Abstract
Medical cannabis is a topic of increasing debate. To investigate this issue, we conducted a content analysis of Israeli news coverage of medical cannabis from 2007 to 2013. A deductive framing analysis examined three elite issue frames—medical, policy, and law enforcement. Additionally, inductive analysis revealed a a fourth, nonelite patients’ frame. Each frame was associated with a distinct pattern of textual elements, including portrayal of patients, references to cannabis, opinion about medical cannabis, and salience of scientific research. The most common and most stable frame was the policy frame. Implications for framing theory are discussed.
Introduction
Medical cannabis is an unconventional and highly contested treatment. On the one hand, the 1961 UN Convention on Narcotic Drugs classifies cannabis as a Schedule I drug, a classification that indicates that it has little accepted medical use and high potential for abuse (United Nations, 1961; see also Ballotta, Bergeron, & Hughes, 2008; Bostwick, 2012). On the other, there is a growing basis of clinical research to suggest that cannabis may offer a range of medical benefits (e.g., Bonn-Miller, Boden, Bucossi, & Babson, 2014; Earlywine, 2002; Hazekamp & Grotenhermen, 2010; Institute of Medicine, 1999).
The increasing evidence base for medical cannabis, along with public pressure, has been followed by the legalization of medical cannabis in 23 states in the United States, as well as in Israel, Canada, and the Netherlands. Some jurisdictions such as Washington and Colorado have gone even further and legalized all uses of cannabis (Belle-Isle, et al., 2014; National Conference of State Legislatures, 2014). As a result of these policy changes, and variance in regulation across different U.S. states and countries, medical cannabis has become an increasingly complicated issue and the subject of much debate.
In contrast to Canada and the United States, where medical cannabis policies have been put in place after public vote, policy related to medical cannabis in Israel has been driven more exclusively by legal and governmental processes. As early as 1995, a subcommittee formed by the Israeli Parliament Drug Committee recommended not only that the government continue to categorize cannabis as illegal but also that it allow and regulate access to medical cannabis for patients who are severely ill (Brinn, 2009). As a result, Israel has been running a state-supported program providing medical cannabis since the late 1990s. Initially serving fewer than 100 people, the medical cannabis program began to grow substantially in 2007 and onward. In 2014, there were approximately 18,000 licensed medical cannabis patients in Israel. This number is expected to grow to 40,000 by 2018 (Efrati, 2014).
In Israel, specialist physician recommendations for medical cannabis are referred to medical professionals in the Medical Cannabis Unit of the Ministry of Health, who authorize or reject recommendations for licenses (Waissengrin, Urban, Leshem, Garty, & Wolf, 2014). In addition, 11 oncologists have the authority to directly issue licenses for cancer patients. Treatment with medical cannabis is more restrictive in Israel compared to some U.S. states and Canada where licenses for medical cannabis can be easily obtained from a physician for virtually any illness (Fischer, Kuganesan, & Room, 2014; W. Hall & Weier, 2015). In Israel, medical cannabis can only be recommended for a predefined list of medical indications, and licenses are typically only granted following the exhaustion of other “conventional” therapeutic options (Ministry of Health, 2014). However, Israel is less restrictive than some states in the United States that only allow the possession of certain cannabis oil extracts rich in cannabidiol, a nonpsychoactive compound of cannabis (Americans for Safe Access, 2014).
At present, when medical cannabis policies are changing rapidly within and across countries, it is important to examine media coverage of this issue. Indeed, media coverage is one of the major factors that influence public policy through shaping the public’s opinions on an issue (McCombs & Reynolds, 2002). The current study investigates framing of medical cannabis in Israeli media coverage. Israel represents an interesting case study as it is a country in which medical cannabis policy is centralized and uniform. Furthermore, the Israeli government has established and continues to oversee a wide-reaching medical cannabis program serving thousands of patients over an extended period of time. In contrast, in the United States, there are a variety of different policies relating to medical cannabis across different states, which complicates an investigation of media framing of medical cannabis. Indeed, the media landscape in each state is likely to be influenced by media coverage in other states with different policies, which may obscure an investigation of the framing of medical cannabis within states. In contrast, the media landscape in Israel is geographically focused on one jurisdiction.
Media Framing: Elite and Nonelite Frames
Seen from the perspective of framing theory, an issue such as medical cannabis can be understood from a range of perspectives and be perceived as having implications for multiple values or considerations (Chong & Druckman, 2007). Framing can make a difference in the way people think about issues—the way in which an issue is described provides advantages to one possible perspective and disadvantages to another (Wolfsfeld, 2011). By emphasizing particular aspects of an issue in news coverage, the media aids audience understanding of an issue and attaches meaning to news content (Entman, 1993; Gitlin, 1980; Goffman, 1974; Reese, Gandy, & Grant, 2001). The impact of this process is to promote a perspective that is consistent with that construction rather than a contrasting perspective.
The process of framing is dynamic and involves mutually influential relationships between journalists, elite groups (Gans, 1979), and social movements (Gitlin, 1980). Although social movements have been identified as influential in framing, framing theory has traditionally emphasized the emphasis of elites on media coverage rather than nonelite groups in shaping media coverage. According to this theoretical perspective, journalists construct frames, in part, in response to external political factors, such as elite discourses and changes in policy and public opinion. Media frames, in turn, influence policy and public opinion. Thus, it is expected that there is a reciprocal relation between frames and policy. This perspective is consistent with Wolfsfeld’s Politics-Media-Politics principle, which proposes that the media responds to changes in the political environment and, in turn, influences further political changes in the political environment (Wolfsfeld, 2011). Thus, media framing of medical cannabis is likely to be influenced by, as well as impact, political changes and shifts in public opinion related to this issue.
Framing theory has traditionally emphasized the influence of elites on media coverage. Elite groups are defined here as a “distinct group within a society which enjoys privileged status and exercises decisive control over the organization of society” (Wolf, 2012, p. 120). Elite groups are those who can exercise a degree of control over the productive assets and institutions, which enables them to influence both the allocation of resources and the allocation of authority (Amsden, DiCaprio, & Robinson, 2009). In the context of medical cannabis, elite groups include policymakers, the medical establishment, and the legal system (the judiciary and law enforcement). Each of these elites represents powerful institutions with access to structural resources. Each of these groups has a stake in medical cannabis and approach this issue from different perspectives and with varying, and often conflicting, interests. Each of these groups is directly involved in the issue of medical cannabis in Israel, and would benefit from media coverage that reflects and promotes their perspective.
For Israel’s medical establishment, interest in medical cannabis focuses on its potential utility as a treatment for patients and evidence basis for medical cannabis, or lack thereof. This group comprises doctors, nurses, researchers, other medical practitioners, and medical staff. In light of the growing empirical research suggesting potential medical benefits of cannabis for some patients (see above), a substantial proportion of doctors have shown support for the continued expansion of the medical cannabis program and its use among some patients (Nussbaum, Boyer, & Konrad, 2011), including world-renowned experts in this field, such as Prof. Raphael Mechoulam and Prof. Ruth Galily (Wilson, 2013). However, this positive attitude toward medical cannabis is not uniform across the medical establishment. A number of studies from the United States indicate that physician beliefs and attitudes regarding medical cannabis vary widely across physician populations, perhaps as a result of the clinical population served and differences in training (Harrold, Field, & Gurwitz, 1999; Kondrad & Reid, 2013; Williams et al., 1999). While no study, to the best of our knowledge, has formally explored attitudes toward medical cannabis within the Israeli medical establishment, some Israeli physicians have argued that more evidence is needed to demonstrate its efficacy as a treatment (e.g., Bar-Sela, Avisar, Batash, & Schaffer, 2014). Others have raised concerns about pressure from patients or their relatives on medical practitioners to provide medical cannabis (Israel Medical Association, 2013). Some medical professionals have also cited fears of a “slippery slope of the evolving list of medical conditions” for which medical cannabis is indicated (Thompson & Koenen, 2011).
Law enforcement officials include police officers, lawyers, and members of the judicial system, among others, whose most pressing concern is to prevent the diversion of medical cannabis from patients to other individuals. In Israel, cannabis is considered an illegal drug, and the medical cannabis program is seen as a threat to the enforcement of sanctions for the use and distribution of cannabis (Knesset Labor, Welfare and Health Committee, 2013). In addition, if medical cannabis finds its way, through sale or otherwise, to the streets, this may lead to increased criminal activity, generating an additional burden on law enforcement officials (Harkov, 2012). Consequently, the perspective of those in law enforcement is likely to reflect and highlight these concerns.
The perspective of policymakers, which include among others, officials at the Ministry of Health and members of various parliamentary committees, reflects the need to maintain control of the medical cannabis industry. Policymakers must also take into account the interests of the medical establishment, as well as the welfare of patients, and the concerns of law enforcement officials. In addition, policymakers are cognizant of the sizeable economic potential of this industry (Short, 2014), which could be of great economic benefit to the Ministry of Health as well as other (aligned) interest groups.
Framing theory has conceptualized framing as a contest between powerful elites (Gamson & Wolfsfeld, 1993). In this sense, much of the research employing framing theory applies a top-down approach to analysis in which journalists are presumed to, often subconsciously, prefer the frame that has a more natural fit with media routines or practices and/or heavy sponsorship by elites (Scheufele & Nisbet, 2007; Tewksbury & Scheufele, 2009). This approach largely neglects to consider the extent to which frames sponsored by nonelite groups may also influence media coverage of an issue. It is, however, theoretically possible to consider framing through a bottom-up approach as well, but this has not yet been widely applied in framing theory and research.
Nonelite frames may be adopted by journalists for reasons unrelated to political, economic or cultural power, or access to other structural resources. In the case of medical cannabis, patients are an example of a nonelite group in Israel. Patients are involved and interested parties with regard to Israel’s medical cannabis program. Tens of thousands of patients in Israel have received licenses for treatment with medical cannabis, and many more are waiting for their license applications to be approved or have seen their applications being rejected (Lis & Efrati, 2015). In addition, patients can offer a personal and compelling perspective on this complex issue. However, in contrast to the three elite groups described above, patients in Israel represent an unorganized group of individuals without agreed-upon spokespersons or organizations to promote their interests in the media or to influence medical cannabis policy. This stands in contrast to the situation in the United States, where patient-led organizations such as Americans for Safe Access have been a central agent in promoting medical cannabis policies.
According to literature on framing theory, structural power is a critical element in an actor’s ability to dominate the media text (Entman, 1993). While patients in Israel may lack this type of power, they may have other forms of power, as yet unexplored in framing theory. For example, they may have emotional power—the ability to generate sympathy and to elicit a strong emotional response from audiences. Given the critical role that patients play in the debate surrounding this issue, this study contributes to our understanding of framing processes by examining the extent to which Israeli news coverage is influenced by this bottom-up nonelite frame (in addition to examining the prevalence of elite frames).
Media Framing and Medical Cannabis
A number of studies have shown that framing influences policy support for drug policies (Elliot & Chapman, 2000; Forsyth, 2012; Lawrence, Bammer, & Chapman, 2000; McArthur, 1999). Despite the increasing debate surrounding the topic, we found only one study that has analyzed the framing of medical cannabis specifically in news media. Kaiser (2011) examined the framing of medical cannabis in U.S. print media, focusing on the question of whether the issue was portrayed as a policy issue or a medical issue (Kaiser, 2011). This study focused only on news coverage from the 19 states in which medical cannabis had been legalized at that time. The author found that the majority of news items framed the issue as policy related. The Kaiser study suggests that the dominance of the policy frame for medical cannabis may sway the debate over this issue and its legalization to one in which policy concerns take precedence over its (potential) risks or benefits as a medical treatment, thus possibly delaying its “path to legitimization” (Kaiser, 2011, p. 45).
The current study builds and expands upon this previous work, first by examining media framing of medical cannabis in another geographical and policy context (Israeli) over an extended period of time. In addition, the current study extends the framing analysis of medical cannabis to examine the extent to which this issue is framed as a law enforcement issue. Certainly, the blurring of boundaries between illegal and medical uses of cannabis is likely to be an issue of interest to law enforcement professionals—an elite group who may influence media coverage on medical cannabis. The inclusion of a law enforcement frame is also motivated by research on the media construction of issues related to crime and its influence on policy. For example, the classic study by Hall and colleagues (S. Hall, Critcher, Jefferson, Clarke, & Robert, 1978) found that U.K. media coverage that presented “mugging” as a moral panic racially associated with African Americans and Black immigrants to Britain boosted the legitimacy of harsh law enforcement measures. Finally, the study extends research on medical cannabis specifically, and research on framing in general, by considering the extent to which framing of medical cannabis is influenced not only by elite groups but also by a nonelite group—patients.
Study Objectives and Analytic Approach
One objective of the current study is to identify and measure the occurrence of three issue-specific elite frames (De Vreese, 2001)—a medical frame, a policy frame, and a law enforcement frame. In contrast to generic news frames, which are broadly applicable to a variety of different news topics, issue-specific news frames focus on specific topics, such as medical cannabis (De Vreese, 2001, 2005). In addition, the study tests whether a traditional, top-down analysis of elite frames may limit our understanding of framing processes in the context of medical cannabis by examining and contrasting the prevalence of a fourth frame—nonelite patients.
Frames can be best understood as groups of various independent textual elements working together to create an overall frame. In addition to making up a frame, each independent textual element can also signal different meanings to the reader (Entman, 1993). This approach draws from the sociological foundation of framing that has been used by researchers including Entman (1993), Gamson and Modigliani (1989), Gitlin (1980), and Goffman (1974). This body of research investigates the way in which news is presented to audiences by journalists (Kaiser, 2011). This approach to framing research includes “words, imagines, phrases, and presentation styles” that are used by journalists to create news stories (Druckman, 2001, p. 227). Choice of words, imagines, phrases, and presentation styles affect the interpretation and definitions of issues, such as medical cannabis, and may even shape moral judgments and shape policy. To better understand how the news media frames the issue of medical cannabis, we investigate the association between textual elements within news coverage and the overall framing of medical cannabis. Another objective of this study is to examine changes in the relative proportion and stability of these four frames during the study period (January 2007 through June 2013).
To achieve these objectives, we separately coded and examined a news item’s overall frame as well as the association between an overall measure of framing and a range of distinct textual elements that are expected to be associated with the overall frame. We examined the associations between each of the overall frames for medical cannabis and (1) the characterization of cannabis (i.e., as a drug or as a medicine), (2) the way in which patients were portrayed in each news item (e.g., as needy or in distress, or untrustworthy), (3) support for medical cannabis and/or the medical cannabis program, and (4) references to scientific research. By examining the association between overall frames and distinct textual elements, we aim to identify subtle patterns in terms of how medical cannabis is reported on in the press and how the overall frame is achieved. We also aim to attenuate the (potential) influence of the researcher’s subjective perceptions (Kohring & Matthes, 2002).
Hypotheses
Framing of Cannabis
We expected that there would be an association between the (overall) frame for medical cannabis and references to cannabis (i.e., as a medicine, a drug, or other reference). The following hypotheses are proposed:
Portrayal of Patients
We also examine associations between the portrayal of patients and framing of medical cannabis. The following hypotheses are proposed:
Opinion About Medical Cannabis
We examined associations between the (explicitly or implicitly expressed) opinion about medical cannabis and/or the medical cannabis program in the news item and overall frame. The following hypotheses are proposed:
Research on Medical Cannabis
We examined the extent to which news items within each frame included references to scientific research on medical cannabis. The following hypothesis is proposed:
In the second stage of the analysis and with regard to relative proportions, we propose the following hypothesis, based on prior research (Kaiser, 2011):
We also asked the following research questions:
Method
Sample
To quantify and analyze news coverage of medical cannabis, we sampled all news stories that focused on medical cannabis that were published in the three highest circulation national newspapers in Israel (Yediot, Maariv, and Haaretz) from January 2007 through June 2013. We began our study period in 2007 to capture the period in which medical cannabis licenses first began to increase. Our sampling frame included both print and online articles accessible through the newspapers’ websites (ynet.co.il, nrg.co.il, and haaretz.co.il). For news items that were published in a newspaper’s print and online site, we included the print item only in order to avoid duplication in our sample. Based on the most recent available data, print circulation ranged from 300,000 to 600,000 (Yediot), 160,000 to 270,000 (Maariv), and 65,000 to 75,000 (Haaretz). According to Alexa (n.d.), Yediot’s website (ynet.co.il) receives about 310,841 unique visitors and 1,059,967 page views per day; Maariv’s website (nrg.co.il) receives about 101,365 unique visitors and 304,095 page views per day; and Haaretz’s website (haaretz.co.il) has approximately 53,082 unique visitors and 159,245 page views per day.
News Coverage Selection
We used the online archives of each of the three newspapers to collect news media stories, as well as the periodicals library in Bet Ariella, an archive of Israeli press coverage located in Tel-Aviv. To generate a sample frame of news stories, we used the following search terms: “cannabis” or “medical cannabis” and “marijuana” or “medical marijuana” (N = 1,233). We searched entire articles, including titles. From this sampling frame, we selected the articles that related specifically to medical cannabis and excluded those that referred only to marijuana or cannabis generally but not to medical usages specifically. News stories that were classified as letters to the editor, duplicate items, or book reviews were also excluded. The search led to a total sample of 229 news stories and editorials related to medical cannabis. Twenty-one of these articles did not relate to the framing of cannabis as a medical, policy, legal, or patient issue (i.e., focused on diverse issues, such as medical cannabis as a religious practice or as a commercial industry) and were excluded from the current analysis, leading to a final analytic sample of 208 articles.
Measures
Overall Frame
The creation of the overall frame variable was conducted in two stages. During the first stage, we applied a deductive approach, which involved defining specific frames prior to coding (Semetko & Valkenburg, 2000). During this stage, coders assessed whether medical cannabis was discussed within the context of medicine, regulation, or a law enforcement issue in news coverage (the elite frames that were identified prior to analysis). Following this stage, we employed an inductive, bottom-up approach. This approach allows alternative frames to emerge from the data during the course of analysis (De Vreese, 2005). We found that patients, a nonelite source, were a highly prevalent source in media coverage (28% of articles cited patient directly, and 38% cited patients indirectly). Thus, we reexamined all news items using an inductive analytic approach to identify items framing medical cannabis from the perspective of patients (n = 45), which were coded as a fourth (nonelite) patients’ frame.
The measure of overall frame took into account the primary focus of the news item, the most prominent sources cited, the context in which medical cannabis was discussed, and other references to medical cannabis throughout the article. For example, articles that situated the issue of medical cannabis in the context of medical treatment and/or included sources from the medical establishment were coded as having an elite medical frame. In contrast, articles describing medical cannabis in the context of health but focusing on patients (patients were the most prominent sources) were categorized as having a nonelite patient frame.
News items that emphasized the diversion of medical cannabis to the black market and other aspects of criminal activities, and cited sources from the judiciary or law enforcement officers, were coded as having a law enforcement frame. Articles that focused on the regulatory or political context of medical cannabis, and cited sources from the political establishment, were coded as having a policy frame. These articles focused on governmental regulations, policy decisions by the Ministry of Health, or administrative issues related to licensing of medical cannabis. The coding was based on the understanding that framing of medical cannabis may include elements from each of these constructs, but that the overall framing of a news item will tend to be dominated by one of these perspectives.
Correlates With Overall Frame
In addition to describing the distribution of each of the overall frames, this study tests hypotheses relating to other elements within the body of the news item and their associations with the overall frame (see Hypotheses 1-4). These hypotheses serve to validate our measure of overall frame of medical cannabis by testing proposed associations between the overall measure and other textual elements. They also enable an investigation of how the overall frames were conceptualized. For this purpose we measured four textual elements.
Coders judged references to cannabis, a measure which included the following categories: (1) cannabis as an illicit drug (e.g., used terms such as “grass” or “joint”), (2) cannabis as a medicine (e.g., used terms such as “medical cannabis,” “medicine,” or “treatment”), or (3) other (e.g., cannabis is described as a religious practice or an economic product). In addition, coders assessed the portrayal of patients. This variable included the following categories: (1) neutral description of patients, (2) negative portrayal—patients as untrustworthy, (3) patients as victims of disease, or (4) no mention of patients.
The third textual element evaluated by coders was opinion about medical cannabis, which includes the following categories: (1) negative (i.e., the news item is critical of medical cannabis and/or the medical cannabis program), (2) mixed (i.e., the item describes both positive and negative aspects of medical cannabis and/or the medical cannabis program), (3) neutral (i.e., the item presents facts about medical cannabis and/or the medical cannabis program without criticism or endorsement), and (4) positive (i.e., the item is supportive of medical cannabis and/or the medical cannabis program). Finally, coders assessed the extent to which the item refers to research on medical cannabis, which includes the following categories: (1) low salience of research (i.e., one or no references to scientific research) and (2) high salience (i.e., two or more references to research on medical cannabis).
Time Trends
To examine trends over time, a variable was created that indicated the 6-month time period in which articles were published. As only 11 articles were published between January 2007 and June 2008, the four 6-month intervals that comprise this time period were collapsed. To explore potential curvilinear relationships, squared and cubed versions of the time trend variable were created.
Coding Procedure and Reliability
A coding instrument was developed that sought to identify an exhaustive list of textual elements that were expected to be related to the overall frame of medical cannabis. Two graduate student coders were extensively trained by two researchers during a 2-month period to become familiar with the codebook measures and to practice coding news stories. In order to establish proper training and intercoder reliability before coding the entire sample, the coders met on several occasions during the coding process, checking reliability for measures on a subset of articles. Adjustments were made to the coding protocol during this process to reduce discrepancies. Coders then used the revised codebook to independently code a random sample of 24 news stories (12% of the sample), as recommended by Wimmer and Dominick (2006), and consistent with methods used in similar studies (e.g., McKeever, 2012). Intercoder reliability was calculated for each item in the coding protocol using Krippendorff’s Alpha (Hayes & Krippendorff, 2007). Reliability for each item was measured using K statistics, which all met conventional standards for adequate reliability (≥.80; Hayes & Krippendorff, 2007) and ranged from 0.83 to 1.00. The final coefficients were as follows: overall frame (α = 1.00), reference to cannabis (α = 1.00), portrayal of patients (α = .83), opinion about medical cannabis (α = .99), and research cited (α = .99). These reliability scores are also consistent with other published content analyses (e.g., Matthes, 2009; McKeever, 2012).
Data Analysis
The unit of analysis for this content analysis was individual newspaper articles. In order to identify and validate the overall framing of cannabis as a medical issue, a policy issue, a law enforcement issue, or a patient issue, and to examine the proportional differences, we relied on chi-square tests of independence. To examine the distributions of frames over time, we graphed the trends in framing. Furthermore, to examine potential linear and curvilinear time trends statistically, stepwise logistic regression was used, where each of the frames were separately entered as the dependent variable, and a variable indicating the linear time trend was entered as the independent variable at the first step. In the second step, the quadratic time trend independent variable was added, and in the third step, the cubed time trend variable was added. The time trend variables were standardized in order to avoid multicollinearity (Marquardt, 1980).
Results
The sample consisted of 208 articles, of which 190 were news articles, 14 were editorials, and 9 were “other” category articles. Table 1 shows the percentage of news items for each frame according to the parameters studied.
Comparison of Elite and Nonelite Frames.
Note. Each subscript letter denotes a subset of overall frame categories whose column proportions do not differ significantly from each other at the .05 level.
Results show that references to cannabis in news items differed by overall frame (Hypothesis 1), χ2(6) = 48.33, p < .001. Medical cannabis was referred to as a medicine or treatment in the vast majority (91.4%) of news items in which medical cannabis was framed as a medical issue, confirming Hypothesis 1a. In contrast, in news items framing cannabis as a law enforcement issue, medical cannabis was described as a drug in the majority of cases (84.6%), compared with other references, confirming Hypothesis 1b. Cannabis was described as a medicine in approximately two thirds (67.3%) of news items with a policy frame, and a patient frame (64.4%), a distribution that is less extreme than was observed in the other frames.
Results also show that the portrayal of patients differed by overall frame (Hypothesis 2), χ2(9) = 70.39, p < .001. Patients were described as needy or in distress in the majority of news items with a medical frame (52.8%), supporting Hypothesis 2a. In contrast, the majority of news items (61.5%) in which medical cannabis is framed as a law enforcement issue described patients in a negative manner, supporting Hypothesis 2b. Patients were referred to in neutral terms in almost half of news items with a policy frame (48.5%), which is consistent with expectations (Hypothesis 2c), although the difference in proportions across frames did not reach significance (p > .05). Results also show that patients were portrayed as distressed and in need of treatment in approximately two thirds of items using the patient frame (64.4%), a pattern similar to that observed for the elite medical frame, supporting Hypothesis 2d.
Results also found differences regarding the evaluation (pros and cons) of medical cannabis and/or the medical cannabis program across frames (Hypothesis 3), χ2(9) = 43.94, p < .001. In news items with a medical frame, the majority of items expressed a positive opinion toward medical cannabis (69.4%), supporting Hypothesis 3a. The majority of news items framing medical cannabis as a law enforcement issue expressed a neutral opinion (61.5), while only two items (7.7%) emphasized the negative aspects of medical cannabis. Thus, Hypothesis 3b is not supported. The majority of news items with a policy frame expressed a neutral (56.4%) opinion about medical cannabis (Hypothesis 3c); however, the difference in proportions across frames did not reach significance (p > .05). Results also show that the majority of news items with a patient frame emphasized the positive aspects of medical cannabis and/or the medical cannabis program (64.4%), supporting Hypothesis 3d.
Results also show an association between references to scientific research and the overall frame. News items with a medical frame were more likely to include references to medical research (44.4% of all items with this frame), compared to other frames, supporting Hypothesis 4.
Focusing more specifically on the top-down medical elite frame and the bottom-up patient frame, the results suggests that there are points of convergence and divergence between news items employing these frames (see Table 1). For both the elite medical frame and the nonelite patient frame, approximately two thirds of all news items expressed a positive opinion toward medical cannabis (69.4% and 64.4%, respectively). In comparison, for items using the policyframe or the law enforcement frame, the majority of news items expressed a neutral position with regard to medical cannabis (56.4% and 61.5%, respectively). Another point of similarity between the elite medical and nonelite patients frame relates to the portrayal of patients. In the majority of news items with an elite medical frame, when patients are discussed, it is in relation to their distress and need for treatment (52.8%). Similarly, in items employing a bottom-up patients’ frame, approximately two thirds of all items portrayed patients in relation to their suffering (64.4%). Thus, both the elite medical frame and the bottom-up patients’ frame appear to promote a positive perspective on medical cannabis and acknowledge the need to find a treatment for patients suffering from chronic or terminal diseases.
However, these frames diverged with regard to the emphasis placed on scientific research on medical cannabis (see Table 1). The elite medical frame referred to research more frequently (research was highly salient in 44.4% of items) compared with news items using the patients’ frame (13.3%). On a related note, a qualitative analysis of news items using these frames reveals a recurring pattern relating to the rhetoric strategies employed by the sources cited. In most news items with a patient-focused frame, patients interviewed by journalists described their poor health (prior to beginning treatment with medical cannabis) and the dramatic improvement in their health as a result of using medical cannabis. Patient testimonials are rich in emotional, nonscientific language and appeal to the reader’s sympathy and emotion. For example, one patient describes how “medical cannabis changed my life. Finally I was able to sleep. . . . It was the only light in my life” (Koren, July 13, 2011). In another news item, a patient reports, “Ever since I began using medical cannabis my life has changed. There are no side effects and I can sleep” (Levi, May 22, 2011). According to another elderly female patient, “The cannabis helps me connect with my feelings” (Even, January 4, 2009). In some cases, even the title of the patient-focused news items reflects an appeal to readers’ emotions—“Ronen’s plea—Don’t take away my medical cannabis!” (Alzon, May 21, 2013), or “This plant is a present from god” (Har Noi, June 12, 2007). In contrast, when medical sources are cited (in both the elite medical and the patient frames), most sources are quoted using language that is precise (including medical terminology and statistics) and refer to medical research to support their claims.
Relative Proportions of Frames
The hypothesis that, overall, there would be a greater number of articles that framed medical cannabis as a policy issue (Hypothesis 5), was confirmed; 48.5% (n = 101) framed the issue as a policy issue. Proportional tests showed that the policy frame was significantly more common than the other frames (p < .001). The second most common frame was the nonelite patient frame (22%, n = 45), followed by the medical frame (17%, n = 36) and the law enforcement frame (12.5%, n = 26).
Time Trends in Framing
Figure 1 shows the trends in framing over time and confirms that the policy frame was more prevalent across the entire study period. Furthermore, it shows that while the policy and patient frames were relatively stable over time, the prevalence of the other frames fluctuated substantially over the study period. In order to explore these time trends more carefully, we ran logistic regression models for each of the three frames where linear, quadratic, and cubed versions of the time trend variable were entered stepwise as predictors. Results confirm the visual representation of the data, in that none of the time trend variables were significantly related to the policy or the patient frames (p > .05), whereas the cubed time trend variable was significant for the law enforcement (p < .005) and the medical frame (p < .006).
Mean proportions of overall frame for medical cannabis in news coverage (January 2007 to June 2013).
Discussion
This study describes a first attempt to examine the framing of medical cannabis in Israeli news coverage. The study is based on established approaches in framing research (e.g., Martins et al., 2013) and examines the overall frame as well as a range of textual features. Research on the framing of medical cannabis is important as the way in which an issue is framed may shape public opinion (Price, Tewksbury, & Powers, 1997), including opinion about policy relating to medical cannabis, which, in turn is likely to affect medical cannabis policy.
Results of the current study show that the competing frames for medical cannabis are shaped by, and comprised of, a number of contextual elements within the news items. These elements include the references to cannabis (i.e., a drug, a medicine, or an economic commodity), the portrayal of patients, opinion regarding medical cannabis, and references to scientific research. These elements work together to promote a particular view of medical cannabis, which is likely to influence perceptions of this issue and of patients. For example, by referring to medical cannabis as a drug or a medicine, journalists may affect how audiences receive, process, comprehend, and form opinions about various issues (e.g., Gamson, 1992; Kinder, 2007), including medical cannabis. Similarly, by portraying patients as suffering and in need of treatment, the media encourages sympathy for patient demand for medical cannabis. In contrast, the portrayal of patients as untrustworthy (i.e., the most frequent reference to patients in the law enforcement frame) promotes the notion that the provision of medical cannabis should be carefully restricted and requires the oversight of law enforcement to prevent potential abuse.
The results of this study show that patients were featured prominently as sources in media coverage of medical cannabis in Israel and are an important part of the medical cannabis story in Israel. In fact, patients were the most frequently cited source in our sample. The findings of this study indicate that Israeli news media coverage of medical cannabis was less dominated by elite frames than may be expected based on traditional top-down framing theory. The nonelite patient frame accounted for more than one fifth of all news items examined and represented the second most prominent frame. The findings that elites were not the only influences on framing suggests that framing researchers should reconsider the conceptualization of power and the focus on elites. As Carragee and Roefs (2004) suggest, it is necessary to move away from thinking about framing in terms of “mechanistic definitions of hegemony.” For some issues, and in some contexts, nonelite frames may also influence framing construction. The power to influence media discourse may stem from factors that are not, as yet, accounted for in framing theory. For example, patients are a nonelite frame (in Israel) and lack economic and cultural power, but the bottom-up patient frame has the advantage of emotional power—an important form of persuasive currency (see Berger, 2013). As such, patients are likely to engender sympathy and identification among readers. Patients are also likely to be seen as credible sources by readers, as (unlike other elites) they are unlikely to be perceived by readers as having an economic or political stake in the debate.
This study illustrates the way in which framing medical cannabis as a medical issue, a policy issue, a law enforcement issue, or a patients’ issue promotes (likely unintentionally on the part of journalists) the interests and perspectives of particular stakeholders involved in this issue. As the most dominant frame in Israeli news coverage, the policy frame promotes a relatively neutral view of medical cannabis and of patients. This is somewhat in contrast to the second and third most prevalent frames (the patients’ frame and the elite medical frame) for which medical cannabis is first and foremost presented as a treatment option for patients and in which patients are frequently described as victims of disease who are suffering and in need of treatment. Last, the criminal frame, which was the least prominent frame in news coverage, promotes a more negative view of medical cannabis as it describes cannabis as an illicit drug and patients as untrustworthy or criminal. These findings may be interpreted as a positive sign for stakeholders who are interested in the continued expansion of the Israeli medical cannabis program. The relative scarcity of media items in which medical cannabis was framed in the context of crime or law enforcement may also suggest that there is a distinction made, among media professionals, between cannabis for recreational purposes and medical cannabis.
The finding that the policy frame was the most prominent is consistent with research by Kaiser (2011), which found that the policy frame was the dominant issue frame in U.S. newspaper coverage of medical cannabis. One explanation for this finding may relate to the differences in access to journalists across different sources. Policymakers have a vested interest in receiving media coverage that is consistent with their policy goals and an elite status that would enable relatively great access to journalists. In contrast, medical professionals and law enforcement agents are not dependent, in the same way, on public support and opinion. Thus, there may be a structural advantage for policymakers as far as promoting a policy frame for medical cannabis compared with politicians and regulators or law enforcement agents. However, as noted above, results indicate that Israeli media coverage was more “democratic” than expected, and (in the context of medical cannabis) was responsive to nonelite frames.
When examining the stability of frames over time, the difference between the stable policy frame and patient frame and the volatile medical and law enforcement frame is striking (see Figure 1). One explanation for this finding may be that journalists covering medical cannabis may be influenced, in their choice of frame, by breaking events such as a medical research breakthrough regarding medical cannabis, or by a crime story relating to medical cannabis (e.g., the arrest of medical cannabis users who were driving under the influence). The effect of these events is likely to influence the choice of frame for medical cannabis in immediate news coverage and also be short-lived. This is consistent with the sudden changes in the proportions of news items using these frames. In contrast, the policy frame focuses on long-term issues and regulatory debates, which may not only be less controversial and attention grabbing in the short-term but may also be less likely to be influenced by onetime events. Similarly, the patient frame is likely to generate reader interest and sympathy in a manner that is independent upon particular onetime events.
Limitations
A frequently raised concern regarding the analysis of news media is that the advent of online social media and the shift from broadcasting to narrowcasting have altered what we know about media effects, particularly the effects of the “mass” media, such as framing through news coverage (Bennett & Iyengar, 2008; Chaffee & Metzger, 2001). Fragmentation of media and increasingly selective exposure of audience to niche content provides audiences with the opportunity to choose news coverage that is more likely to frame issues in a way that resonates with, and reinforces, their existing opinion, thus mitigating framing effects.
However, as Wolfsfeld (2011) notes, many of the issues that are discussed in new media are first raised in the mass media, including in newspaper coverage. Newspapers (and their online versions) remain a useful proxy for news reporting, as they are likely to set the agenda for other news formats (Clegg Smith et al., 2002; Wakefield, Flay, Nichter, & Giovino, 2003). Among some audiences, and for some topics, selective exposure to particular frames may attenuate the effects of framing; however, the basic concept remains important within the study of media effects (Wolfsfeld, 2011). For example, the news media is an important source of health information (Kenterelidou, 2012), especially for topics that are not easily comprehended by the majority of individuals, such as medical cannabis. Furthermore, newspapers (online and/or in print) are read by 95% of Israeli adults, which in an international perspective is extremely high (European Neighborhood Journalism Network, 2014). As such, newspapers are clearly an important channel for shaping the public opinion in relation to medical cannabis.
This study has a number of strengths, including the use of rigorous methods of analysis to examine an important, but unexplored research area. However, despite these strengths, this study has a number of limitations that should be noted. One limitation is the relatively small sample size. It should be noted that Israeli (Hebrew language) newspaper coverage tends to focus rather heavily on issues relating to the security and political arena, including the Palestinian-Israeli conflict. This overarching focus on political issues is at the expense of the quantity of coverage of other topics, including health, which receives relatively little coverage, compared with media coverage of health in other countries. Consequently, the total number of articles focusing on this health topic (i.e., medical cannabis) is fairly small (N = 229). However, news items focusing specifically on medical cannabis comprised approximately 20% of all coverage about cannabis (N = 1,332), which suggests that the topic of medical cannabis is an important policy issue. In addition, medical cannabis and related policy affects tens of thousands of patients, such that the volume of coverage alone should not be seen as an indicator of the significance of this topic in Israel.
This study focuses on Israeli news media coverage from three popular news sources. Although there appear to be some similarities as far as the prevalence of the policy frame in coverage of medical cannabis, we do not know whether the findings reported here are generalizable to news media coverage from other news sources and in other countries. Another limitation of the study relates to the choice to focus on a limited number of textual elements in order to define the frames examined here. The elements were chosen (prior to analysis) on the basis of their centrality to the policy debate surrounding medical cannabis (i.e., Can patients be trusted? Is medical cannabis a drug or treatment?). However, we acknowledge that other elements (e.g., use of metaphors, selection of images) may also have influenced media framing and should be included in future research. Furthermore, the sample included only Hebrew language news articles, and it is unclear whether the current results are generalizable to Arabic and other language news articles published in Israel.
In addition, although this study sheds light on trends in framing of medical cannabis in Israeli media coverage, it does not investigate the readers’ interpretation of the competing frames and the effect of these frames on public opinion and policy discussion related to medical cannabis. Nevertheless, mass communication theory suggests that the news coverage is likely to influence public opinion related to medical cannabis (Lancaster, Hughes, Spicer, Matthew-Simmons, & Dillon, 2011). As the competing frames represent competing interests in and perspectives on medical cannabis, it is important for researchers to track which frames are dominant and which of these perspectives are reflected in news media coverage.
Finally, another limitation of this study is that it focuses only on media framing of medical cannabis and as such the analysis focuses on one aspect of the broader public debate surrounding cannabis. The public’s understanding of medical cannabis, and media coverage of this issue, are likely to be informed by this wider discourse (e.g., debate surrounding the legalization of cannabis for recreational purposes and other related issues), a factor that is not examined here. Future research on media framing should expand the focus of analysis to include media discourse beyond a specific issue (e.g., medical cannabis). This approach will offer a more nuanced perspective of the relationship between media discourse and the way in which we understand specific issues.
Conclusions
As Kaiser (2011) notes, certain medical issues are, by their nature, also political issues. The issue of medical cannabis is a particularly interesting and important case study for framing analysis as it can be approached from a range of different, and competing, perspectives. It is also a complex issue that the public may be unfamiliar with, and thus will rely more heavily on media coverage for information and interpretation. As a result, the media can be expected to play an important role in shaping public perceptions of and attitudes toward medical cannabis. Indeed, media framing of medical cannabis in news coverage may influence the debate on this topic by placing greater emphasis on considerations and perspectives that are consistent with the perspective of one group (e.g., politicians or patients), while relegating others (e.g., the medical establishment or law enforcement) to the background.
The findings of this study have implications for framing theory, in that it shows that framing may not (always) be a process whereby elites compete with each other to influence media coverage. Instead, other nonelite frames may also influence the construction of issues. In the case of controversial scientific issues, such as medical cannabis, it may be particularly important to consider the extent to which the perspectives and voices of the nonelites most affected by the issue—the patients, the primary beneficiaries of medical cannabis—influence framing processes.
Footnotes
Acknowledgements
The authors are grateful to Keren Sereno, Elinor Tram, and Yuval Zolotov for their assistance with this study and to the anonymous reviewers for their insightful comments on the article.
Authors’ Note
The funding sources had no involvement in the collection, analysis, or interpretation of data; in the writing of the article; or in the decision to submit it for publication.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the University of Haifa’s Department of Research and by an Israeli Pfizer Health Policy grant.