A large Australian survey found that patients who received prescribed cannabis through specialist cannabis clinics reported lower satisfaction on several care dimensions than those treated in general practice settings. However, the two patient groups differed substantially in age, health status, and reasons for treatment, and the study design cannot determine whether the clinic type itself caused the satisfaction gap.

Specialist medicinal cannabis clinics have proliferated rapidly in Australia and are expanding internationally, yet we know very little about how patients experience these clinics compared to traditional general practice. If a significant portion of cannabis patients are less satisfied with the information, time, and holistic care they receive from specialist clinics, that has immediate implications for how these services are structured, regulated, and improved. Understanding where patients feel underserved is a prerequisite for building better care models in cannabinoid medicine.

The CAMS-22 survey recruited Australian adults who self-reported using prescribed medicinal cannabis, collecting responses online between December 2022 and April 2023. Participants were categorized by their primary care setting: specialist medicinal cannabis clinics (MCC) or generalist health settings (GHS) such as general practitioners. Of the 2,394 eligible respondents, 79.3% accessed treatment through MCCs. The study aimed to characterize demographic and clinical differences between the two groups and to compare their satisfaction with care using Bayesian multilevel regression models, providing an exploratory look at how the emerging cannabis clinic model is perceived by consumers.

MCC consumers were on average 3.5 years younger, more likely to be employed (OR 2.1, 95% CI: 1.8 to 2.5), more likely to meet criteria for cannabis use disorder (OR 1.5, 95% CI: 1.2 to 1.8), and more often treated for mental health conditions rather than chronic pain. Critically, MCC consumers reported significantly lower satisfaction on four dimensions: consultation duration (OR 0.8), information about potential harms and benefits (OR 0.7), integration with other treatment approaches (OR 0.5), and cost (OR 0.6). However, the two populations differed markedly in baseline characteristics, and because patients self-selected into their care setting, these satisfaction differences may reflect population differences rather than care quality. The authors emphasize that these findings are hypothesis-generating and call for controlled studies that can disentangle patient characteristics from service model effects.

This study highlights something I have seen first-hand: when cannabis care is siloed into a specialty visit disconnected from a patient’s broader medical context, satisfaction suffers. The finding that cannabis clinic patients felt less informed about risks and benefits, received less integration with other therapies, and felt their consultations were too brief resonates with a common complaint. It is not that specialist clinics are bad, but when the model is transactional rather than relational, patients notice. The study’s design limitations are real and substantial, though, so I would not use these data to condemn any particular care model.

In my own practice, I treat cannabinoid medicine as one tool within a comprehensive care plan. I spend time discussing both benefits and risks, I integrate cannabis recommendations with other therapies a patient is already using, and I make sure consultation time is adequate for the conversation these decisions require. This study reinforces my conviction that the most effective cannabis care is care that does not treat cannabis as a standalone product but as part of a whole-person strategy.

For clinicians working in or referring to cannabis-specific care settings, this study sits early in the research arc. It is among the first to formally compare patient satisfaction across service models for prescribed cannabis. While the Bayesian approach and large sample are methodological strengths, the cross-sectional convenience-sample design places it near the base of the evidence hierarchy. The fact that GHS users were older, sicker, and more likely to be treated for chronic pain means that any satisfaction comparison is thoroughly confounded by indication and selection bias. Clinicians should view these findings as flags for further investigation, not as evidence that one care model outperforms another.

From a pharmacological standpoint, the observation that MCC consumers were less likely to receive oral cannabis preparations (OR 0.4) and more likely to use inhalation routes raises questions about prescribing norms across settings. Different product formats carry different safety profiles, onset characteristics, and drug-interaction considerations, and clinicians should remain attentive to whether their patients are receiving adequate counseling on these distinctions regardless of setting. The most actionable takeaway from this study is straightforward: clinicians prescribing cannabis should actively ensure that their consultations include explicit discussion of potential harms, integration with existing treatments, and sufficient time for patient questions, areas where cannabis clinic patients in this survey felt underserved.