#88
High Clinical Relevance
This paper addresses a clinically important real-world question at end of life, but the design only supports cautious, association-level interpretation.
This is a cross-sectional survey of terminally diagnosed medical cannabis patients, not a treatment trial. It is useful for understanding patterns of use and care access, but it cannot prove that cannabis improves outcomes or that it delays or replaces palliative care.
Medical Cannabis
End-of-Life Care
Observational Research
Opioid Use
| Audience | Clinicians, patients, caregivers, palliative care teams, cannabis medicine readers |
| Primary Topic | Medical cannabis and palliative care utilization at end of life |
| Source | Read the full article |
Medical Cannabis and Palliative Care at End of Life: What This Illinois Study Suggests, and What It Cannot Prove
Medical cannabis and palliative care are often discussed in the same breath, especially for people living with severe symptoms near the end of life. This paper asks an important health-services question: among terminally diagnosed patients in Illinois enrolled in a medical cannabis program, was cannabis being used alongside nonhospice palliative care, or instead of it, and what patterns were associated with each path?
This was a cross-sectional survey of 708 terminally diagnosed Illinois Medical Cannabis Program participants who were not enrolled in hospice, and only 115 of them reported receiving nonhospice palliative care. The study does not test whether cannabis works, nor whether palliative care causes better outcomes. What it does offer is a snapshot suggesting that many terminal patients in this state program were using cannabis without formal palliative care, while those in palliative care tended to look clinically more complex and reported somewhat higher self-rated improvement in pain and ability to manage health outcomes. Its biggest limitation is that all major exposure and outcome measures were self-reported at one point in time, so selection effects and confounding are substantial.
This question matters because palliative care is not just symptom control. Good palliative care can also include care coordination, psychosocial support, family support, practical planning, and help aligning treatments with patient goals. If some patients are using cannabis in ways that partly substitute for entering supportive care systems, that has implications well beyond cannabis itself. It touches quality of care, access, stigma, health-system design, and how clinicians frame serious-illness treatment options.
For clinicians and informed lay readers, the deeper lesson is methodological. Studies about medical cannabis and palliative care can sound more decisive than they really are. A patient who is sicker, has cancer, has more gastrointestinal symptoms, or already has better access to certain systems may be more likely to end up in palliative care and also more likely to report different cannabis experiences. That means this paper is useful for hypothesis generation and clinical conversation, but not for making strong claims that cannabis replaces palliative care safely or that adding cannabis clearly improves end-of-life outcomes.
| Study Type | Cross-sectional survey analysis |
| Population | Terminally diagnosed patients enrolled in the Illinois Medical Cannabis Program, largely older adults, excluding hospice enrollees |
| Exposure or Intervention | Reported use of nonhospice palliative care alongside medical cannabis, plus analyses of concurrent opioid use and self-reported cannabis outcomes |
| Comparator | Terminally diagnosed program participants not using nonhospice palliative care |
| Primary Outcomes | Correlates of palliative care utilization, self-reported symptom improvement ratings, and pain levels in patients using cannabis with or without opioids |
| Sample Size or Scope | 708 terminally diagnosed participants included; 115 reported nonhospice palliative care and 593 did not |
| Journal | Innovation in Aging |
| Year | 2022 |
| DOI | 10.1093/geroni/igab048 |
| Funding or Conflicts | Funded by the Illinois Department of Public Health; authors reported no conflicts of interest |
This paper suggests that, within one state medical cannabis program, many terminal patients were using cannabis outside formal nonhospice palliative care, while those in palliative care reported somewhat better pain-related and self-management outcomes. It does not show that cannabis should replace palliative care, and it does not prove that cannabis or palliative care caused those differences.
The authors surveyed adults enrolled in the Illinois Medical Cannabis Program and focused here on 708 respondents who had a terminal diagnosis and were not in hospice. They compared those who reported receiving nonhospice palliative care with those receiving standard care without supportive palliation. They then used logistic regression to identify characteristics associated with palliative care use, ordinary least squares models to examine self-reported symptom improvement scores, and t-tests to compare pain levels among palliative care patients using cannabis with or without opioids. In plain terms, this was a health-services study about care patterns and reported experiences, not a trial of a cannabis product.
Only 16% of these terminally diagnosed medical cannabis participants reported nonhospice palliative care, while 84% did not. In adjusted analyses, palliative care use was more likely among people with cancer, low psychological well-being, medical complexity, and prior military service, and less likely among college graduates, married participants, those without financial insecurity, people using the fast-track application, and those whose certification visit was covered by insurance. Palliative care utilization was associated with higher self-reported improvement scores for pain and ability to manage health outcomes. Among palliative care patients, those using opioids alongside cannabis reported higher pain levels when they initiated cannabis than those not using opioids, which may simply reflect greater baseline symptom burden. The study also found that more frequent cannabis use was consistently associated with higher self-reported improvement scores across several outcomes, but this again is associative and may reflect reverse causation or selection.
This is lower-to-moderate strength evidence for generating clinical questions, not for establishing treatment efficacy. Cross-sectional surveys can be valuable for showing how patients and systems behave in the real world, especially when randomized trials are lacking, but they sit well below randomized trials and well-conducted longitudinal cohort studies for causal inference. Here, both the exposure variables and the outcome measures are largely self-reported, the sample comes from a convenience survey within one state program, and the key clinical interpretation depends heavily on unmeasured confounding.
Several caution flags matter here. First, this was a cross-sectional design, so exposure and outcome were captured at the same general time, making temporality unclear. Second, the data are self-reported, including cannabis use, symptom changes, palliative care use, and pain ratings, which opens the door to recall bias, social desirability bias, and misclassification. Third, there is strong risk of selection bias and nonresponse bias, especially in an end-of-life population answering an online survey. Fourth, the study cannot tell us much about the actual palliative care received, the timing of illness trajectory, local service availability, or whether someone later died or transitioned in care. Fifth, the phrase that cannabis is an “alternative” to palliative care is more an interpretive frame than a directly demonstrated finding. The study shows that many people used cannabis without formal nonhospice palliative care, but it cannot establish that cannabis was the reason they did not receive palliation. Finally, the positive associations with more frequent cannabis use and better self-reported outcomes are especially vulnerable to confounding by expectancy, indication, survivorship, and reporting style.
It does not show that medical cannabis improves end-of-life outcomes better than palliative care. It does not show that cannabis safely substitutes for palliative care, opioids, or other supportive services. It does not establish that medical cannabis and palliative care interact synergistically, nor that access to a cannabis program causes delay in palliative care uptake. It also does not identify which cannabis formulations, doses, cannabinoid ratios, or routes produced the reported experiences.
This study is best read as a window into care patterns, not as a verdict on treatment effect. Among terminally diagnosed participants in one state medical cannabis program, most were using cannabis without formal nonhospice palliative care, and those in palliative care appeared more clinically complex and reported some better self-rated outcomes. That is clinically interesting. It is not proof that cannabis improves those outcomes, nor that cannabis delays or replaces palliative care. The careful takeaway is that clinicians should ask more explicitly about cannabis use in serious illness, ask more explicitly about palliative care access, and avoid treating those as separate conversations.
? Join the Conversation
Have a question about how this applies to your situation? Ask Dr. Caplan →
Want to discuss this topic with other patients and caregivers? Join the forum discussion →
Have thoughts on this? Share it:
Frequently Asked Questions
What kind of study was this?
It was a cross-sectional survey study of terminally diagnosed participants in the Illinois Medical Cannabis Program. That means it captured self-reported information at one broad point in time rather than following patients prospectively.
Did this study prove that cannabis works in palliative care?
No. It found associations between palliative care status, cannabis use patterns, and self-reported outcomes, but it did not test a defined cannabis intervention against a control group.
Did the paper show that cannabis replaces palliative care?
No. The authors suggest cannabis may function as an alternative for some patients because many participants were using cannabis without formal nonhospice palliative care, but the study cannot establish why those patients were not receiving palliative care.
What did the strongest findings actually show?
Participants using nonhospice palliative care were more likely to have cancer, low psychological well-being, medical complexity, and prior military service, and they reported higher self-rated improvement in pain and in their ability to manage health outcomes.
Why is self-report such an important limitation here?
Because patients were reporting their own cannabis use, symptoms, and perceived improvement, the data can be shaped by memory, expectations, coping style, and social desirability. That does not make the data useless, but it does make causal interpretation much weaker.
What does “nonhospice palliative care” mean in this paper?
It refers to palliative care services short of hospice enrollment. The authors excluded hospice participants so they could focus on patients receiving palliative support without being in hospice.
What is the difference between symptom relief and palliative care?
Symptom relief is one piece of palliative care, but palliative care is broader. It can include coordination, communication, emotional support, caregiver support, and goal-aligned planning, not just relief of pain, nausea, or sleep problems.
What should clinicians take from this study?
Clinicians should ask explicitly about cannabis use in serious illness, ask whether palliative care has been offered or accessed, and avoid assuming that one conversation covers the other. This is a study about care patterns and reported experiences, not a basis for sweeping treatment claims.
Does this paper tell us which cannabis product, dose, or ratio is best?
No. It does not provide product-specific efficacy data, cannabinoid ratios, or dosing guidance that could support clinical prescribing decisions.
What kind of research would answer this question better?
Prospective cohort studies and pragmatic trials would be much stronger, especially if they tracked timing of palliative care referral, symptom trajectories, cannabis product details, opioid exposure, caregiver outcomes, and health-system use over time.
