This first study following the historical passing of legislative change in Lebanon to legalize cannabis cultivation for medicinal use has highlighted and documented important differences between non-cannabis users, recreational users only, and dual motive users (both recreational and medicinal users). Our findings confirm that dual motive users are a subgroup that warrants further attention. In their study, Turna et al. (2020) found that compared to medical users, the dual motives group reported more daily cannabis use, used more alcohol and tobacco, and were more likely to use cannabis to treat psychiatric conditions [5]. In this paper, we note that even when compared to recreational users only, dual motive users have a riskier behavioural profile and a less conservative general attitude towards legalization and perception of cannabis-related harms. Worth noting that all the dual motive users in our study had first used cannabis recreationally (on average, 2 years before medicinal use) compared to about 75% of an Australian sample [17]. From a harm reduction perspective, this could mean that the “fear of diversion/conversion” from medical to recreational use is less relevant in the case of cannabis since most users begin to us cannabis for non-medical purposes. Indeed, the overwhelming majority of cannabis users reported “experimentation” as their main reason for initiating use. Nonetheless, despite their “benign motives”, all cannabis users in the study are in fact using cannabis “illegally” and are at risk of being prosecuted by law. Understanding patterns of use and motives could, therefore, inform national drug policies, including the Lebanese drug law, which has been revisited and is being reviewed by the Parliamentary Committee for Human Rights [18].

In the United States (US), trends of annual cannabis prevalence in college students have been fluctuating since the 1960s, reaching an all-time high (44%) in 2020 (Schulenberg et al., 2021). In Europe, cannabis remains the most widely consumed substance, with about 16% of young adults aged 15–34 reporting past-year use (ranging between 3 and 23% across different European countries) (EMCDDA, 2022). In Lebanon, though country-level data on cannabis use date back to 2003, [19], more recent university surveys have consistently found cannabis to be the most widely used illicit drug [20]. Besides impact on consumption, legalization of cannabis use, namely recreational, has been associated with more positive health perceptions of cannabis use [21]. Studies from Europe and North America have shown that the legalization of cannabis use for medicinal purposes did influence the perception of risk regarding cannabis [22]. Latest data from the US show that the percentage of young adults (19–30 years) residing in the US who reported marijuana use as “risky” has reached an all-time low (5–8%) in 2020 [23]. In addition, personal disapproval of cannabis use, whether experimental, occasional, or even regular is on a declining trend [23]. Though cannabis is an illegal substance in Lebanon, close to 55% of young adults are supportive of its legalization for recreational use and three-quarters for medicinal use. This is similar to the US estimates where voters supported the legalization of medicinal marijuana at significantly higher rates than recreational marijuana [24]. Overall, the support for some form of legalization is also in line with observations from the US [25], where attitudes have shifted in a liberal direction over time [26].

Studies have shown that about half of lifetime cannabis users proceed to use other illicit drugs [27, 28], which is similar to our findings, although we further highlight an important distinction between the prevalence of other illicit drug use in recreational users only (36%) versus dual motive users (58%). Other cannabis-related harms include a substantial proportion of cannabis users admitting to operating a vehicle within 2 h of cannabis consumption. The previous literature clearly shows how drug and/or alcohol combinations contribute to increased car crashes [29]. Recreational use (with or without concomitant medicinal use) can, therefore, be harmful beyond consumption per se, particularly when users get behind the wheel, highlighting the need for injury prevention programmes to reduce driving under the influence of cannabis regardless of its legal status [30]. The fact that the majority also admitted to intentionally combining cannabis with alcohol or drugs to heighten drug effects also warrants being addressed through harm reduction policies and programmes especially that a recent review concluded that the co-use of alcohol and cannabis is associated with more impairments, higher and more frequent consumption levels, and greater likelihood of comorbidity [31]. Use of cannabis has also extended beyond traditional private locations due to the spread of new methods of consumptions, through edibles and vaporizers, leading to potential for indirect harm [32].

The issue of early substance use, including cannabis, must be addressed within the context of social determinants of health, which recognizes the broader social context of marijuana use as per the socio-ecological model approach [33, 34]. In our study, the predominant majority (almost eight in 10 young adults) reported that they were in the company of a close friend when they first tried cannabis is an indication that young adults select peers with similar current marijuana consumption patterns, and/or adjust to their peers’ group behaviour over time [35].

Other important findings relate to the possibility of significant cannabis uptake following the multiple crises that Lebanon witnessed in 2020, similar to what was witnessed following the 9/11 incident in the US [36]. During the year preceding our online survey, Lebanon had experienced a series of crises starting with the ongoing financial crisis (October 2019), followed by the pandemic (February 2020), and ending with the landmark Beirut Port explosion (August 2020). Inevitably, the multiple crises have negatively affected the mental health of the population, both young people and adults [37, 38]. The link between this succession of adverse events and cannabis initiation cannot be ascertained from our cross-sectional survey. Still, a quarter of the lifetime recreational cannabis users, and 40% of the medicinal users used cannabis for the first time within the year prior to the survey date, and we know from other local data that increase in substance use initiation/frequency were reported [39].

From a health policy standpoint, the extent of self-medication with cannabis among the medicinal users is quite significant, with only 2% having been “prescribed” cannabis by a healthcare provider/pharmacist. While the fact that cannabis use remains illegal in the country and not formally available for prescribing, a prescription may have been issued by a physician from a country where such service is available, or young adults may have understood “prescribed” interchangeably with “recommended” and not an issuance of a formal prescription. Though not prescribed for the predominant majority, this fact does not exclude the possibility that cannabis may have been “recommended” for use, by a clinician or a pharmacist. This is of particular concern since most physicians, as per a recent systematic review, are unaware of the substance’s clinical benefits and adverse effects and hence are not in a position to provide advice [40].

The study has its limitations. Intrinsic to most online surveys is the risk of selection bias. Such an online sample not representing all residents of Lebanon within that age group. The study was conducted following the news of legalization cannabis cultivation for medicinal use, which coincided with the COVID-19 lockdowns; thus, an online survey was the only feasible method of data collection. The second limitation is the complexity in defining recreational vs. medicinal cannabis use. We defined medicinal use as “any consumption to cure, treat or prevent a physical or psychiatric disease” and recreational use as “any consumption for pleasure or to induce an alternate state of consciousness”. In the case of medicinal use, our definition did not distinguish between the use of THC- and CBD-based products. Overlap between these products is anyway well documented and quality control almost impossible [41]. The third limitation, though insightful, is the substantial percentage (36%) of young adults who chose not to disclose their cannabis use status, despite the fact that the survey was online, self-administered, and anonymous. Nonetheless, they completed the survey, and our analyses showed that their profile was similar to the non-cannabis users, which is most likely their cannabis use status. Therefore, their reluctance to report recreational cannabis consumption (and respond to the medicinal use question and all other survey questions) could be due to fear of being legally pursued, as safety and ethical concerns are not entirely resolved in online surveys despite researcher efforts to maintain privacy and integrity [42]. Further analyses showed that non-disclosers were more likely to be younger, of female gender, Lebanese, unemployed, living with parents or guardians, and not using any substances (an additional file shows this in more detail [see Additional file 1]). One solution moving forward is to employ different data collection methods. One study comparing findings from a traditional survey to indirect survey methods (the cross-wise model) found that the reported lifetime prevalence of cannabis use was two–three folds higher when using the indirect survey method, concluding that the latter may provide more accurate estimates keeping in mind both were administered online [43].