Rhetorical and regulatory boundary-work: The case of medical cannabis policy-making in Israel

Social Science & Medicine 217 (2018) 1–9

Contents lists available at ScienceDirect

Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed

Rhetorical and regulatory boundary-work: The case of medical cannabis policy-making in Israel

T

Dana Zarhina,∗, Maya Negevb, Simon Vulfsonsc,d, Sharon R. Sznitmanb a

Department of Sociology, Faculty of Social Sciences, University of Haifa, Social Science Building, Mt. Carmel, Haifa, 31905, Israel School of Public Health, Faculty of Social Welfare and Health Sciences, University of Haifa, Eshkol Tower, Mt. Carmel, Haifa, 3190501, Israel c Institute for Pain Medicine, Rambam Health Care Campus, Israel d Faculty of Medicine Technion, Israel Institute of Technology, Haifa, Israel b

ARTICLE INFO

ABSTRACT

Keywords: Israel Boundary-work Medical cannabis policy Medicalization Legalization Qualitative research methods

Recent studies have explored how professionals draw boundaries to reach workable solutions in conflictual and contested areas. Yet they neglected to explore the relationships and dynamics between how boundaries are demarcated in rhetoric and in policy. This article examines these relationships empirically through the case of medical cannabis (MC) policy-making in Israel. Drawing on interviews with key stakeholders in the MC policy field, formal policy documents, and observations of MC conferences, this article sheds light on the dynamics between rhetorical boundary-work and what we term regulatory boundary-work, namely setting rules and regulations to demarcate boundaries in actual practice. Results show how certain definitions of and rationales for a discursive separation between “medical” and “recreational” cannabis and between cannabis “medicalization” and “legalization” prevailed and were translated into formal policy, as well as how stakeholders’ reactions to this boundary-work produced policy changes and the shifting of boundaries. Both rhetorical and regulatory boundary-works emerge as ongoing contested processes of negotiation, which are linked in a pattern of reciprocal influence. These processes are dominated by certain actors who have greater power to determine how and why specific boundaries should be drawn instead of others.

1. Introduction The term “boundary-work” was coined by Thomas Gieryn (1983) to capture how scientists demarcate discursive or ideological boundaries between science and non-science to maintain an image of expertise, authority and credibility, thereby eliciting financial and political backing. However, the boundary-work concept has since been used to explain a variety of processes and phenomena within and without science (Lamont and Molnar, 2002; Pachucki et al., 2007). Indeed, boundary demarcation is now recognized as a pervasive strategy people use for diverse purposes. Several studies have shown how professionals draw boundaries to reach workable solutions in contested or conflictual areas (Brosnan et al., 2013; Duke, 2016; Ehrich et al., 2006; Hallowell et al., 2009; Wainwright et al., 2006). For example, Wainwright et al. (2006) showed how stem cell scientists presented themselves as ethical by drawing lines around which sources of embryos they agreed to work

with, as well as by deferring to regulatory frameworks. Others found that professionals deflect responsibility onto regulatory frameworks and an abstract notion of “society” in contentious and highly regulated fields (Brosnan et al., 2013; Kerr et al., 1997). Thus “displacement of responsibility” has been recognized as a boundary-work repertoire, i.e., a recurrent pattern of how boundary-work is employed (Frith et al., 2011). Still, regulation does not always map neatly onto individuals’ use of boundary-work as some individuals may draw less or more permissive boundaries in their own work/lives than those of the relevant regulatory framework (Brosnan et al., 2013). Nonetheless, the actual relationships and dynamics between boundary demarcation in rhetoric and in policy remain largely underexplored. This article examines these dynamics empirically through the case of medical cannabis (MC) policy-making in Israel. Using cannabis for medical purposes is a highly contentious issue, replete with scientific uncertainties and ethical ambiguities; the formation of related policies is an ongoing process unfolding before our eyes in many states and

Corresponding author. Department of Sociology, Faculty of Social Sciences, University of Haifa, Social Science Building, 7th floor, Mt. Carmel, Haifa, 31905, Israel. E-mail addresses: [email protected], [email protected] (D. Zarhin), [email protected] (M. Negev), [email protected] (S. Vulfsons), [email protected] (S.R. Sznitman). ∗

https://doi.org/10.1016/j.socscimed.2018.09.047 Received 19 September 2017; Received in revised form 24 September 2018; Accepted 24 September 2018 Available online 25 September 2018 0277-9536/ © 2018 Elsevier Ltd. All rights reserved.

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

countries around the world (Abuhasira et al., 2018; Philipsen et al., 2014; Zarhin et al., 2017). In most places where MC programs have been established recreational use of cannabis remains illicit, so policymakers must decide whether and how to draw boundaries between what are often termed “medical” and “recreational” cannabis. This is not an easy task as these boundaries are blurred (Bostwick, 2012). Thus, MC serves as a good case study to explore how stakeholders demarcate boundaries in both rhetoric and policy in contested and conflictual areas. Reviewing existing MC programs reveals that some programs draw clear and firm boundaries between legitimate and illegitimate uses of cannabis; others do so less adamantly. For example, MC programs in California and Canada have been criticized for making MC so easy to obtain that they produced de facto legalization of all types of cannabis usage (Fischer et al., 2015; Hall, 2015). Williams et al. (2016) have found that programs with a stronger medical orientation have fewer enrollees than nonmedical programs, and argued that states with medicalized programs are less likely than older nonmedical programs to legalize recreational use. Yet whereas some studies have discussed the relationship between medical and recreational cannabis use (Cerdá et al., 2012; Chu, 2014; Johnson et al., 2017; Pacula et al., 2015; Sznitman, 2017), to our knowledge no study has examined how stakeholders address their boundaries in MC policy formation. We draw on the Israeli case of MC policy-making to explore the following questions: (1) how and why do stakeholders demarcate boundaries between medical and recreational cannabis? (2) Is this boundary-work translated into policy, and if so, how? (3) How do stakeholders react to the ways boundaries are drawn and maintained in formal policy and actual practice?

individuals: physicians (n = 8), patient activists (n = 6), past and present government officials in the MC field (n = 3), a Member of Knesset (n = 1), a central figure in the Israeli Pharmacist Association (n = 1), and two individuals at the Israeli Anti-Drug Authority (n = 2). Of these, 18 interviewees were identified as key stakeholders because of their participation in Knesset committees on MC policies. Three more physicians were interviewed because they had previous experience with recommending MC for patients or directly granting MC licenses to patients. Fourteen interviewees were interviewed face-to-face at workplaces or coffee shops; seven chose to be interviewed by phone. Interviews lasted 40–120 min and each provided rich and detailed data. Prior to interviews participants were informed of the overarching research goal (to understand the MC policy process in Israel) as well as the voluntary nature of the interview. Participants signed informed consent forms (face-to-face interviews) or gave oral consent (phone interviews) before interviews. Interviewees were guaranteed confidentiality and anonymity. As the number of persons actively involved in the MC field in Israel is comparatively small, painstaking measures had to be taken to protect their privacy. We removed all identifiers such as sociodemographic characteristics or details of any affiliation with organizations or institutions. We avoided using pseudonyms, which were judged ineffective in disguising participants' identities (Morse and Coulehan, 2015, 151), and instead referred to participants by their social location in the field of MC (such as patient activists, MoH representatives, and physicians). Furthermore, we analyzed protocols of Knesset Committees dealing with MC regulation. The Knesset is the nation's legislative authority and has played a major role in shaping the discourse around MC and its policies. Individuals and groups with a stake in an upcoming topic on the agenda can participate in Knesset committees. The public can access protocols, allowing researchers to explore the diverse stakeholders' voices and perspectives. We searched the Knesset website (http://main. knesset.gov.il/Activity/committees/Pages/AllCommitteeProtocols. aspx), entering the words “cannabis,” “marijuana” and “hashish” in the topics of committee discussions from 1980 to November 2016. Of the 26 results, 21 focused on MC. Document analysis began in December 2016. Note that these data reflect the political and public discourse around MC, but are limited in their excluding stakeholders who did not participate in these committees. Our data also include all documents available on the IMCA website (http://www.health.gov.il/UnitsOffice/HD/cannabis/Pages/default. aspx). These data include government resolutions on MC, and other formal policy documents such as Form 106 which lists diagnoses approved by the MoH as MC indications and the most comprehensive MC document produced by Israeli policymakers to date entitled “Cannabis for medicinal use: An information booklet and medical guidelines” (also known as “The Green Book”). Lastly, the first author attended three international MC conferences that took place in Israel, as well as a lecture by the MoH at the National Institute for Health Policy. The data compiled included materials in both Hebrew and English.

1.1. The Israeli context Israel has run a MC program since the late 1990s, where the Ministry of Health (MoH) is responsible for MC regulation, including granting licenses to patients, growers, and suppliers. Since then, demand for MC has increased steeply, leading to a rise in the number of license applications (approximately 300 a day according to a MoH representative, Welfare committee, January 9, 2017) and licensed patients (approximately 28,000 in March 2017 according to a MoH representative, Medicalization of Cannabis conference, 2017). This expansion pressured policymakers to upgrade MC regulation and monitoring in Israel. The legal status of cannabis in Israel is contradictory. Israeli law follows the UN Single Convention on Narcotic Drugs (UN, 1961) and classifies cannabis as an illicit drug with no medicinal value. However, the government has issued three resolutions acknowledging its medicinal effects. Hence, cannabis remains a liminal substance: a legitimate medicinal alternative and an illegitimate dangerous drug. Responsibility for setting the boundaries between the two was assigned to the MoH, which established the Israeli Medical Cannabis Agency (IMCA). This agency has exclusive authority to authorize licenses for cultivating, producing, dispensing, and researching, as well as using, cannabis. Most physicians cannot authorize patients' licenses, but rather recommend that the IMCA issue these licenses to specific patients. Next we draw on a variety of sources showing how and why the IMCA, and other MC policy stakeholders, engage in boundary-work in both rhetoric and practice.

2.2. Data analysis Data from all of these sources were uploaded to ATLAS.ti (a qualitative data analyses program) and were analyzed following the principles of constructivist grounded theory method, including systematic conceptualization, constant comparisons, coding and memo-writing (Charmaz, 2006; Morse et al., 2009). Coding was inductive and exploratory, aiming to generate theory. In the initial coding stage all data enjoyed equal consideration and all theoretical possibilities were explored. Here a variety of codes emerged, describing how stakeholders defined and distinguished “medical” from “recreational” cannabis, as well as “medicalization” and “legalization” (all emerged as in vivo codes, meaning they were taken directly from respondents' accounts (Charmaz, 2006, 92)). The later stages of coding (focused and

2. Data and methods 2.1. Data collection Data collection began after the research received ethical approval from the Research Ethics Committee at the University of Haifa (#357/ 15). Between December 2016 and September 2017, the first author (D.Z.) and a trained interviewer conducted interviews with 21 2

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

theoretical) included densifying codes into categories and concepts, and moving from a descriptive to a more abstract theoretical level (Charmaz, 2006). The categories “rhetorical boundary-work” and “regulatory boundary-work” were then formulated, with their properties and relationships identified and illuminated. Several strategies were applied to ensure trustworthiness, the main criterion for the evaluation of qualitative studies (Guba and Lincoln, 1989; Rolfe, 2006). First, at the start of the coding two of the authors (D.Z. and M.N.) coded parts of the materials independently, then compared results and settled disagreements to ensure inter-coder reliability. Secondly, interviews continued until theoretical saturation was reached, meaning until the derived data provided a detailed and thick description, properties of emerging categories were developed, and additional interviews did not generate novel insights into the concepts that had emerged (Guest et al., 2006; LaRossa, 2005; Morse, 1995). Thirdly, three of the authors (D.Z., M.N. and S.S.) met regularly for peer-debriefing sessions where they discussed the data and the emerging codes and categories. Finally, the first author kept a diary in which she reflected on the team's decisions and actions throughout the research process, maintaining awareness of whether and how they may have influenced the research project (Charmaz, 2009). By these strategies the authors could maintain a systematic and transparent approach to data collection and analysis.

questions [regarding medical and recreational cannabis] are completely separate … [I]f they were united, our [the MoH's] point of view would be to leave this game because it is not our game. We do not want to be in this game.” The Deputy Director-General was adamant that medical and recreational cannabis could and should be separated, but rather than explain why that was so, he justified the distinction by emphasizing the MoH's reluctance to be part of the “game” of legalizing an illicit drug. In sum, moral and ethical assumptions about cannabis and about the role of medicine, rather than ontological assumptions, were the drivers of boundary-work. Nonetheless, as they clarified in their interviews, MoH representatives realized that to integrate cannabis into medicine without legalizing the substance altogether, an alternative regulatory framework was required. The MoH thus formulated a plan entitled the “medicalization of cannabis”: We introduced the concept of “medicalization” […] to describe a process in which we take something that is more social than medical and turn it into some kind of medical platform. That is, we take something that has a recreational use and turn it into something that is seemingly medical. (Interview with a MoH representative) “Medicalization” thus became a boundary-work mechanism whereby regulators could draw boundaries between “medical cannabis” and “recreational cannabis.” In this process, the loaded term “marijuana” (Thompson, 2013) was replaced with the more neutral term “cannabis,” the name of the plant itself:

3. Rhetorical boundary-work The majority of stakeholders across the spectrum of groups in the MC arena used “medicalization” as a boundary-work mechanism to distinguish MC from recreational cannabis, and emphasized that these should remain separate in both rhetoric and practice. However, stakeholders defined these concepts differently and presented dissimilar motivations and rationales for the required boundary demarcation.

There is a lack of understanding of what medicalization is, as opposed to legalization, which we are not discussing today. Medicalization is precisely an investigation in the light of medical issues, and indeed where medical issues are concerned there are situations in which you get [a medication] and situations in which you do not […] If everyone could get medical cannabis, what kind of a medical process would we have here? (MoH Director-General, MoH, Drugs Committee, March 22, 2016)

3.1. Regulators’ rhetorical boundary-work: advocating medicalization, renouncing legalization

For MoH representatives, legalization is all-encompassing whereas medicalization is restrictive. Medicalization, they stress, is the process through which cannabis is made into a “medication” and thus should follow the well-established biomedical model by prioritizing scientific knowledge and medical expertise, as well as standardization and regulation in defining what MC is (and is not), who can get it or provide it, how and under what conditions (for a further read about stakeholders’ interpretations of “medicalization” please see Zarhin et al., 2017). Applied as a boundary-work mechanism, however, “medicalization” also became a line of demarcation in and of itself, as stated in the English version of the MoH's “Policy Principles of Cannabis for Medical Use”: “Medicalization—the use of Cannabis for medical purposes, not legalization nor decriminalization. The MoH's position is against legalization in view of the potential negative health consequences and harmful effects [of cannabis], especially at a young age” (non-dated, 2). Hence, MoH representatives attempted to separate medicalization from both de jure and de facto legalization:

Ever since the inception of the MC program in Israel, MoH representatives have insisted that medical and recreational cannabis be distinguished. In the first Knesset committee session dedicated to MC, MoH Deputy Director-General informed Knesset Members of the conclusions of an Inter-Ministerial Committee: The notion that marijuana should be given to patients stems from medical professionals' duty to protect life, promote health, prevent suffering, and enable a higher quality of life […] Marijuana use must be for a medical indication and for the purposes of alleviating the patients' suffering from disease […] People who think they will feel better if they take marijuana are not the subject of the Inter-Ministerial Committee. It is a matter of public debate whether or not the use of marijuana should be permitted to the public. This is not the [Inter-Ministerial] committee's business. Rather, its concern is with suffering due to disease and the kind of suffering for which there is evidence that marijuana can ease it. (Drugs Committee, Nov 2, 1999)

[…] We are not discussing the social side [recreational use] because we have no intention of legitimizing any form of legalization, as some try to do by using cannabis for medical purposes as a stepping stone to legalization. Cannabis for medical use is meant to serve sick people and patients. (IMCA Director, Drugs Committee, July 14, 2015)

With this statement the Deputy Director-General defined and demarcated the boundaries between medical and recreational cannabis. As previously observed in other contexts, the distinction rests on the purpose for which cannabis is provided: curative or remedial effects versus pleasure production (Lancaster et al., 2017). This distinction also constitutes the subjects who consume “medical marijuana” as deserving patients diagnosed with specific diseases, in contrast to illicit drug users, i.e., deviant criminals. The former, the Deputy Director-General asserts, should concern the Committee, whereas the latter should be the topic of public debate. The Director thereby called for a discursive separation. In response, Knesset Members and other participants expressed doubt that such a distinction was maintainable, to which the Deputy Director-General unequivocally replied: “For me, these

Other MoH representatives similarly emphasized that decisions on recreational cannabis must be made by the Knesset and the larger society, not the medical profession or the MoH. Legitimization of legalization or de-criminalization through medicalization was presented as an unethical boundary crossing. Through such statements MoH representatives incorporated MC into their jurisdiction, while constructing a different entity—“recreational cannabis”—and excluding it from their responsibility. They thereby expanded medicine's boundaries 3

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

to include MC, while reinforcing medicine as a profession that provided legitimate treatment for diseases rather than allowing or supporting the illegitimate use of illicit drugs for non-medical purposes. Thus the MoH opposed “medicalization” to “legalization” in its rhetorical boundarywork, to displace moral responsibility for the formation of recreational cannabis policies; it assigned it instead to politicians and the “larger society.” The MoH thereby sanctioned the preservation of medicine's social legitimacy and support, as illicit drug use remains highly stigmatized and legalization a contentious issue. Representatives of the MoH also called for the restriction of access to MC to prevent expansion of the field which may blur the boundaries between medicalization and de facto legalization; they cautioned: “if we are not careful, there will be cannabis in every household”:

users. Others worried that such boundary-blurring would deter regulators and physicians from easing patients’ access to MC. Therefore several MC advocates specifically requested that recreational cannabis and its legalization not be discussed in Knesset committees they attended, and emphasized the importance of distinguishing between the two: Look, in all my activism and public speaking, I never refer to legalization at all […] I'll explain why. It is unfathomable that a question about providing medical care be a question about whether the substance is criminalized or not. It just cannot be. It does not exist in any other part of the medical world. (Interview with a patient activist) Similarly, other advocates clarified that the guiding principle in MC policy should be the patient's good, not the substance's legal or moral status. They also argued that using MC as a stepping-stone to legalization would be a manipulative, therefore immoral, use of patients' suffering:

The issue of expansion [increasing numbers of patients and indications] is an issue that stands before you as Members of Knesset. At some point in time—I cannot say when— there will be ten thousand patients, forty thousand patients. Only yesterday, the Police Superintendent and I discussed this subject—we are actually legalizing cannabis through the backdoor. This is a social decision that the Knesset needs to consider, and perhaps especially this committee. (MoH representative, Drug Committee, March 6, 2012)

Making patients the first group to lead a process of legalization is manipulating patients. We must not manipulate patients politically this way because a decision about legalization is a political decision and a social decision and is distinct from a medical decision to allow patients to use cannabis for medical purposes. (Interview with a physician)

Regulators' underlying assumption was that such expansion of cannabis use as well as MC diversion and patients' misuse of cannabis all endanger public health and safety. This perception is also reflected in the MoH's declaration of dual commitment to “provide patients with an appropriate supply of cannabis for medical purposes while protecting public safety and preventing the diversion of the drug to nonmedical purposes” (GR # 1050, 2013). To strike a balance between the potential benefit to individual patients and the potential risk to public health, officials concluded, boundaries between medical and recreational cannabis as well as medicalization and legalization must be demarcated and maintained.

In such statements, MC advocates made moral claims regarding how discussions around MC should be handled. They emphasized the importance of separating medicalization and legalization for the good of the patients, constituted as worthy subjects who used MC because they needed rather than desired it. 3.3. Counter-discourse: boundary-blurring Despite these calls, some stakeholders, including patient activists, Knesset Members and physicians, presented a counter-discourse: medical and recreational cannabis, as well as medicalization and legalization, were and should be intertwined. For example, some Knesset Members emphasized that as long as recreational cannabis is criminalized, cannabis stigma would persist and affect not only recreational users, but also MC patients (Cann10 2nd International Medical Cannabis Conference, 2017). Medicalization, they hoped, would serve as a gateway to legalization. Similarly, a patient activist explained in an interview that to his mind, recreational legalization was desirable and interwoven with medicalization. He hoped that because of medicalization “everyone will see that they needn't be afraid of cannabis, and that it doesn't result in drug addicts on the streets […] Legalization will then stop being a dirty word.” These stakeholders believed that both medicalization and legalization were required to protect public health and wellbeing. “In reality,” they argued, no boundaries exist between medical and recreational cannabis because the same consumers may use the same substance for both purposes. Hence, pretending that boundaries exist harms all cannabis users as well as society at large, for example through wrongful arrests with their negative financial and social consequences, as well as the formation of contradictory and incoherent policies regarding cannabis use (for further read on the effects of incoherent MC policies in Israel, please see Zolotov et al., 2018).

3.2. MC advocates’ rhetorical boundary-work: lobbying for patients MC advocates, including patient activists, certain medical professionals, and a few Knesset Members, also used medicalization as a boundary-work mechanism and called for its separation from legalization. Yet they differed from the MoH in how they defined medicalization and in their motivations and rationales for employing boundarywork. These advocates viewed medicalization as the integration of cannabis into medicine (rather than its being made a medication) and suggested that this be done through co-optation. They argued that cannabis should be incorporated into medicine like other non-medication remedies lacking the scientific basis required by evidence-based medicine, such as complementary and alternative medicine. Cannabis, they asserted, should be considered a viable medicinal option provided by physicians in medical facilities (for a further read about medicalization as co-optation please see Zarhin et al., 2017). Unlike the MoH, most of these advocates did not necessarily see wider cannabis use as problematic or worrisome per se, but continued to call for a discursive separation between medical and recreational cannabis as well as medicalization and legalization. Their motivation for this boundary-work focused on patients’ benefits. As one patient activist put it:

4. Regulatory boundary-work The authority to demarcate the boundaries between medical and recreational cannabis was assigned to the MoH from the start of the Israeli MC program. The MoH thus became empowered to “impose” boundary change by drawing lines where they did not previously exist (Tilly, 2004). Through a process we term regulatory boundary-work, the MoH has translated its rhetorical boundary-work into rules and regulations intended to maintain this separation in practice (see Fig. 1). Specifically, the MoH has used several strategies to define what MC is, who can get it, how and when, thereby demarcating both symbolic and

As long as legalization and medical cannabis are not separated, nothing will come of [our discussions], and nothing will help us [patients]. The two must be separated. Legalization should be discussed separately and medical cannabis needs to be discussed separately. (Patient activist, Drugs Committee, March 22, 2016) Some advocates were concerned that blurring the boundaries between medical and recreational cannabis would reinforce stigmatization of MC patients because they could be “mistaken for” recreational 4

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

social boundaries. The emerging construct of “medical cannabis” thus refers not to a specific substance (as the same substance can be used recreationally as well) but to a classification system that lists the conditions under which cannabis is a legitimate and legal treatment while creating a hierarchy of who can classify cannabis as such.

prevent overarching expansion. 4.3. Physicians as gatekeepers To obtain a MC license, patients must find a specialist who supports their application. Specialists have an important role in the MC arena, but are still subject to the MoH and its medical experts. Specialists wishing to recommend MC for their patients must meet specific requirements: they need to establish a bona fide relationship with patients by treating them for a period of time before requesting the IMCA to approve these patients' license applications (the specific treatment period varies by condition). Physicians must review patients' medical records and add an “Annual Medication Issue Report” to each application attesting that the patient has indeed bought the prescribed medications. Patients’ accounts of whether and how they took their prescribed medications are not considered sufficiently reliable. Physicians are also expected to probe for a history of drug use and avoid recommending MC for patients with a history of addiction (“The Green Book,” 2016). Licenses must be renewed every few months (from 3 to 12 depending on the condition) through a similar renewal process. These requirements allow the MoH to distinguish medicalization from legalization by ensuring MC is used only for medical purposes and as a last resort, identifying malingerers and actual or potential drug dealers, and detecting the potential for addiction.

4.1. State-authorized license system The state, represented by the MoH, heads the hierarchy of decisionmakers empowered to authorize or refuse MC licenses. The IMCA along with other authorities, such as the Ministry of Public Security and the Ministry of Agriculture, set particular demands on growers and patients to ensure that cannabis is used only for medicinal and not recreational purposes. Patients’ licenses to grow cannabis for medicinal use have been revoked since 2007 for better monitoring of the field (Interview with a MoH representative). Commercial growers, who since then have been responsible for cultivating and dispensing cannabis, are required to uphold high security standards to prevent diversion, misuse by workers and theft. They are certainly not allowed to sell cannabis to unauthorized individuals. Security requirements include the installation of cameras, armed guards, high walls, sudden police checkups, etc. As policy documents clarify, anyone wishing to engage in MC cultivation, production or dispensing must be approved by the police. They are checked for criminal records and connections with criminals that may constitute grounds for rejection of their license applications or revocation of existing licenses. These strategies are part of the process of demarcating the boundaries between legitimate MC industry and illegitimate black market. The IMCA also functions as the main authority that grants or rejects patients' MC license applications. Currently, except for 11 oncologists, most physicians can only recommend cannabis for their patients. To do so, specialists must provide the IMCA a computerized form detailing patients’ condition, indication of the failure of a specified number of conventional treatments, and absence of contraindications (such as previous drug abuse and addiction). IMCA staff physicians review the applications and decide whether to approve or reject it. This approval system is meant to ensure that only patients who meet specific criteria receive a license, thereby curbing the number of cannabis users, restricting expansion of the MC field, and preventing “legalization through the backdoor.”

5. Envisioning alternative boundaries As portrayed below, the way the MoH draws and maintains boundaries in policy and actual practice was highly controversial for other stakeholders, including patients, physicians, and Knesset Members. In their critiques of the regulatory boundary-work they envisaged boundaries shifting and a variety of alternative boundary-work strategies. Analysis of their accounts illuminates how MC might have been enacted and constituted otherwise (Lancaster et al., 2017) and how such vision helped reconfigure boundaries, at least to some extent (see Fig. 1). 5.1. Redefining the state's role MC advocates, including patients and physicians, criticized conferment of full control on the state. They argued that physicians' inability to authorize MC licenses directly deprived them of their medical judgment, and the encroachment on medical jurisdiction jeopardized medicine's status as a profession and patients' health; their treatment was determined by bureaucrats rather than their own physicians:

4.2. Use of medical experts’ knowledge To help draw the boundaries between legitimate and illegitimate uses of cannabis, the MoH founded the “Indications Committee,” comprised of medical experts who review the available evidence and determine which diagnoses should be approved as MC indications. As the “Policy Principles of Cannabis for Medical Use” document specifies, “Medical indications for prescribing Cannabis are established by professional authorities in the MoH and supervised for adequacy of knowledge and medical practice” (Form 105, sic). Thus, following the logic of biomedicine (Kleinman, 1995), policymakers use specialists’ authority to distinguish medical from recreational cannabis while the MoH holds a supervisory role. The list of approved indications undergoes updating when accumulated information justifies it. Indeed, demarcation between approved and disapproved indications constantly shifts as previously included indications are omitted (e.g., Glaucoma) and new ones inserted (e.g., Post Traumatic Stress Disorder). Nonetheless, except for treating side-effects of chemotherapy, cannabis is approved only as a last-resort treatment and patients must show that they have exhausted “conventional” treatments: “As a general rule, license to use cannabis will be given only after exhausting all usual treatments and only for approved indications.” MC is thus defined as a treatment available to patients with specific diseases that resist conventional treatments. This definition narrows down the number of eligible patients in a way intended to

The Minister of Health comes [to the Committee] and says: “You physicians, you're exaggerating. Cannabis is a drug. I will not allow using it massively. It will poison the population. We [the MoH] will decide who deserves to get it.” I don't think the Minister understands what he is talking about. You [the Minister] want to take therapeutic responsibility? […] The MoH should not enter into the relationship between caregiver and patient because that is not the Ministry's role; its role is regulating. (Interview with a physician) MC advocates likewise criticized the MoH's contradictory and paradoxical position. On the one hand the MoH is a regulatory body, with authority to decide national policy including MC policy. But on the other hand the IMCA also intervenes in practicing physicians' medical decision-making. Advocates pointed out that this was a unique situation in the Israeli medical system, and stressed that it should be changed. MC advocates also argued that the consequences of giving full control to state agencies led to over-bureaucratization and inefficiency of the process and general lack of transparency, including rejection of patients' applications without proper explanation. In this “over-regulated” bureaucratic regime, advocates concluded, the patient's good ceased being the guiding principle of medical practice: “the patient gets 5

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

lost in the process.” To re-humanize MC medicine, they insisted, physicians should replace the state as the main authority. By contrast, other stakeholders, including representatives of the Israeli Anti-Drug Authority and several physicians, expressed concern about public health, insisting that state agencies should take greater control of the MC arena. MC regulation, they asserted, was too permissive, and its implementation too tolerant. By their account the MoH “turns a blind eye” to violations of the law and to transgressions of the boundaries between medical and recreational cannabis. For example, security measures at growing sites were not properly enforced, and licensed patients caught misusing their cannabis supply were not indicted. These stakeholders expressed concern that medicalization in Israel had contributed to a rise in recreational use of cannabis and had essentially become de facto legalization.

believe their patients or not. Their decisions are based not on biomedical or scientific criteria but on trust. However, trust, both physicians and patients emphasized, was eroding in the MC arena in which all stakeholders were suspected of having hidden agendas. They explained that the MoH's concern about maintaining boundaries between medical and recreational cannabis led to constant suspicion that patients and physicians would cross these boundaries, thereby jeopardizing the reputation of both. For example, physicians had been suspected of recommending MC for financial gain and were derogated by their peers: You are actually being ostracized by the community, rejected by the medical community. For example, I know that there are some people, colleagues of mine, who look at me and say, “Ah, this is a cannabis doctor.” They say it with derision. They have contempt for this whole thing. (Interview with a physician)

5.2. Considering types of knowledge in addition to medical experts' opinions

This “contempt” derives from suspicion of physicians’ ulterior (financial) motives and from lack of solid scientific evidence of MC efficacy and safety: evidence that the Western medical community deems essential for the practice of medicine and its ethical grounds. Patient activists, in turn, feel they are frequently suspected of abusing cannabis, by personally using it for non-medical purposes or by supplying it to unauthorized others:

MoH critics challenged the epistemic authority given to certain medical experts in boundary demarcation, and expressed reservations about the approved list of MC indications. For example, in interviews several physicians argued that including indications in the absence of scientific evidence, primarily randomized clinical trials, had led to a reality where “we are essentially conducting a clinical trial on humans on a national level.” Similarly, at an international conference held in Israel, a physician asked the MoH Director-General why Inflammatory Bowel Disease was included in the list of indications despite the “weak evidence,” and accused the MoH of being irresponsible. By contrast, MC advocates criticized the omission of several medical conditions that could be improved if treated with cannabis products. Cannabis, these stakeholders insisted, was a much more potent medicine than current definitions allowed. Patient activists and some physicians suspected that MC policy and the medical experts who implemented it were not guided by medical or scientific considerations, but by social or moral ones (e.g., concerns about legitimizing and increasing cannabis use by the public, diversion, etc.):

The physicians' approach is not to look at the patient and see if anything else helped him or to check if the patient really needs cannabis or if you, as a physician, think the patient needs it. Rather, it is to go and check if your patient is actually a drug dealer who is fooling you in trying to get you to increase his dosage from 20 to 30 g a month. Why does the doctor have to think about such things? Does he think about these things when he increases patients' dosages of morphine? (Interview with a patient activist) As these excerpts make clear, patients and physicians alike feel their experience is read through a moral discourse that casts them as potential offenders who pose a threat to society. Hence, in their public appearances, such as in the media or Knesset committees, patient activists strive to present their MC use as legitimate, fending off accusations that they or other MC patients are immoral individuals; other patients simply choose to conceal their use of MC, thereby unintentionally reinforcing the cannabis stigma. Still, although they shared criticism of the current regulatory boundary-work, physicians envisioned shifting the boundaries in different directions. Some said they opposed the use of cannabis for recreational purposes but hoped for legalization because they believed it would simplify medical practice with cannabis. As one physician explained:

It is true that there is an Indications Committee and that the committee is comprised of experts from each specialty. However, these experts are guided to a certain degree or to a very large degree by a rather conservative worldview. (Physician, Welfare Committee, July 13, 2015) MC advocates argued that policy-making processes around MC should encompass a larger number of voices, including those of clinicians and patients who had gained expertise through experience. Clinicians gained knowledge through their hands-on experience treating patients with cannabis, whereas MC patients acquired embodied knowledge obtainable only through regular use of cannabis. These types of knowledge, advocates maintained, should be guiding forces in regulatory boundary-work as they evinced the effectiveness and safety of cannabis use and indicated that it should be given much more liberally—for a greater variety of indications and as a first or second line treatment rather than last.

Leave the doctors out of it […] I will support medicalization after there is legalization, not before. Medicalization is a euphemism, a laundered word. Legalization will clean up all the people who come here with ulterior motives. But other physicians believed legalization was not the answer, yet opposed medicalization until cannabis products resembled conventional medications, namely they were standardized, their makeup was wholly known and understood, and their efficacy and safety were proven against scientific standards. These boundary contestations helped reconfigure boundaries as they related to various changes in MC policies over time. For example, additional diagnoses were approved as MC indications; the MoH has initiated a training course for 100 physicians who on completing it will be authorized to grant licenses directly to patients; and patients are now allowed to carry up to 5 g of MC for personal use with them and to use non-smoked cannabis outside their homes—all of which were forbidden until recently. Moreover, the MoH has announced its plan for an elaborate reform whereby MC will include only standardized products that will be sold in pharmacies across the country. These changes led a number of interviewees to argue that “we are currently in a transitional period, but we don't know yet where we will end up.” The boundaries, in other words, are not yet stable but in flux.

5.3. Redefining physician-patient relationship Physicians expressed discomfort with their role as gatekeepers. Some were worried that in practice drawing the boundary between medical and recreational cannabis usage was difficult; they objected to their dual role as physicians and “cops,” charged with detecting which patients were honest and which malingerers or drug dealers. Physicians declared that this role also put them at personal risk because patients whose cannabis requests were rejected sometimes became aggressive and violent. Physicians were also concerned about the “gray areas,” where it was difficult to ascertain whether cannabis use was medically necessary. For example, evaluating patients’ pain and suffering is problematic because these experiences are subjective and cannot be measured with objective tools; physicians have to decide whether they 6

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

Fig. 1. Boundary-work in policy-making.

leads to multiple conflicts among other stakeholders. The significance of rhetorical boundary-work is highlighted by presentation of its practical consequences for policy development, which emerges as an ongoing contested process of negotiation. Identification of the dominant actors in this process elucidates how and why specific boundaries are demarcated instead of others, underscoring the role and significance of the specific socio-cultural and political contexts in which boundarywork is performed, including the existing power differentials and moral order. Thus, this study extends Gireyn's insight—that boundaries are ambiguous, flexible, contextually variable, and disputed (1983)—to regulatory boundary-work, and sheds light on how and why specific boundaries are continually drawn, negotiated and modified. In the case of Israel, the MoH is the dominant actor in policy-making processes around MC. MoH representatives attempted to draw boundaries between medical and recreational cannabis by using medicalization as a boundary-work mechanism. They emphasized that decisions on MC should be taken by medical professionals who should follow medical considerations, whereas decisions regarding the regulation of recreational use should be taken by the larger society, more particularly its representatives, Knesset Members. The latter should take into account medical considerations, but also social and moral ones. Through this rhetorical strategy MoH representatives displaced moral responsibility for the formation and monitoring of recreational cannabis policies, incorporating under its jurisdiction cannabis use for medical purposes only. Therefore, as shown in other studies (Frith et al., 2011), the displacement of responsibility appears as a boundary-work repertoire. The MoH thus expanded medicine's boundaries by incorporating MC, while solidifying medicine as a profession that treats sick patients and preserving its social legitimacy and support. Nonetheless, this study shows that the attempt to maintain the boundaries between medical and recreational cannabis in effect ensures that recreational cannabis use is ever-present in MC policy processes. Hence, although policymakers argue that medical policy should be guided by

6. Discussion and conclusion This paper draws on, and makes theoretical and empirical contributions to, a few bodies of knowledge, including boundary-work, sociology of health and illness, science and technology studies, and MC literature. As the analysis indicates, the making of MC shares similarities with other contested areas in medicine and science characterized by ongoing boundary-work, such as stem cell research (Wainwright et al., 2006), complementary and alternative medicine (Shuval et al., 2012), infertility (Ehrich et al., 2006) and the use of methadone in addiction treatment (Fraser and Valentine, 2008). Like these and other phenomena (Barad, 1998), MC does not preexist as such and is not an ontologically stable object, but is constituted through and within particular material-discursive practices of boundary demarcation. In this sense, this paper shares Gieryn's original goal of understanding how boundary-work is accomplished in practice (1983), but goes further, to examine the relation among different types of boundary-work, a venture that so far has received relatively little research attention (Lindberg et al., 2017). Specifically, although a large body of work discusses boundarywork (e.g., Hobson-West, 2012; Wainwright et al., 2006; Wilson et al., 2014), we know little about the dynamics between rhetorical boundarywork and what we term regulatory boundary-work. This paper addresses this gap by showing that both types of boundary-work are processes in the making, where different stakeholders draw on dissimilar definitions, motivations and rationales, as well as mechanisms and strategies. Findings further indicate that rhetorical and regulatory boundary-works are linked in a pattern of reciprocal influence. Stakeholders make claims about how and why boundaries should be drawn, and these are then translated into policy which stakeholders may accept or contest, leading to policy formation, changes or stagnation (see Fig. 1). As this case shows, regulatory boundary-work may reflect the rhetorical boundary-work of a specific group or actors in a way that 7

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al.

medical considerations only, it is in fact guided by social and moral considerations as well, thereby reproducing a social and moral order in which cannabis is seen as a dangerous drug. This research therefore joins previous efforts (Fraser and Valentine, 2008) to enhance understanding of how the social, the moral, and the medical are mutually constituted. Yet this paper goes beyond current literature, to highlight how regulators translated their own understanding of and rationales for demarcating boundaries into formal policy by setting rules and regulations to achieve this separation in practice. Regulatory boundarywork was therefore employed to distinguish medicalization from legalization/de-criminalization in a process guided by the assumption that liberal use of cannabis, as well as its diversion and misuse, threatened public safety and health, and that MC policies should focus on preventing these effects while providing a specific and limited patient population with MC. Here policymakers defined what MC was and what it was not, and drew boundaries between the two categories. The emerging construct of MC refers not to a specific substance, as is often perceived, because the same substance can be used for non-medical purposes too. This construct refers instead to a classification system that encapsulates guidelines on how to categorize legitimate versus illegitimate uses of cannabis as well as a hierarchy among actors who may devise and implement this classification. So whereas the term MC may be used universally to signify the use of cannabis for medical purposes, its local uses and meanings differ across time and place. This is significant, as the particular ways in which MC is enacted constitute specific objects and subjects in any given context. Findings indicate that stakeholders were generally critical of the ways the MoH drew and maintained boundaries, as well as of the hierarchy constructed by regulatory boundary-work: a hierarchy in which the IMCA as a representative of the state was at the top, followed by medical experts and physicians. Stakeholders expressed reservations about MC policies, accusing the MoH of harming patients’ wellbeing by excessive lenience or stricture. Furthermore, the concern that stakeholders would trespass upon the boundaries between the medical and recreational uses of cannabis created a moral landscape rife with suspicion and distrust. Thus, in this case boundary-work serves as a practicable solution allowing the state to provide cannabis to specific individuals while maintaining its illegal status; however, this boundarywork also has negative consequences for patients and physicians who feel they are stigmatized as potential criminals, immoral human beings, and a danger to society. In their critical accounts stakeholders envisioned the shifting of boundaries, but in different directions. Though non-consensual, the rhetorical boundary-work done in response to regulatory boundary-work fed back into regulatory boundary-work and helped modify it. This analysis therefore echoes previous findings that elucidate how MC might have been enacted and constituted otherwise, in a way that opens up hitherto silenced discourses and highlights the political, contingent and emergent nature of this new entity (Lancaster et al., 2017). Research shows that where recreational cannabis remains illegal, MC programs continue to be contested. Some stakeholders call for greater tightening of the boundaries between medical and recreational cannabis; others call for their loosening (Fischer et al., 2015; Hall, 2015). Some stakeholder groups actively try to change boundaries by engaging in lay research (Sobo, 2017); others do so through participation in expert committees (Taylor, 2016). This article sheds light on the sources of and processes in which these controversies emerge and develop, drawing attention to who is responsible for demarcating and policing boundaries in rhetoric, formal policy and its implementation, and how this is accomplished. The study further highlights the role of stigma in the emergence of such contestations: stigma was a concern for most research participants, and was employed to justify both the relaxing and the tightening of boundaries. Accordingly, how different regulatory frameworks affect stigmatization remains an empirical question for future research. This study also opens the door to further

research on the relationship and dynamics between rhetorical and regulatory boundary-works as well as their consequences for involved stakeholders inside and outside the global MC arena. Acknowledgements This research was funded by the Israel National Institute for Health Policy Research [grant number #131-15]. The funding source had no involvement in the study design, analysis or interpretation of the data. We wish to thank a number of individuals for their important contribution to this paper: Yuval Zolotov for his assistance in the data collection phase; SS&M anonymous reviewers for their constructive and insightful comments; and all the interviewees who participated in this research and shared their thoughts and experiences with us. References Abuhasira, R., Shbiro, L., Landschaft, Y., 2018. Medical use of cannabis and cannabinoids containing products - regulations in Europe and North America. Eur. J. Intern. Med. 49, 2–6. Barad, K., 1998. Getting real: technoscientific practices and the materialization of reality. Differences. J. Feminist Cult. Stud. 10, 87–128. Bostwick, J.M., 2012. Blurred boundaries: the therapeutics and politics of medical marijuana. Mayo Clin. Proc. 87, 172–186. Brosnan, C., Cribb, A., Wainwright, S.P., Williams, C., 2013. Neuroscientists' everyday experiences of ethics: the interplay of regulatory, professional, personal and tangible ethical spheres. Sociol. Health Illness 35, 1133–1148. Cerdá, M., Wall, M., Keyes, K.M., Galea, S., Hasin, D., 2012. Medical marijuana laws in 50 states: investigating the relationship between state legalization of medical marijuana and marijuana use, abuse and dependence. Drug Alcohol Depend. 120, 22–27. Charmaz, K., 2006. Constructing Grounded Theory: a Practical Guide through Qualitative Analysis. Sage Publications, London; Thousand Oaks, California. Charmaz, K., 2009. Shifting the grounds: constructivist grounded theory methods. In: Morse, J.M., Stern, P.N., Corbin, J., Bowers, B., Charmaz, K., C.A, E. (Eds.), Developing Grounded Theory: the Second Generation. Left Coast Press, Walnut Creek, CA, pp. 127–147. Chu, Y.-W.L., 2014. The effects of medical marijuana laws on illegal marijuana use. J. Health Econ. 38, 43–61. Duke, K., 2016. Exchanging expertise and constructing boundaries: the development of a transnational knowledge network around heroin-assisted treatment. Int. J. Drug Pol. 31, 56–63. Ehrich, K., Williams, C., Scott, R., Sandall, J., Farsides, B., 2006. Social welfare, genetic welfare? Boundary-work in the IVF/PGD clinic. Soc. Sci. Med. 63, 1213–1224. Fischer, B., Kuganesan, S., Room, R., 2015. Medical Marijuana programs: implications for cannabis control policy - observations from Canada. Int. J. Drug Pol. 26, 15–19. Fraser, S., Valentine, K., 2008. Substance and Substitution: Methadone Subjects in Liberal Societies. Palgrave Macmillan, Basingstoke, England. Frith, L., Jacoby, A., Gabbay, M., 2011. Ethical boundary-work in the infertility clinic. Sociol. Health Illness 33, 570–585. Gieryn, T.F., 1983. Boundary-work and the demarcation of science from non-science strains and interests in professional ideologies of scientists. Am. Socio. Rev. 48, 781–795. Guba, E.G., Lincoln, Y.S., 1989. Fourth Generation Evaluation. Sage Publications, Newbury Park, California. Guest, G., Bunce, A., Johnson, L., 2006. How many interviews are enough? An experiment with data saturation and variability. Field Meth. 18, 59–82. Hall, W., 2015. U.S. Policy responses to calls for the medical use of cannabis. Yale J. Biol. Med. 88, 257–264. Hallowell, N., Cooke, S., Crawford, G., Lucassen, A., Parker, M., 2009. Distinguishing research from clinical care in cancer genetics: theoretical justifications and practical strategies. Soc. Sci. Med. 68, 2010–2017. Hobson-West, P., 2012. Ethical boundary-work in the animal research laboratory. Sociology 46, 649–663. Johnson, J., Hodgkin, D., Harris, S.K., 2017. The design of medical marijuana laws and adolescent use and heavy use of marijuana: analysis of 45 states from 1991 to 2011. Drug Alcohol Depend. 170, 1–8. Kerr, A., CunninghamBurley, S., Amos, A., 1997. The new genetics: professionals' discursive boundaries. Sociol. Rev. 45, 279–303. Kleinman, A., 1995. From Margin to Center: from Medical Anthropology to Cultural Studies of Science. University of California Press, Berkeley. Lamont, M., Molnar, V., 2002. The study of boundaries in the social sciences. Annu. Rev. Sociol. 28, 167–195. Lancaster, K., Seear, K., Ritter, A., 2017. Making medicine; producing pleasure: a critical examination of medicinal cannabis policy and law in Victoria, Australia. Int. J. Drug Pol. 49, 117–125. LaRossa, R., 2005. Grounded theory methods and qualitative family research. J. Marriage Fam. 67, 837–857. Lindberg, K., Walter, L., Raviola, E., 2017. Performing boundary work: the emergence of a new practice in a hybrid operating room. Soc. Sci. Med. 182, 81–88. Morse, J.M., 1995. The significance of saturation. Qual. Health Res. 5, 147–149.

8

Social Science & Medicine 217 (2018) 1–9

D. Zarhin et al. Morse, J.M., Coulehan, J., 2015. Maintaining confidentiality in qualitative publications. Qual. Health Res. 25, 151–152. Morse, J.M., Stern, P.N., Corbin, J., Bowers, B., Charmaz, K., E, C.A., 2009. Developing Grounded Theory: the Second Generation. Left Coast Press, Walnut Creek, CA. Pachucki, M.A., Pendergrass, S., Lamont, M.l., 2007. Boundary processes: recent theoretical developments and new contributions. Poetics 35, 331–351. Pacula, R.L., Powell, D., Heaton, P., Sevigny, E.L., 2015. Assessing the effects of medical marijuana laws on marijuana use: the devil is in the details. J. Pol. Anal. Manag. 34, 7–31. Philipsen, N., Butler, R.D., Simon-Waterman, C., Artis, J., 2014. Medical marijuana: a primer on ethics, evidence, and politics. JNP-J. Nurse Pract. 10, 29–36. Rolfe, G., 2006. Validity, trustworthiness and rigour: quality and the idea of qualitative research. J. Adv. Nurs. 53, 304–310. Shuval, J.T., Gross, R., Ashkenazi, Y., Schachter, L., 2012. Integrating CAM and biomedicine in primary care settings: physicians' perspectives on boundaries and boundary work. Qual. Health Res. 22, 1317–1329. Sobo, E.J., 2017. Parent use of cannabis for intractable pediatric epilepsy: everyday empiricism and the boundaries of scientific medicine. Soc. Sci. Med. 190, 190–198. Sznitman, S.R., 2017. Do recreational cannabis users, unlicensed and licensed medical

cannabis users form distinct groups? Int. J. Drug Pol. 42, 15–21. Taylor, S.L., 2016. Evidence-based policy? The re-medicalization of cannabis and the role of expert committees in the UK, 1972–1982. Int. J. Drug Pol. 37, 129–135. Thompson, M., 2013. The Mysterious History of 'Marijuana'. NPR. Tilly, C., 2004. Social boundary mechanisms. Philos. Soc. Sci. 34, 211–236. UN, 1961. Single Convention on Narcotic Drugs. Wainwright, S.P., Williams, C., Michael, M., Farsides, B., Cribb, A., 2006. Ethical boundary-work in the embryonic stem cell laboratory. Sociol. Health Illness 28, 732–748. Williams, A.R., Olfson, M., Kim, J.H., Martins, S.S., Kleber, H.D., 2016. Older, less regulated medical marijuana programs have much greater enrollment rates than newer 'medicalized' programs. Health Aff. 35, 480–488. Wilson, N., Pope, C., Roberts, L., Crouch, R., 2014. Governing healthcare: finding meaning in a clinical practice guideline for the management of non-specific low back pain. Soc. Sci. Med. 102, 138–145. Zarhin, D., Negev, M., Vulfsons, S., Sznitman, S., 2017. Medicalization of cannabis: what does it mean? Int. J. Drug Pol. 49, 54–57. Zolotov, Y., Vulfsons, S., Zarhin, D., Sznitman, S.R., 2018. Medical cannabis: an oxymoron? Physicians' perceptions of medical cannabis. Int. J. Drug Pol. 57, 4–10.

9