We explored patterns of cannabis use among workers with a work-related injury or illness, a group who commonly experience physical and mental health symptoms for which they may perceive cannabis to be of therapeutic benefit. We found that 14.1% of our sample were using cannabis for their work-related condition, and another 13.3% were using cannabis unrelated to their condition. Compared with other respondents, workers using cannabis for their work-related condition demonstrated poorer health on a range of measures, yet consistently perceived cannabis to have a positive impact on their health. Importantly, among those using cannabis for their work-related condition, only one-fifth reported having a medical authorisation to use cannabis, much like respondents using cannabis unrelated to their condition. Two-thirds of respondents using cannabis for their work-related condition had not received guidance from a healthcare provider.

Findings of this study demonstrate workers are turning to cannabis many months following the onset of their original work-related condition, mostly without medical guidance. While we are not aware of similar studies among workers with work-related injuries and illnesses, results are consistent with studies of other patient populations, which have found that individuals using cannabis for medical purposes are frequently not receiving guidance on their cannabis use from a healthcare provider.37–39 Recent findings from the National Cannabis Survey also found that, among the 14% of Canadian respondents 16 years of age and over using cannabis for medical purposes, 78% did so without medical authorisation.40 Yet, some adults who self-medicate with cannabis also report a desire to access health professionals for advice related to their use.41 A qualitative study of family physicians demonstrated they were often hesitant to guide patients on medical cannabis use, including a reluctance to authorise its use, due to concerns about the potential for harm and a feeling they lacked education about its therapeutic benefit.42 Studies conducted in other jurisdictions and with other specialties have reported similar findings.43 44

Workers in this sample who reported using cannabis for their work-related conditions appear to be struggling with their recovery, demonstrating higher levels of pain, poor mental health and sleep difficulties than other workers, which also parallel the most common therapeutic reasons reported for using cannabis. They were also more likely to report not working at the time of interview. Conversely, when asked about the perceived impact of their cannabis use, the majority of workers using cannabis for their work-related condition reported beneficial effects on their physical and mental health. Approximately a third of these workers also reported decreasing their use of prescription medications and alcohol as a result of their cannabis use, consistent with a larger body of research demonstrating the substitution effects of cannabis.45–47 Unfortunately, as a result of the cross-sectional design of this study, we lack temporal information on cannabis use and these recovery indices, limiting interpretation of the direction of these findings. One potential explanation for these paradoxical findings is that workers with more severe symptoms may be more likely to pursue the use of cannabis as a therapeutic measure, particularly if other treatments have failed.48–50 Furthermore, given workers were asked their perceptions of the impact of use, it is possible that benefits were overstated. Longitudinal studies should be conducted to better understand the impact of cannabis use in this population on objective recovery measures, including the impact on return to work.

Differences in patterns of use among workers using cannabis were also apparent. Workers using cannabis for their work-related condition used cannabis more frequently (70.8% at least once per week) compared with workers using cannabis unrelated to their condition (55.0%). Daily use was also more common among workers using cannabis for their work-related condition (39.5% vs 27.1%). Prior studies have demonstrated variability in frequency of use among adults using cannabis therapeutically.51–57 While daily use among workers using cannabis for their condition in our study was less common than in some studies of medical cannabis use,54–56 it is likely a function of the nature of recruitment, whereby we included workers with varying durations of work disability. As such, workers who participated in this study are not necessarily individuals with chronic, daily symptoms, for whom we may expect daily use to be more prominent. Furthermore, we defined workers as using cannabis if they reported use more than once in the previous year, to ensure we were able to capture the extent and nature of cannabis use in this population for which little data exist. As a result, it is likely we included a diverse group of workers in our sample who ascribe varying degrees of importance to cannabis in treating their work-related condition.

Workers using cannabis for their work-related condition were also more likely to use oral oils, tinctures and capsules, although inhalation methods were still most common, similar to other studies of individuals using cannabis for medical reasons.54 55 57 58 They were also more likely to report either medical use only (44.9%) or mixed medical and non-medical use (46.3%). These findings are consistent with other research, which has shown a large degree of overlap with non-medical cannabis use among adults using cannabis medically,8 58–62 and the blurring of boundaries between medical and non-medical use.63

Our study underscores the need for healthcare providers to be aware that members of their patient population who have experienced a work-related injury/illness may be using cannabis as a therapeutic measure to manage the physical and mental difficulties arising from their condition. With growing public interest in the use of cannabis for medical purposes,4–6 greater accessibility to legal sources of cannabis64 and workers’ compensation organisations increasingly developing formal policies on medical cannabis,65 it is reasonable to assume healthcare providers will continue to encounter workers using cannabis for their work-related conditions in their practice. This represents an important opportunity for healthcare providers to speak candidly with workers about their cannabis use and ensure they are supplied with evidence-based information about potential harms and benefits, to enable an informed choice about safe and effective use. Guidance from healthcare providers is all the more important given information found on the internet directed at medical cannabis consumers may be biased.66–68 In discussing cannabis use with this group of workers, healthcare providers may consider drawing from a recent clinical framework for assessing impairment risk from medical cannabis.69

Our study provides novel information on workers’ use of cannabis for their work-related conditions, a population for which little data exist. Participants were also recruited from a population sampling frame. There are, however, some limitations not yet described. Data were collected at 18 and 36 months after injury/illness onset and we lack information on how soon after injury/illness cannabis use was initiated. We also lack information on the duration of use, including whether there were any interruptions in cannabis use. Information on trajectories of use should be captured in future inception cohorts. Qualitative studies may also help elucidate workers’ decision-making process around using cannabis for their work-related conditions, and the role of their healthcare providers in guiding those decisions. Due to the moderate response rates in this study, selection bias is possible if those who chose to participate differed from those who did not participate. In addition, workers with longer duration claims were oversampled. As such, workers in this study may not be representative of the source population of claimants, potentially limiting the generalisability of the findings. Further, absolute prevalence estimates of cannabis use identified in this study may not reflect the true prevalence of cannabis use in the workers’ compensation system in Ontario. Given this study takes place in the context of legalised cannabis and public healthcare, generalisability of the findings to other policy contexts may also be limited. Finally, the self-reported nature of the data may have led to some recall and social desirability bias.