This study explored the current knowledge, experiences, and attitudes of Australian GPs towards MC, and provides key insights into this emerging field subsequent to our original survey of five years ago [7]. This survey is reasonably comprehensive, with 42-items scoping numerous aspects of MC that are relevant and specific to general practice. Overall, most GPs had received MC enquiries from patients. Yet only a minority felt comfortable managing these enquiries. Only around 21% of the cohort were MC ‘prescribers’, around the same proportion (22%) of the overall cohort reporting adequate self-perceived knowledge around MC in clinical practice. Indeed, many GPs felt they had inadequate knowledge of MC products, MC access pathways and optimal clinical practice. A need for further education in this area was also endorsed in the open-ended comments.

A disproportionate number of MC prescribers resided in Queensland, in agreement with recent government data on prescriber location for SAS-B approvals [5]. Overall, however, GP demographics in the current survey were strikingly similar to our original 2017 survey [7]. More GPs had received recent MC enquiries than in our original survey (85.3% vs. 61.5% respectively in past three months). This observation is consistent with the recent and considerable rise in MC prescribing [2, 10] likely driven by recent initiatives aimed at improving access pathways, as well as the growing acceptance of MC from health professionals [10, 20, 21]. The mental health burden of the COVID-19 pandemic has also been suggested as a contributing factor for the demand for MC products over the past two years [10].

Interestingly, however, GP comfort levels with managing MC enquiries remain low and unchanged since the original survey despite the increase in patient enquiries. Low comfort levels around managing and prescribing cannabis-based therapy are in line with other surveys of physicians [13–17]. Australian GPs currently select from > 240 distinct MC products to treat more than 120 distinct medical conditions and often prescribe in the absence of high-quality supporting evidence of efficacy [3, 21, 22]. The specialised prescribing process and costly nature of MC products may also present hurdles in GP discussions with patients [7, 9, 23]. Most GP participants endorsed these two issues as ongoing problems. Formal education around MC has been suggested to improved comfort levels [14, 15, 24].

The greatest support for the use of MC was in terminal and/or often difficult-to-treat medical conditions such as palliative care, chronic cancer pain and CINV, more so than our original survey (Supplementary Table S4) [7] and similar to other international studies [25–27]. MC is often seen as a therapy of last resort, where conventional treatment options have been exhausted [10, 15, 25, 27, 28]. Furthermore, the evidence around the use of cannabis-based medicine in these indications has been well-reviewed and summarised in a variety of comprehensive guidance documents by organisations such as the TGA [2, 3, 21].

The extent of participating GPs support for the use of MC in anxiety, depression and insomnia was not well-endorsed relative to other indications, perhaps reflecting the limited evidence for MC efficacy in these conditions [29, 30]. Additional factors here may be the availability of relatively safe and effective conventional therapies for these conditions, as well as concerns around the impact of THC use on mental health [16, 27, 31]. When compared to the original survey however, participating GPs demonstrated increased support for the use of MC in most health conditions compared to the original survey (Supplementary Table 4) [7] and this included mental health conditions such as anxiety and PTSD as well as insomnia and chronic pain. This agrees with TGA data showing chronic pain, anxiety and sleep disturbance as the leading indications for SAS-B approvals [5].

Participating GPs who were MC prescribers had prescribed MC most often for chronic pain and anxiety (Supplementary Table 2). However, prescribing for these indications is supported by relatively sparse evidence [32–35]. Indeed, participating GPs were clearly aware of the limited evidence for clinical efficacy of MC, as has been noted in previous surveys [13, 16, 17, 31]. Widespread prescribing of MC in the absence of evidence suggests a patient-driven rather GP-driven process, or possibly that clinical efficacy is readily apparent to GPs but has yet to be properly captured by published clinical trials.

Concerns endorsed by GPs around THC included driving impairment, addiction and dependence; and are legitimate given current evidence [22, 36]. Some GPs expressed concern around THC causing weight gain; however, regular cannabis users tend to have a leaner phenotype than non-users [37, 38]. Legitimate concerns were expressed around drug-drug interactions with CBD [22]. However, concerns expressed around CBD causing driving and cognitive impairment are unfounded [39, 40]. Similarly, concerns existed around addiction and psychosis, yet CBD has shown some promise as a treatment for these conditions [41, 42]. Concerns around MC are therefore not always based on the most current evidence. The lack of endorsement for availability of low dose CBD products in pharmacies may reflect misconceptions around the safety of such products by GPs.

Despite safety concerns, few GPs believed the risk of side effects were “too high” (15.6% in the current survey vs. 19.8% in the original survey). Notably, GPs rated prescription opioids, benzodiazepines, and chemotherapy drugs as more hazardous than MC products, to a greater extent than the original survey [7]. This indicates increasing acknowledgement of the safety of cannabis-based medicines as clinical experience with MC grows.

The low perceived knowledge of GPs around MC, particularly amongst non-prescribers, is perhaps the most striking outcome of the survey and was reiterated in open-ended comments. Notably, most GPs (79%) endorsed the need for compulsory, MC-specific training for prescribers, a belief that has been maintained since the original survey (78.6%). Low knowledge has been highlighted repeatedly in previous surveys of physician around MC [13, 15–17, 26]. It is essential that GPs are supported in developing a sound knowledge around MC, regardless of whether they prescribe MC products or not. Professional organisations will play a leading role in educational initiatives in this area. Knowledge drives optimal patient care [7], and lack thereof may contribute to inadequate provision of support including missed or delayed treatment opportunities [43]. This should be a priority given the ongoing surge in demand for MC therapy.

There are around 4600 MC prescribers in Australia [5] of whom approximately 80% are thought to be GPs (i.e., around 3680) [12]. With a total of 31,000 GPs in Australia [44] MC prescribers were perhaps over-represented in our cohort (21%; 110/505) compared to the overall GP population (12%, 3680/31,000). The 505 GPs included in this study is only a small proportion of the total Australian population of ~ 31,000 GPs [45] such that issues of sample representativeness are worthy of consideration. Government data show Australian GPs to be 41.0% identifying as female, 40.0% aged ≥ 55 years and 32.1% located in NSW [45] compared to 59.8% identifying as female, 51.7% aged ≥ 55 years and 33.5% located in NSW respectively in the current study. Other limitations include recruitment being limited to GPs participating in continuing education and use of an on-line recruitment strategy due to the COVID-19 pandemic that may have biased the cohort towards more technologically adept GPs.