Medical marijuana has been nominally legal in Ohio since September 8, 2016. However, physician engagement, as reflected in rates of CTR acquisition, has been surprisingly weak. This contrasts with pre-program surveys that predicted much more robust support (Ohio Resurvey after Draft Rule 2017; Medical Marijuana Physician Survey 2016). The explicit purpose of the present study was not to explain this discrepancy, but simply to assess the demographic factors that influence physician participation in the OMMCP. The data we present here support the assertion that specialty, medical degree, and age are independently correlated with the decision to obtain credentials to recommend MMJ.

Specialty was expected to be a strong predictor of CTR acquisition. A national questionnaire study from 2005 suggested that OB/Gyn and Internal Medicine physicians were more supportive of marijuana as medical therapy than either Family Medicine physicians or psychiatrists (Charuvastra et al. 2005), although these results were not statistically significant. In the aforementioned New York study (Sideris et al. 2018), it was reported that one-third of registered physicians elected to identify themselves on a public website, and of these, the overwhelming majority consisted of Primary Care and Pain Medicine practitioners. In a small survey of licensed Delaware physicians (who may recommend MMJ without any form of registration or certification), a very different sort of distribution was described, with Emergency Medicine physicians and pediatricians reporting a higher likelihood of recommending MMJ than Primary Care physicians (Michalec et al. 2015). The present study differs from the foregoing reports in that, rather than relying on physician surveys or voluntary reporting, it combines the use of a comprehensive, non-voluntary state database with an objective outcome variable, i.e., CTR acquisition. Despite these methodological differences, some meaningful comparisons to prior reports are possible, if CTR acquisition is taken, at a minimum, to be an indicator of support for MMJ. Indeed, the Ohio State Board of Pharmacy has recently announced that, as of September 2019, 60.2% of CTR holders have issued at least one MMJ recommendation (Miller 2019); clearly the CTR signals both support, and to some extent, an intention to treat with MMJ.

We observe that CTR acquisition is indeed highly specialty-dependent, with a distribution distinct from previous reports in the literature. Based on absolute numbers, Family Medicine and Internal Medicine physicians comprise the majority of CTR holders. However, these specialties collectively represent a plurality of Ohio physicians, and per-specialty CTR prevalence ratios were comparable for Family Medicine, Neurology, Anesthesiology, and Psychiatry. The highest prevalence was found for Physical Medicine and Rehabilitation (PM&R), while the lowest prevalences were noted for Surgery, OB/Gyn, and Pediatrics, and Other. For a small group of specialties (e.g., Dermatology, Radiology, Pathology), very few (0–5) CTRs were acquired.

With respect to medical degree, DO physicians were two to three times more likely than MD physicians to obtain CTRs. This may partly be explained by specialty factors: among CTR holders with the DO degree, a majority were in Family Medicine or Internal Medicine – the specialties with the highest gross numbers of CTRs. However, our multivariate analysis clearly demonstrates a “DO effect” that holds across specialties. To our knowledge, this is the first report in the literature of a difference in MMJ program enrollment between physicians with MD and DO degrees.

Physician age was expected to be another significant demographic factor influencing physician engagement in MMJ programs. In Colorado, a survey of family physicians reported that those between the ages of 20–29 were less likely than other age groups to issue a MMJ recommendation (Kondrad and Reid 2013). In the small cross-sectional study of Delaware physicians, there was no relationship between age and likelihood of recommending; there were, however, differences observed between groups with different levels of practice experience – such that physicians with 11–20 years of experience were most likely to issue recommendations (Michalec et al. 2015). Our study found that physician age correlated independently with CTR acquisition, with CTR holders tending to be, on average, almost 3 years older than non-CTR-holders.

Gender may also be a factor in predicting support for MMJ as well as likelihood of obtaining credentials to recommend it. Although an older US study showed no difference in MMJ attitudes between male and female physicians (Charuvastra et al. 2005), a more recent Israeli survey did find greater support among male physicians (Ebert et al. 2015). A 2018 survey of 167 New York state physicians found no gender differences between registered and non-registered physicians (Sideris et al. 2018). A gender effect was also observed in our data, with male physicians more than 50% more likely to have acquired a CTR than their female colleagues. As gender could not be included in the multivariate analysis, it is not possible to state with certainty that the gender effect is independent of specialty, medical degree. or age. However, given the relatively strong representation of women among the specialties with high CTR counts, as well as among DO physicians, gender is very unlikely to have interacted strongly with specialty or degree in this study. On the other hands, an interaction between gender and age is possible, particularly given that female physicians in the US are younger, on average, than their male counterparts (Young et al. 2019).

The reasons for the specialty, degree, age, and gender distributions reported here were not explicitly addressed in this study. We might speculate that specialty exerts its primary influence as a function of the likelihood of encountering patients with qualifying diagnoses, although specialty-correlated personal and political views, as well as the positions of official organizations (AAP COMMITTEE ON SUBSTANCE ABUSE COA 2015; John et al. 2018; American Psychiatric Association 2018; AAFP 2018) may also play a role. Certainly, the lower prevalences observed for OB/Gyn and Pediatrics are at least in partly attributable to apprehension about using MMJ in sensitive populations – although this seems to contradict findings in previous reports (Michalec et al. 2015; Charuvastra et al. 2005).

More work is needed to help understand the rather more complex questions surrounding specialty differences in CTR acquisition. Initial hypotheses should focus on profiling CTR holders vs. non-holders, exploring the relative likelihoods and experiences of treating patients with qualifying diagnoses, and assessing the knowledge, beliefs, attitudes, and reasoning that drive the CTR decision.

The reasons for the observed DO-MD disparity are likewise unclear. There are no published studies elucidating differences between MDs and DOs with regard to MMJ knowledge, behavior or atttudes. DOs do appear to have more favorable positions on “non-mainstream therapies” (Johnson and Kurtz 2002), but it is unclear whether this translates into a higher level of acceptance of MMJ, or greater willingness to recommend it to patients. Further studies are needed to explore the underlying attitudinal and practice differences that form the basis for the observed “DO effect” on physician MMJ registration.

The finding of age as a positive correlator with CTR acquisition was somewhat surprising to us. The effect, though small, runs counter to the perception that younger physicians will be, in reflecting age-dependent atttitudes toward marijuana in general (Pacek et al. 2015), more favorably disposed toward engagement in an MMJ program. The reasons underlying this finding are likely to be complex. Younger physicians may be less confident about embracing controversial therapies, and more fearful about peer disapproval and perceived threats to their reputations. This is likely counterbalanced, albeit incompletely, by increased conservatism and negative generational attitudes about MMJ among older physicians. The age effect may, however, be a surrogate for experience and time in service, as no age differences in the attitudes toward MMJ as legitimate therapy were observed in a study of Colorado medical students (Chan et al. 2017). And as noted, the potential interaction of age and gender must also be considered. Clearly more work is needed to understand the nuanced factors that form the basis of the effect of age on physician MMJ engagement.

We might speculate that the male-to-female gender predominance among CTR holders relates to gender-specific differences in political or personal views, although female physicians are more likely than male physicians to identify as moderate or liberal (Frank 1999). Female physicians, on the other hand, may be more likely to adhere to evidence-based practice guidelines (Tsugawa et al. 2017), which are notably lacking for most MMJ indications (National Academies of Sciences E, Medicine 2017). These observations might form the basis for further gender-based research exploring the motivations and attitudes of CTR holders vs. non-holders.