, falls, and cognitive side effects. Older patients on multiple medications face pharmacological risks that younger users rarely encounter. Despite this, few clinical models exist for integrating cannabis oversight into geriatric care. Research describing how such models can be built and what patient populations they serve is a necessary precondition for the outcome studies that will eventually determine whether supervised cannabis care actually reduces harm in this age group.

Clinical Summary

Cannabis use in adults over 65 has increased substantially in recent years, driven in part by expanding state legalization and growing interest in cannabis for chronic pain, anxiety, and insomnia. Yet older adults carry disproportionate risk from cannabis-drug interactions and central nervous system depression, particularly when polypharmacy is present. Researchers at Virginia Commonwealth University Health developed a monthly, physician-led medical cannabis certification clinic within the Division of Geriatrics, integrating pharmacy-led drug utilization reviews into each visit. The clinic was designed to bring cannabis certification into the electronic health record and primary care infrastructure rather than leaving it to external dispensaries or unmonitored self-use. This retrospective program implementation report, published in the Journal of the American Geriatrics Society in 2025, describes the clinic’s development and the baseline characteristics of the 144 visits completed over 30 months.

The patient population was medically complex: a mean age of 65 years, an average of 20.9 comorbid conditions, 14.7 concurrent medications, and 4.6 drug interactions identified per visit through pharmacy screening. Pain was the qualifying condition for 88.9% of visits, with anxiety and insomnia also represented. Two-thirds of patients were taking CNS depressants, and more than half were on concurrent pain or psychiatric medications. Importantly, this study reports no clinical outcomes, no patient-reported measures, no follow-up data, and no comparator group. It establishes that the clinic can operate and that it attracts patients with substantial pharmacological complexity, but it does not demonstrate whether the model improves safety, reduces harmful interactions, or changes patient behavior. The authors explicitly note that outcome measurement and replication at other sites are needed before clinical effectiveness can be evaluated.

Dr. Caplan’s Take

This paper addresses something I encounter regularly: older patients with complex medication regimens who are already using cannabis or want to, and who have no clinical home for that conversation. The model described here, integrating pharmacy review, geriatric expertise, and cannabis certification into a single primary care visit, is exactly the kind of infrastructure we need. But the study itself gives us process, not proof. We have visit counts and medication lists, not outcomes. I cannot point to this paper and tell a patient that a structured clinic will make their cannabis use safer. What I can say is that the population it describes mirrors the patients I see, and that the risks it quantifies are real.

In practice, I treat every older patient considering cannabis as a polypharmacy case first. That means a thorough medication review with specific attention to CNS depressant load, fall risk, and hepatic metabolism pathways before any cannabis discussion. I welcome models like this one and I hope they generate the outcome data we need. Until then, the clinical work is the same: careful, individualized risk assessment with honest communication about what we know and what we do not.

Clinical Perspective

This report sits at the very beginning of the research arc for structured geriatric cannabis care. It answers the question of whether such a clinic can function inside an academic primary care division, and it characterizes the patient population that seeks this service. It does not answer whether the clinic reduces drug interactions, prevents adverse events, improves symptom management, or changes prescribing patterns. Clinicians should recognize the descriptive data as hypothesis-generating: the finding that patients averaged 4.6 drug interactions per visit, with two-thirds on CNS depressants, quantifies a risk environment but does not tell us whether identifying those interactions changed clinical management or patient outcomes.

The pharmacological considerations in this population are substantial. Cannabis compounds, particularly cannabidiol, inhibit CYP3A4 and CYP2C19 pathways and can alter levels of common geriatric medications including anticoagulants, benzodiazepines, and certain antidepressants. The additive CNS depression risk with concurrent opioids, gabapentinoids, or sedative-hypnotics is well established. This paper does not grade the clinical significance of the interactions it identified, which limits actionability. The most concrete step clinicians can take now is to ensure that any older patient using or considering cannabis receives a formal drug utilization review with explicit attention to CNS depressant burden and cytochrome P450 interaction potential, regardless of whether a specialized clinic is available.

Study at a Glance

Study Type

Retrospective descriptive case series and program implementation report

Population

Older adults (mean age 65, SD 13.8) seen in a geriatric primary care cannabis clinic

Intervention

Physician-led medical cannabis certification with integrated pharmacy drug utilization review

Comparator

None

Primary Outcomes

Feasibility (visit completion) and patient population characteristics; no clinical outcomes reported

Sample Size

144 visits over 30 months

Journal

Journal of the American Geriatrics Society

Year

2025

DOI or PMID

Not provided in source data

Funding Source

Not reported

What Kind of Evidence Is This

This is a retrospective, single-site program implementation report describing a care model and characterizing its patient population through descriptive statistics drawn from 144 clinic visits. It sits at the lower levels of the evidence hierarchy, functioning as a case series without a comparator group, outcome measures, or inferential analysis. The most important inference constraint is that feasibility of operation cannot be conflated with clinical effectiveness. Nothing in this design permits conclusions about whether the clinic improves patient safety or health outcomes.

How This Fits With the Broader Literature

Prior literature has documented rising cannabis use among older adults and has characterized the theoretical drug interaction risks associated with cannabinoid pharmacology, particularly through CYP enzyme inhibition and additive CNS depression. What has been largely absent is research describing structured clinical models for supervising cannabis use in this population. This paper fills a gap by offering the first detailed description of a geriatric primary care cannabis clinic, including its interprofessional workflow and patient characteristics. It extends the observational literature by quantifying the drug interaction burden in a real clinical cohort rather than modeling it theoretically. However, it does not yet connect to the outcome-oriented studies that would be needed to evaluate whether structured supervision translates into measurable clinical benefit.

Common Misreadings

The most likely overinterpretation is reading the high number of identified drug interactions as evidence that the clinic prevented harm. The study reports interaction counts but does not grade their clinical significance, describe what actions were taken in response, or track whether patients experienced fewer adverse events as a result. Identifying an interaction on a drug utilization review is not the same as preventing a clinically meaningful adverse outcome. Similarly, the completion of 144 visits demonstrates that the clinic can operate, but operational feasibility should not be mistaken for evidence that the model is effective, cost-efficient, or superior to other approaches to cannabis oversight in older adults.

Bottom Line

This study demonstrates that a physician-led, pharmacy-supported medical cannabis clinic can be embedded within geriatric primary care and will attract patients with substantial medical complexity and polypharmacy. It quantifies the drug interaction burden in this population but provides no data on clinical outcomes, patient safety improvements, or behavioral change. The model is promising and addresses a genuine gap in supervised cannabis care for older adults, but replication, outcome measurement, and controlled comparison are needed before clinical effectiveness can be claimed.

References

1. VCU Health CAHM Division of Geriatrics medical cannabis clinic program implementation report. Journal of the American Geriatrics Society, 2025. Models of Geriatric Care, Quality Improvement, and Program Dissemination section. DOI not provided in source data.