The present study aimed to characterize knowledge and attitudes on medical cannabis among a convenience sample of U.S. healthcare providers, broadly defined. It is important to note that this was a self-selected demographically homogenous sample. Our intentions for gathering these data were primarily to inform ongoing efforts in policymaking, institutional decision-making, and clinical training and practice surrounding medical cannabis. Key findings indicated that most respondents (> 85%), regardless of their role in healthcare, believed cannabis to have legitimate therapeutic uses and had patients who used cannabis. Furthermore, most participants were supportive of legal access to medical cannabis, open to using it in their own practice, and interested in receiving further training to do so. Consistent with these descriptive findings, multivariable regression models showed that openness to clinical use of cannabis was positively associated with greater self-related cannabis knowledge and negatively associated with greater concern about cannabis, suggesting that both perceived knowledge and perceived risk can shape clinical attitudes. Notably, even among highly trained medical professionals (e.g., nurses, physicians, pharmacists), less than a third of participants endorsed formal clinical training as a primary source of knowledge for whether or how to integrate medical cannabis in their practice. Instead, many reported relying on personal experience, popular media, and informal conversations as the primary sources of information on the topic, highlighting a critical gap between clinical practice and institutional training.

While participants’ self-reported knowledge did correlate significantly, though weakly, with their performance on a brief objective knowledge check assessing therapeutic indications, risks, and mechanisms related to medical cannabis, there were inconsistencies in self-report vs. objective knowledge. Specifically, although respondents rated themselves as most knowledgeable about cannabis’s risks, objective knowledge scores were lowest in this domain, suggesting overconfidence and a disconnect between perceived and actual awareness about cannabis-related harms. This discrepancy is further contextualized by regression findings indicating that greater objective cannabis knowledge was associated not only with higher self-rated knowledge, but also greater concern about cannabis, suggesting that increased factual understanding may heighten awareness rather than necessarily openness to clinical use. This misalignment is especially concerning given clinicians’ roles in advising patients on safe and effective cannabis use. Similarly, although self-rated knowledge of therapeutic indications for medical cannabis was also high (rated 4 on a 5-point scale), less than a quarter of respondents correctly identified evidence-based indications on objective knowledge checks. These findings highlight the need for more effective educational interventions that go beyond passive exposure to medical cannabis related information and actively test, reinforce, and update clinical knowledge. Overall, physicians and pharmacists tended to provide more correct responses to objective knowledge checks, consistent with their professional roles in prescribing and/or administering medications. Indeed, regression models showed that physicians performed better than all other health professionals on objective cannabis knowledge questions. Academic research centers, experienced clinicians, and professional organizations were cited as the most trusted sources of information on medical cannabis, suggesting an important opportunity for these groups to develop and offer structured trainings on medical cannabis. Although some healthcare training programs have begun to incorporate cannabis-related content, there is a general lack of structured and standardized cannabis education in health curricula. Additionally, healthcare trainees report limited formal instruction and low confidence in counseling patients about medical cannabis (Zolotov et al. 2021).

A perceived lack of adequately trained providers, potential patient exploitation, and cannabis misuse among patients were commonly noted concerns among healthcare providers. These concerns likely reflect perceived gaps in clinical preparedness and confidence regarding medical cannabis use and are particularly relevant when considering the growing complexity in clinical cannabis care, including a growing variety of available cannabinoid products (Wadsworth et al. 2022), the presence of unregulated cannabis products with documented heavy metal contamination and other toxic impurities (Dryburgh et al. 2018; Gardener et al. 2022), rising cannabis use (Mattingly et al. 2024), and increasing rates of cannabis use disorder (CUD) (Haley 2024). These findings highlight the clear need for structured, evidence-based cannabis education across all levels of healthcare (Marcu 2025). Clinician training should include legal and regulatory frameworks, therapeutic dosing, pharmacokinetics, drug interactions, management of adverse effects, and effective patient communication (Kruger et al. 2022; Baral et al. 2025; Cronin et al. 2025).

Respondents with higher self-rated knowledge and younger age were more likely to express openness to clinical use of cannabis in their practice, possibly reflecting generational differences in medical training as well as general attitudes. Unsurprisingly, those who reported greater concerns were less open to recommending medical cannabis in their practice, with physicians endorsing less openness generally than APPs and registered nurses. Across the board, nearly all participants (96%) supported additional research on cannabis, highlighting a large need for rigorous sources of clinical data regarding both risks and therapeutic applications. Indeed, 76% of respondents reported that their knowledge of cannabis primarily came from personal experience rather than formal education or clinical training. Given the extensive transition to electronic health records (EHRs), we recommend providers be diligent in asking patients about cannabis use and making detailed documentation in the EHR of their cannabis use status, including the type of cannabis (THC-dominant, CBD-dominant, other), source (dispensary vs. unregulated source), route of administration, dose, and frequency of use.

These results are limited by several factors. First, the study used a convenience sample of individuals who responded to online advertising and social media posts related to psychedelic medicines and most participants indicated openness to using cannabis or other cannabinoids therapeutically in their clinical practice. Thus, data likely reflect interest and potentially favorable bias among the sample towards novel therapies or plant-based medicines such as cannabis. Because this survey relied on unverified self-report data, results cannot be strongly substantiated or support robust conclusions. However, survey data showed good internal consistency as well as conceptually compatible associations (e.g., lower openness to clinical use among respondents who reported greater concerns about medical cannabis), providing some evidence for valid responses over and above other data integrity measures. Second, given that eligibility was determined solely by self-report, we were unable to verify that participants were in fact healthcare professionals, which limits the confidence with which findings can be attributed to a true healthcare provider sample. Third, the entire questionnaire was developed specifically for this study, meaning that the self-rated and objective knowledge questions have not undergone any formal validation. Although our findings revealed inconsistencies between self-reported and objective cannabis knowledge, future research would benefit from the development and psychometric validation of standardized tools to assess self-perceived and objective cannabis-related knowledge and attitudes across healthcare professionals. Fourth, the categorization of participants’ professional roles presents a limitation. Participants selected their role from a predefined list of broad designations (e.g., therapist, counselor), which include a wide range of training backgrounds, licensing requirements, and clinical responsibilities. Furthermore, these roles were classified as general categories (e.g., Mental Health Professionals) for analyses. Thus, our findings cannot be used to draw meaningful conclusions across professions. Future research should collect more specific information about professional credentials and responsibilities. Fifth, because the sample was primarily comprised of white female respondents, the data may not generalize broadly to U.S. healthcare providers. Additionally, the heterogeneity of the sample, which included a wide range of healthcare providers, cannot be taken to represent the perceptions of any specific healthcare profession. Nevertheless, the survey was designed to gather information from a broad base of professionals working in the healthcare space in part to examine potential differences across sectors.