by Michael D. Anestis, M.S.
Much of my early career has been devoted to an understanding of how negative emotions influence problematic behavioral outcomes. Because of this, I have spent a fairly substantial amount of time on PBB discussing topics such as distress tolerance, negative urgency, and drinking motives. That being said, every once in a while, I come across a study that reminds me that the relationship between emotions and maladaptive behaviors can not always be reduced to the drive to make uncomfortable emotions fade away. Today, I would like to discuss one such study, recently published in Depression and Anxiety by Julia Buckner of Louisiana State University and Brad Schmidt of Florida State University. Dr. Schmidt is is a world renowned anxiety researcher and clinician and Dr. Buckner is quickly establishing herself as an expert in the understanding of social anxiety disorder (SAD) in general and marijuana use in SAD in particular. In this particular study, the authors examined why individuals with SAD exhibit such an elevated marijuana use rate and the answer, quite frankly, was not what I expected.
You might remember from earlier PBB articles on the topic that SAD is characterized by the fear of negative evaluation by others. In the generalized form, this can extend to any social or performance situation in which embarrassing behavior or failures are believed to have undesirable consequences. In the non-generalized form, the anxiety is specific to a particular situation (e.g., public speaking, taking an exam). Regardless of the subtype, the data from the National Comorbidity Survey indicate that individuals with SAD are seven times more likely to use marijuana than the general population (Agosti, Nunes, & Levin, 2002). Similarly, a longitudinal study of adolescents diagnosed with SAD found that such individuals were five times more likely to develop marijuana use problems during young adulthood than were adolescents not diagnosed with SAD (Buckner et al., 2008).
My gut impulse when reading that information is to assume that individuals with SAD use marijuana to reduce their feelings of anxiety and to make social situations easier to manage. Furthermore, I would anticipate that individuals with SAD would report that they fully anticipate marijuana to accomplish these feats. In the only prior work examining this point, however, Buckner and Schmidt (2008) found an entirely different answer. In their study, conducted on a nonclinical undergraduate population, they found that individuals with higher levels of social anxiety reported greater levels of cognitive and behavioral impairment as a result of marijuana use and that those same individuals expected that marijuana would fail to reduce tension and facilitate social interactions. So, despite that fact that higher levels of social anxiety are positively correlated with marijuana use frequency, there is no evidence that socially anxious individuals engage in this behavior with the belief that it will make them more social and less anxious.
Before explaining why the authors believe they found these surprising results, let's talk about their recently published follow-up study. This time, Buckner and Schmidt (2009) examined the same variables, but with a clinical population. They again recruited from an undergraduate population, with 107 individuals (43% female) taking part in the study. This time, however, in addition to having the participants fill out self-report questionnaires, the authors conducted structured diagnostic interviews (Anxiety Disorders Interview Schedule - IV; DiNardo, Brown, & Barlow, 1994) in order to determine what percentage of their sample actually met criteria for SAD and other mental illnesses (depression and other anxiety disorders, in this case). 26.2% of the sample met DSM-IV-TR criteria for SAD, 96.2% reported having used marijuana in the past month, 78.3% reported using marijuana at least weekly, and 32.1% reported using marijuana on a daily basis. Participants reported using marijuana, on average, 1-4 times per week throughout their lifetime and 3-4 times per week during the past month. As such, a substantial portion of the sample in this study was clinically anxious and using marijuana regularly.
In their first analysis, Buckner and Schmidt (2009) found that whether or not an individual met criteria for SAD predicted marijuana use problems (e.g., problems with significant other, missing work or classes, financial difficulties) above and beyond the effects of biological sex, marijuana use frequency, depression, generalized anxiety disorder, obsessive-compulsive disorder, panic disorder, specific phobia, and number of SAD symptoms. In other words, even when you consider the impact of sex, frequency of marijuana use, and all of those mental illnesses, whether or not an individual met diagnostic criteria for SAD still significantly predicted the degree to which that individual experienced problems as a result of their marijuana use. This, quite clearly, is evidence supportive of the utility of diagnoses.
In their second analysis, Buckner and Schmidt (2009) controlled for all of the same variables they used before, but this time examined the relationship between SAD status and expectancies of cognitive and behavioral problems as well as the tension reduction potential of marijuana. Just as they did before, the authors found that being diagnosed with SAD increased the likelihood that an individual would expect marijuana to result in global negative consequences and cognitive and behavioral problems.
The authors' third analysis was the most interesting. Here, the authors examined whether the belief that marijuana would result in cognitive and behavioral problems as well as global negative consequences would mediate the relationship between SAD and marijuana use. In other words, Buckner and Schmidt (2009) wanted to see if the relationship between SAD and marijuana use was explained by such expectations. The results, in fact, confirmed that this was the case. The evidence thus indicates that SAD is related to marijuana use because individuals with SAD expect negative outcomes.
Now, if you're like me, you likely read that and thought "this makes no sense." After all, why would people use a substance if they expect that it will result in negative consequences? The authors provided a few different theories. First of all, the scale that measures cognitive difficulties includes items such as "marijuana slows my thinking and actions" and "marijuana alters my personality." As such, these results might, to some extent, reflect a desire on the part of individuals with SAD to slow down anxious thoughts. This, however, does not explain the finding regarding global negative consequences. Buckner and Schmidt (2009) addressed this through the lens of the self-handicapping theory of substance use (Jones & Berglas, 1978). As applied to these findings, this theory posits that, because individuals with SAD believe marijuana will have a negative impact on their behavior, they anticipate that others will hold that belief as well. As a result, using marijuana might provide them with an opportunity to blame any anticipated negative outcomes on marijuana use rather than some flaw in their character. As a result, marijuana use can become an appealing safety aid for individuals with SAD even though they often expect it to result in problems, as they believe marijuana use makes others less likely to see those problems as a reflection of their character. Now, obviously this is not the most sound reasoning on the part of the individuals choosing this behavior for this reason, but how many of us use sound reason when we are highly anxious? If you think about it from the perspective of the individual who is paralyzed by the fear of being negatively evaluated by others, the function of this behavior becomes clear, even as its consequences stare us in the face.
An important consideration to keep in mind is that some prior work has, in fact, linked SAD to coping motives for marijuana use (smoking in order to reduce negative emotions; Buckner et al., 2007). That being said, the scales used to measure that motive make no mention of social situations whereas the ones used in the Buckner and Schmidt (2009) study do. As such, the authors believe this might reflect a tendency on the part of individuals with SAD to smoke when alone (e.g., prior to or after a social event) in order to regulate emotions as well as a belief on the part of individuals with SAD that marijuana is unlikely to help with anxiety prompted by actual social interactions.
So, what can we take home from this study? First of all, this is a great example of why research is so important. Left to only use our intuition, we are prone to making assumptions about why two variables are related. Because I tend to research populations less relevant to SAD (e.g., borderline personality disorder), I am used to investigating people who tend to engage in problematic behaviors in response to negative affect. As such, I looked at the relationship between SAD and marijuana use and immediately assumed that they were linked because of the drive to reduce anxiety and more easily navigate social and/or performance situations. Clearly, however, this is not the case. Additionally, this study helps shine a light on the fact that our motivations for using a particular behavior may, at times, hinge upon the context. In other words, when alone, individuals with SAD may smoke for different reasons than they do when they are faced with a situation in which they might be evaluated by others. As such, when working with individuals with SAD in treatment, it might be important for clinicians not only to help clients develop better emotion regulation skills, but also to help them adjust their beliefs about the utility of self-handicapping.
If you would like to learn more about social anxiety disorder, we recommend the following resources, all of which are available through our online store:
- Anxiety and Its Disorders, Second Edition: The Nature and Treatment of Anxiety and Panic by David Barlow
- Cognitive Behavioral Therapy for Social Anxiety Disorder by Stefan Hofmann and Michael Otto
- Managing Social Anxiety: A Cognitive-Behavioral Therapy Approach Therapist Guide by Debra Hope, Richard Heimberg, and Cynthia Turk
- Cognitive-Behavioral Group Therapy for Social Phobia: Basic Mechanisms and Clinical Strategies by Richard Heimberg and Robert Becker
- Social Phobia: Diagnosis, Assessment, and Treatment by Richard Heimberg, Michael Liebowitz, Debra Hope, and Franklin Schneier