This statewide survey of 153 oncology care providers in Minnesota provides important insight into current registration and certification patterns, identifies cost to patients and inadequate research as key clinician-identified barriers to medical cannabis use, and highlights the receptiveness of many providers to additional education and research to help them best provide information about medical cannabis to their cancer patients. Medical oncologists are the main providers registering with the MMCP and certifying cancer patients for use of cannabis, and they are overall highly supportive of our current randomized observational study with medical cannabis.¹¹

In a 1990 survey of more than 1000 oncologists on the use of cannabis for chemotherapy-induced nausea/vomiting, “almost one half (48%) would prescribe marijuana to some of their patients if it were legal.”¹⁵ In a recent survey, more than 90% of pediatric oncologists expressed willingness to help children access medical cannabis.¹⁶ A sample of general healthcare practitioners showed that 83% of respondents support the use of medical cannabis for cancer patients with poorly controlled symptoms, and another recent survey of medical oncologists shows high levels of discussion occurring with patients and support for cannabis use by oncologists.^10,17 Our survey showed that most adult oncology care providers continue to support cannabis by stating they would recommend it in the NEJM case scenario. Sixty-eight respondents identified as medical oncologists were registered with the MMCP (about half of our total survey respondents) at the time of survey completion. According to an ad hoc analysis conducted by the Minnesota Office of Medical Cannabis, as of December 21, 2017, 109 medical oncologists had registered in the program and, of these, 97 had certified at least one patient for a cancer-related condition. Approximately 302 active hematology/oncology physicians are practicing in Minnesota; thus, about a third have certified a patient to date.¹⁸

Although there is growing support for cannabis use in the cancer setting, significant barriers remain unaddressed. A separate survey of pediatric oncologists listed the following concerns about medical cannabis use: (1) absence of standards around cannabis dosing (46% of respondents); (2) children abusing medical cannabis (37%); and (3) fear of being prosecuted by the federal government (20%).¹⁶ Our results indicate that perceived cost to patients for medical cannabis use was the largest barrier to discussing medical cannabis with qualified patients. More than 50% of respondents stated this was a large barrier (rated 4 or 5 on a 1 to 5 scale).

In February 2017, we surveyed 320 patients in our clinic and found that 47% indicated cost as a barrier to use. In Minnesota, patients pay an annual $50–200 registration fee to be in the MMCP, and they typically pay anywhere from $100 to $300 per month (depending on patterns of use) to obtain cannabis from the two licensed manufacturers, who set the prices. These costs translate to an out-of-pocket expense of $3,000 to $3,500 per year for patients who use medical cannabis. These estimates are similar to those from chronic pain patients in New England who reported annual average spending for cannabis of $3,064 (median spending was $2,320).¹⁹

Respondents listed “my health group/leadership does not allow/support providers certifying patients” as the lowest overall barrier to discussing cannabis with their patients. However, about 24% of respondents felt it was a barrier (rating 3, 4, or 5). Of the 19 respondents who said that their primary organization does not allow registration/certification, 17 said that a formal policy precludes registration/certification (5 of those 17 were from the Minneapolis VA, a federal institution that does not allow cannabis use), while two said an informal policy prevented it. Those who noted that the primary organization did not allow registration/certification represented five different healthcare organizations. Even though this was the lowest overall barrier among all respondents, this could be a potentially large barrier in some systems, affecting numerous providers and patients alike.

Inadequate research into the benefits and risks of medical cannabis is seen as both a barrier and an opportunity. A number of cannabis products (with differing amounts of THC/CBD) have been studied in patients and assessed symptoms such as pain, nausea, and appetite loss.^20–26 A few of the largest, randomized controlled trials (RCTs) using nabiximols focused mainly on pain and showed somewhat mixed results.^27,28 A sentinel article from 1980 showed that 80% of patients prefer THC over prochlorperazine when used for chemotherapy-induced nausea.⁵ Subsequent small RCTs in the current antiemetic era have also shown promising results.^29,30

Of note, respondents did not appear to be concerned with the current safety of medical cannabis, as seen through its status as a minor barrier. Data from the Minnesota Department of Health (MDH) support this finding. According to the MDH's report on the first year of the MMCP, from July 2015 to June 2016, “around 20 to 25% of enrolled patients report negative physical or mental side effects of some kind…the vast majority of adverse side effects, around 90%, are mild to moderate in severity.”²

Currently, medical oncologists get little to no formal training in medical school, residency, or fellowship on medical cannabis. However, support for ongoing education was high in our survey, with only 15% of respondents indicating that they were not interested in receiving further education. Furthermore, pharmacists in Minnesota also desire further training and education.³¹ In Minnesota, oncology providers only certify that a patient has a medical condition that allows them access to cannabis, while the cannabis distribution center pharmacist recommends the doses/types of cannabis and monitors for dose modifications and side effects long term. With more states legalizing medical/recreational use, patients are requesting information about cannabis from their cancer team, yet only 15% report receiving it.³ One reason for this lack of dialogue may be due to providers' lack of comfort with their own knowledge. A previous healthcare provider survey conducted in Washington State revealed a low level of self-reported knowledge and comfort in recommending medical cannabis.³² Clinicians lack of solid knowledge may come from many factors, including (1) difficulty tracking the many different types of cannabis programs/requirements (e.g., state laws on recreational vs medical program), (2) the fact that cannabis remains a schedule one drug and, consequently, clinicians have little direct experience overseeing its use, and (3) the number of large, high-quality randomized studies showing both efficacy and safety of cannabis in the cancer setting is limited. While the respondents of this broader discipline survey agreed that clinicians should have structured training and education on medical cannabis, less than a quarter (24%) received information about medical cannabis through lectures or continuing medical education programming. Education seemed to have a clear effect on engagement in medical cannabis prescribing; clinicians who had written authorizations for medical cannabis were more likely to have received training than those who did not write authorizations.³²

Studies conducted among clinicians in Colorado³³ and Canada³⁴ have yielded findings similar to that of the Washington State survey.³² Through this previous research, two clear takeaways emerge: the primary source of surveyed clinicians' source of information is informal channels (e.g., the media, other clinicians, and patients), however, if training were available, clinicians would participate. While our Minnesota survey did not inquire about where respondents were receiving information on medical cannabis, our participants expressed strong interest in receiving additional education about medical cannabis and our state's medical cannabis program. Educating clinicians with reliable, objective information on cannabis is paramount to facilitate improved discussions with patients about benefits, risks, and cost considerations. We plan to use open-ended feedback responses from this survey to conduct focus groups with clinicians and leadership at the MDH to improve educational opportunities regarding cannabis use in the oncology setting.

This study had several limitations. The sample size was small and may not be generalizable to other states. Our response rate was lower than other studies. One reason for the low response rate could be because we did not provide cash incentives for survey completion. While all major health systems throughout Minnesota were well represented through survey participation, we were not able to extrapolate the survey results in a statistically significant way across the Minnesota healthcare landscape. Response bias, implicit in self-selected survey enrollment, presented another inherent limitation for this study; participants could have a disproportionate level of awareness and motivation around the topic of medical cannabis programs and participation. Such response bias could be reflected in participants' overwhelming support for our study and the high support for ongoing education. Of note, however, were respondents' gaps in knowledge about medical cannabis and the MMCP, which could be readily discerned through their responses, particularly the open-ended ones. Moreover, a case could be made that those who are decidedly opposed to cannabis might be more likely to voice their opinions in this survey. That said, ultimately, our study results afforded little opportunity to compare responders and nonresponders. Finally, the NEJM scenario and the research opportunity questions offer limited answer choices and focus on specific indications for cannabis. For example, the trial opportunity discussed only pertained to patients with stage IV cancers requiring opioids. Thus, our survey may not have allowed respondents to give more detailed answers for use of cannabis in a wide variety of situations.