Nearly Half of Older Cancer Survivors Have Used Cannabis, But Fewer Than 1 in 10 Discussed It With Their Doctor

A cross-sectional survey of 524 cancer survivors aged 65 and older at a South Carolina cancer center reveals substantial cannabis use for pain and insomnia, even in a state without legal cannabis access, while exposing a striking gap in patient-provider communication that carries real implications for safety monitoring in a medically complex population.

Why This Matters

Older adults with cancer histories represent one of the fastest-growing populations using cannabis for symptom management, yet they are also among the most pharmacologically vulnerable, with polypharmacy, fall risk, and cardiovascular concerns making unmonitored use particularly hazardous. The fact that substantial cannabis use persists even in states without legal marketplaces underscores that regulatory status alone does not determine patient behavior. This study arrives at a moment when oncology practices are under increasing pressure to develop screening and communication frameworks around cannabis, and the near-total absence of provider conversations documented here makes that need urgent.

Clinical Summary

Cannabis use among older adults has risen sharply over the past decade, but its prevalence and patterns among cancer survivors in this age group, particularly in states without legal cannabis programs, remain poorly characterized. Brooks et al., publishing in the Journal of Geriatric Oncology (2025), conducted a cross-sectional survey of 524 cancer survivors aged 65 and older drawn from the Hollings Cancer Center registry at the Medical University of South Carolina. The study was a secondary subgroup analysis of a larger NCI Cannabis Supplement initiative involving 12 cancer centers, using a probability-sampled design with demographic weighting to partially address non-response. The broad definition of cannabis used in the survey encompassed THC-containing products, CBD-only products, delta-8 THC, and pharmaceutical cannabinoids, reflecting the diverse and often unregulated product landscape patients actually encounter.

Weighted estimates showed that 46% of respondents reported lifetime cannabis use, 18% had used cannabis after their cancer diagnosis, and 10% were current users. Among post-diagnosis users, pain (44%) and insomnia (43%) were the most commonly cited reasons for use, and smoking was the most common mode of administration (40%). Fewer than 3% reported that cannabis worsened any symptom. Critically, only 8% of participants had discussed cannabis use with a healthcare provider. The study’s chief limitations include a 13.4% response rate, a single-center design, reliance on self-report in a state where cannabis remains unregulated (likely biasing prevalence estimates downward), and the inability to disaggregate findings by product type. The authors emphasize that these findings are descriptive and hypothesis-generating, and they call for longitudinal research and clinical communication interventions before any practice recommendations can be drawn.

Dr. Caplan’s Take

What strikes me most about this study is not the prevalence figure itself, which is consistent with trends we have seen elsewhere, but the 8% provider-discussion rate. I regularly encounter older cancer survivors who are using some form of cannabis product and have never mentioned it to their oncologist, their primary care physician, or their pharmacist. When I ask about it directly, patients often say they assumed their doctor would disapprove or that it was not relevant. This study confirms what many of us suspect: the communication gap is vast, and it exists independent of legal status. The finding that fewer than 3% reported symptom worsening sounds reassuring but cannot be interpreted as evidence of safety given the design limitations.

In practice, I treat this as a screening issue. Every older adult with a cancer history should be asked about cannabis use in a nonjudgmental, structured way, just as we screen for supplement use or alcohol. I do not recommend cannabis as a first-line intervention for pain or insomnia in this population given the limited controlled evidence and the real risks of drug interactions, sedation, and falls. But when a patient is already using it, the clinical priority is documentation, safety review, and honest conversation about what we know and what we do not.

Clinical Perspective

This study sits early in the research arc for cannabis use specifically among older cancer survivors, filling a notable gap by documenting prevalence in a state without legal access. It confirms patterns seen in broader geriatric cannabis surveys: pain and sleep are the dominant motivators, patients overwhelmingly perceive benefit, and provider communication is exceptionally rare. What it cannot do is validate those perceptions. The self-report design, inclusive cannabis definition, and absence of clinical outcome data mean that no efficacy or safety conclusions can be drawn. Clinicians should treat the prevalence and communication findings as credible signals warranting institutional response, while recognizing that the low response rate and single-center origin constrain generalizability.

From a pharmacological standpoint, the finding that smoking remains the most common mode of administration among older cancer survivors is concerning, given the respiratory risks and the unpredictable pharmacokinetics of inhaled cannabis in patients who may be on anticoagulants, immunotherapies, or CYP3A4-metabolized agents. The absence of product-type disaggregation is a meaningful limitation because CBD-only products and high-THC products carry fundamentally different risk profiles. The single most actionable step for oncology practices is to integrate routine, nonjudgmental cannabis screening into intake workflows for older patients, ensuring that any use is documented and reviewed against the patient’s current medication list and comorbidity profile.

Study at a Glance

Study Type

Cross-sectional survey (secondary subgroup analysis of NCI Cannabis Supplement initiative)

Population

524 cancer survivors aged 65 and older (51% of parent study, N=1,036)

Intervention

Not applicable (observational survey of self-reported cannabis use patterns)

Comparator

Exploratory age-subgroup comparison: 65 to 74 years versus 75 years and older

Primary Outcomes

Weighted prevalence of lifetime, post-diagnosis, and current cannabis use; reasons for use; provider communication rates

Sample Size

N=524 (from 8,000 randomly selected registry patients; 13.4% response rate)

Journal

Journal of Geriatric Oncology

Year

2025

DOI or PMID

DOI: 10.1016/j.jgo.2025.102212

Funding Source

NCI Cancer Center Support Grant (P30 CA138313) and related NCI supplement funding

What Kind of Evidence Is This

This is a cross-sectional survey study providing weighted prevalence estimates from a single NCI-designated cancer center. It occupies a descriptive tier in the evidence hierarchy, positioned below cohort studies and far below randomized trials. The single most important inference constraint is that this design supports only descriptive characterization of self-reported behaviors and perceptions; it cannot establish causation, temporal relationships, or the efficacy or safety of cannabis use in this population.

How This Fits With the Broader Literature

The prevalence figures reported here are broadly consistent with other recent surveys documenting rising cannabis use among older adults, including work by Han and Palamar (2020) showing sharp increases in past-year cannabis use among adults 65 and older nationally. The provider-communication gap aligns with findings from Pergam et al. (2017), who documented similarly low rates of patient-initiated cannabis discussions in an oncology setting in Washington state, a jurisdiction with legal access. What this study adds is the demonstration that these patterns persist robustly even in the absence of a legal cannabis marketplace, suggesting that legal status may be a weaker determinant of patient behavior than commonly assumed.

The study extends prior work by using a multi-center NCI-harmonized survey instrument, which may enable future cross-site comparisons. However, the single-center design and very low response rate mean these findings should be viewed as preliminary rather than definitive prevalence benchmarks for the broader older cancer survivor population.

Common Misreadings

The most likely overinterpretation is reading the finding that fewer than 3% of post-diagnosis users reported symptom worsening as evidence that cannabis is safe or effective for older cancer survivors. This figure reflects subjective self-report from a self-selected group of users in a cross-sectional survey with a 13.4% response rate. Individuals who experienced adverse effects may have stopped using cannabis and thus would not appear among current users, introducing survivorship bias. The study design cannot distinguish between genuine benefit, placebo response, and recall bias. Similarly, the 46% lifetime prevalence figure should not be equated with 46% of older cancer survivors nationally using cannabis, given the single-center origin and low response rate.

Bottom Line

This study provides credible, hypothesis-generating evidence that cannabis use among older cancer survivors is common and largely invisible to their clinical teams, even in a state without legal access. It does not establish that cannabis is safe or effective for this population. Its most actionable contribution is the documentation of a provider-communication gap so wide that it constitutes an unaddressed patient safety concern, one that oncology and geriatric practices can begin to close through routine, nonjudgmental screening.

References

1. Brooks CM, et al. Cannabis use, perceptions, and provider communication among older adult cancer survivors at an NCI-designated cancer center in a state without legal cannabis. Journal of Geriatric Oncology. 2025;16(4):102212. DOI: 10.1016/j.jgo.2025.102212
2. Han BH, Palamar JJ. Trends in cannabis use among older adults in the United States, 2015-2018. JAMA Internal Medicine. 2020;180(4):609-611. DOI: 10.1001/jamainternmed.2019.7517
3. Pergam SA, Woodfield MC, Lee CM, et al. Cannabis use among patients at a comprehensive cancer center in a state with legalized medicinal and recreational use. Cancer. 2017;123(22):4488-4497. DOI: 10.1002/cncr.30879