Cancer Survivors in Pain Are More Likely to Use Cannabis—But Less Likely to Drink | Research

The takeaway

Across two national samples totaling over 5,000 cancer survivors, higher pain intensity was associated with more cannabis and cigarette use but less alcohol use—suggesting pain drives substance-specific self-medication patterns.

Oncologists managing pain in cancer survivors, substance use researchers studying self-medication, and cancer survivors navigating pain management options.

What the researchers found

This study used two large national datasets to examine how pain relates to substance use among cancer survivors—a population with high rates of both pain and substance use but surprisingly little research on the connection.

The pattern was consistent across both samples: cancer survivors with more pain were significantly more likely to use cannabis and cigarettes, but less likely to drink alcohol. This wasn't a random scatter—it suggests deliberate self-medication choices. People in pain may turn to cannabis and nicotine (both of which have analgesic properties) while avoiding alcohol (which can interfere with pain medications and worsen certain types of pain).

The implications go beyond substance use itself. In both studies, cigarette smoking combined with pain was associated with worse mental health outcomes and lower quality of life. Cannabis use combined with pain showed a different pattern—not as consistently linked to negative mental health outcomes as smoking.

The study also examined treatment-related side effects and found that pain and substance use together created a compounding effect on overall burden. Cancer survivors managing pain, substance use, and treatment side effects simultaneously face a triple challenge that current care systems often address in silos.

Why it matters

Cancer pain is undertreated in many patients, particularly as opioid prescribing has become more conservative. Understanding that pain drives cancer survivors toward cannabis (and away from alcohol) has practical implications: if patients are self-medicating pain with cannabis, that's information oncologists need for treatment planning—not just a substance use issue to be addressed separately.

The numbers in context

Study 1: N = 1,252 (88% White, 55% female, 60% age ≥65). Study 2: N = 4,130 (83% White, 56% female, mean age 66). Pain associated with increased cannabis use (p < .003 in Study 1), increased cigarette use, and decreased alcohol use (p < .001).

How the study worked

Cross-sectional analysis of two national datasets: Study 1—1,252 adults from Wave 6 (2021) of the Population Assessment of Tobacco and Health Study; Study 2—4,130 adults from the 2020 National Health Interview Survey. All had lifetime cancer diagnosis. Regression analyses examined associations between pain and use of cigarettes, e-cigarettes, cannabis, and alcohol.

Who was studied

N=5,382 adults aged 18-85, 55% female, US cancer survivors from two national surveys.

What this study cannot tell us

Cross-sectional design can't establish whether pain causes cannabis use or whether cannabis-using cancer survivors report more pain. Both datasets relied on self-reported pain and substance use. Predominantly White samples (83–88%) limit racial/ethnic generalizability. Cannabis use was measured as any use, without dose or frequency detail. "Lifetime cancer diagnosis" includes survivors at various stages, from active treatment to long-term survivorship.

How to read the evidence

Two large national datasets with consistent findings—strengthened by replication but limited by cross-sectional design and self-report.

When this study was published

Published in 2025 with data from 2020–2021, reflecting current substance use patterns among cancer survivors.

The bigger picture

This connects to the cancer-cannabis cluster: RTHC-00158 (THC:CBD for cancer symptoms), RTHC-00161 (medical marijuana laws reducing opioid prescriptions), and RTHC-00177 (patient perceptions of vaporized cannabis for appetite). Together, they paint a picture of cancer patients increasingly turning to cannabis for multiple symptoms—and the medical system slowly catching up to that reality. The substance-specific pattern (more cannabis, less alcohol) also adds nuance to the self-medication discussion.

Replication

Not stated in abstract.

Funding

Not reported in abstract.

Conflicts of interest

Not reported in abstract.

Questions still open

Is the shift from alcohol to cannabis among cancer survivors in pain a positive development from a health perspective? Should oncology pain management protocols formally address cannabis as a potential self-medication tool? Does the combination of pain and cannabis use predict different long-term outcomes than pain and cigarette use?

Read the original research

Relationship between pain and nonopioid substance use in two national samples of cancer survivors.

Cancer, 131(4), e35701

Cancer is a well-respected journal focused on oncology research.

View on PubMed ↗Publisher’s paper ↗

Citation

Powers, Jessica M; LaRowe, Lisa R; Rubenstein, Dana; Paice, Judith A; Hitsman, Brian; Rini, Christine M. (2025). Relationship between pain and nonopioid substance use in two national samples of cancer survivors.. Cancer, 131(4), e35701. https://doi.org/10.1002/cncr.35701