This article was downloaded by: [University of Texas Libraries] On: 26 November 2014, At: 13:54 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Psychoactive Drugs Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ujpd20
Marijuana, Expectancies, and Post-Traumatic Stress Symptoms: A Preliminary Investigation a
Mitch Earleywine Ph.D. & Jamie R. Bolles B.A.
b
a
Professor of Psychology and Director of Clinical Training, University at Albany, State University of New York, Albany, NY b
Graduate Associate Researcher in Psychology, University at Albany, State University of New York, Albany, NY Published online: 14 Jul 2014.
To cite this article: Mitch Earleywine Ph.D. & Jamie R. Bolles B.A. (2014) Marijuana, Expectancies, and Post-Traumatic Stress Symptoms: A Preliminary Investigation, Journal of Psychoactive Drugs, 46:3, 171-177, DOI: 10.1080/02791072.2014.920118 To link to this article: http://dx.doi.org/10.1080/02791072.2014.920118
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Psychoactive Drugs, 46 (3), 171–177, 2014 Copyright © Taylor & Francis Group, LLC ISSN: 0279-1072 print / 2159-9777 online DOI: 10.1080/02791072.2014.920118
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Marijuana, Expectancies, and Post-Traumatic Stress Symptoms: A Preliminary Investigation Mitch Earleywine, Ph.D.a & Jamie R. Bolles, B.A.b
Abstract — Previous work suggests that people might turn to marijuana to alleviate the symptoms of post-traumatic stress disorder (PTSD) and associated distress. Expectancy theories emphasize that the use of drugs correlates with their anticipated effects. The current study examined multivariate links among marijuana use, PTSD symptoms, and expectancies for marijuana-induced changes in those symptoms. Over 650 combat-exposed, male veterans who used marijuana at least once per week completed measures of PTSD symptoms, marijuana expectancies, and marijuana use in an Internet survey. Participants generally expected marijuana to relieve PTSD symptoms, especially those related to intrusions and arousal. Symptoms, expectancies for relief of symptoms, and marijuana consumption correlated significantly. Regressions revealed significant indirect effects of symptoms on use via expectancies, but no significant interactions of expectancies and symptoms. Combat-exposed veterans who use marijuana appear to use more as the magnitude of PTSD symptoms and their expectations of marijuana-induced relief of those symptoms increase. These results emphasize the importance of PTSD treatments in an effort to keep potential negative effects of marijuana to a minimum. They also underscore the import of expectancies in predicting marijuana use. Keywords — cannabis, expectancies, marijuana, PTSD symptoms
As Operation Iraqi Freedom approaches its second decade, a clear understanding of post-traumatic stress disorder (PTSD) is increasingly important. One estimate of the prevalence of PTSD among soldiers who have served in Iraq exceeds 30% (Sundin et al. 2010), suggesting that the correlates, causes, and treatments of PTSD
symptoms are worthy of investigation. Many who experience symptoms of the disorder also use marijuana (Calhoun et al. 2000; McFall, Mackay & Donovan 1992), and the National Comorbidity Study revealed that 18.5% of dependent users also have a PTSD diagnosis (Agosti, Nunes & Levin 2002). In addition, many people with PTSD have marijuana receptor genes that differ from those found in people without the disorder, suggesting that the cannabinoid system may be involved in intriguing ways (Onaivi 2009). Previous work suggests that marijuana has the potential to alleviate some of the symptoms of PTSD (Bonn-Miller et al. 2007; Bremner et al. 1996; Chilcoat & Breslau 1998), which might lead those who expect the plant to decrease their suffering to consume more marijuana. Even among the general population, expectancies about marijuana’s effects consistently covary with its use.
We thank Elana B. Gordis for comments on a previous draft. of Psychology and Director of Clinical Training, University at Albany, State University of New York, Albany, NY. b Graduate Associate Researcher in Psychology, University at Albany, State University of New York, Albany, NY. Please address correspondence to Mitch Earleywine, Ph.D., Habits and Lifestyles Laboratory, Department of Psychology, University at Albany, State University of New York, Albany, NY 12222; phone: +1-518-442-4836; fax: +1-518-442-4867; email: [email protected] a Professor
Journal of Psychoactive Drugs
171
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Those who anticipate that the plant will create a positive outcome tend to use more than those who expect otherwise (Hawkins 2007; Metrik et al. 2009; Skenderian et al. 2008; Vangsness, Bry & LaBouvie 2005). These results suggest that standard expectancy theories of marijuana use might apply to a self-medication hypothesis of marijuana consumption for PTSD symptoms. This self-medication could manifest multiple ways. Expectancies might moderate the link between PTSD symptoms and marijuana use. Thus, PTSD symptoms may covary with marijuana use primarily in those who hold expectancies that the plant will alleviate symptoms, but less so in those who do not hold these expectancies. Alternatively, the impact of PTSD symptoms on marijuana consumption might arise via an indirect effect through expectancies. Those with worse symptoms might develop stronger expectancies about marijuana-induced relief through personal experience, modeling, media, or word of mouth. These expectancies would then lead to more consumption of marijuana. Despite effusive clinical lore, few large-sample studies address expectancies about marijuana’s palliative effects for symptoms related to trauma. We sought to examine expectancies about marijuana that were specific to the relief of PTSD symptoms in combat-exposed veterans. We also examined these expectancies as a potential moderator, or as an intervening variable, for the link between PTSD symptoms and marijuana use. Finally, we sought to explore which clusters of symptoms could be the ones that veterans anticipate that the plant would help. Previous work has yet to address which symptoms of PTSD might be most related to marijuana in the minds of veterans. Expectations about marijuana’s effects on some symptoms first appeared millennia ago (e.g., insomnia, irritability; Earleywine 2005), but the plant’s impact on other symptoms has not been examined extensively (e.g., re-experiencing, numbing). The diagnosis has changed with the advent of the Diagnostic and Statistical Manual’s fifth edition (APA 2013), but many symptoms remain the same. The Posttraumatic Stress Disorder Checklist-Military Version (PCL-M; Weathers et al. 1993) remains one of the most frequently investigated assessments of symptoms. Recent confirmatory factor analyses on the scale in over 15,000 U.S. veterans revealed a four-factor structure of Intrusions (with items addressing disturbed memories or sudden reliving of traumatic events), Hyperarousal (trouble sleeping, irritability), Avoidance (avoiding thoughts or activities related to the trauma), and Numbing (feeling distant, emotionally numb; Mansfield et al. 2010). In an effort to see if marijuana, PTSD symptoms, and expectancies might covary differently for some symptoms than others, we examined the role of these individual factors and their associated expectancies on marijuana use.
Journal of Psychoactive Drugs
METHOD Procedure To identify veterans who also used marijuana, we recruited participants from a listserv devoted to altering marijuana prohibition, the one maintained by the National Organization for Reform of Marijuana Laws (NORML). An e-mail that included a link to the survey was sent to all listserv subscribers with a request for military veterans to participate for a chance to receive a prize.
Participants We sought to identify veterans who had combat experience and who used marijuana at least once per week. A total of 1,337 people completed the survey. Responses revealed that 29 of these were not veterans. An additional 344 veterans had no combat experience based on scores on the Combat Experiences Scale (described below). Unfortunately, only 68 veterans were women, making gender comparisons problematic. The small number of women led us to drop them from the analyses. Two men failed to provide consent (n = 2). Others gave inconsistent reports of marijuana use (n = 15), used marijuana less than once per week (n = 189), or failed infrequency items (n = 66). (Participants were asked to respond with a specific number or answer “T” explicitly on three items. Those who failed any of these simple questions were dropped.) We also screened for duplicate cases; none were evident. This left a sample of 653 combat-exposed, male veterans. Compared to those who were not selected, this selected sample had a younger mean age (39.85 (14.26) vs. 44.28 (13.63), t(1247) = 5.61), had higher mean scores on the combat experiences scale (13.70 vs. 3.29, t(1266) = 22.12, p < .001), and used slightly less marijuana per month (.84 oz/month (.57) vs. .96 oz/month (.57), t(1266) = 2.95, p < .05). They were also more likely to be single (Chisquared) = 17.27, p < .05). The groups did not differ on any other variables. Participants ranged in age from 18 to 87 (M = 40.11, SD = 14.27). Ethnic backgrounds included Caucasian (85%), African or Caribbean Descent (2%), Asian or Pacific Islander (1%), Hispanic or Latino (3%), Native American (1%), and mixed or multiple ethnic groups (8%). Marital status included 24% who were never married, 3% separated, 21% divorced, and 52% married. Fortyeight percent were junior enlisted men, with 45% noncommissioned officers and 7% officers. Forty-two percent were Army, 25% Navy, 16% Air Force, 15% Marine, and 2% Coast Guard. Years in the military ranged from 1 to 44 (M = 6.10 SD = 4.60). Education ranged from some high school through advanced degree, with mode of some college (45%).
172
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Measures PTSD symptoms. The Posttraumatic Stress Disorder Checklist-Military Version (PCL-M; Weathers et al. 1993) assessed PTSD symptoms associated with military experiences. Respondents were asked to refer to their military experiences in general rather than focusing on one specific traumatic event. The PCL-M consists of 17 items rated on a five-point Likert scale ranging from “1 = Not at all” to “5 = Extremely.” We treated scores as continuous rather than focusing on cut-offs, an approach consistent with taxometric work on PTSD (Broman-Fulks et al. 2006). Cronbach’s alpha was .95 in this sample. Average score was 45.07 (SD = 16.36), which exceeds recommended cutoffs for screening for the diagnosis in soldiers returning from combat (Bliese et al. 2008). Recent confirmatory factor analyses on PCL scores in over 15,000 U.S. veterans revealed a four-factor structure (Mansfield et al. 2010). We computed factor scores for each of these subscales. Average scores were 12.22 (SD = 5.33) for Intrusions, 14. 50 (SD = 5.06) for Hyperarousal, 4.92 (SD = 2.44) for Avoidance, and 13.42 (SD = 5.26) for Numbing. Internal consistency was high for all despite a smaller number of items (Cronbach’s alpha = .92 for Intrusions, .84 for Hyperarousal, .83 for Avoidance, and .84 for numbing). PTSD expectancies. Marijuana expectancy effects were assessed for each PTSD symptom by following every PCL item with a question stating “Marijuana makes this. . . .” Expectancy items were rated on a five-point Likert scale ranging from “−2 = Much Worse” to “2 = Much better.” Each item had scores reflecting the full range. Observed total scores ranged from −19 (suggesting that marijuana made symptoms worse) to +34 (suggesting marijuana improved symptoms dramatically), out of a possible range of −34 to +34 for the 17 items. The mean score was 18.43 (SD = 10.26), suggesting that, on average, these veterans expected improvement of PTSD symptoms from marijuana. Cronbach’s alpha was .93. Average scores for the expectancy subscales were 6.19 (SD = 3.59) for Intrusions Expectancies (EIntrude), 5.57 (SD = 3.03) for Hyperarousal Expectancies (EHyper), 2.10 (SD = 1.59) for Avoidance Expectancies (EAvoid), and 4.69 (SD = 3.51) for Numbing Expectancies (ENumb). Internal consistencies for the subscale expectancies were acceptable despite the smaller number of items (Cronbach’s alpha = .89 for EIntrude, .70 for EHyper, .77 for EAvoid, and .83 for ENumb). Marijuana use. Marijuana use was assessed by responses to the question “Approximately how much marijuana do you use in a month?” (1 = less than 1/4 ounce to 14 = more than 3 ounces, each response option increased by one-fourth of an ounce). Amounts ranged from one-fourth of an ounce to over three ounces per month. The modal response was three-fourths of an ounce per month. The distribution had a positive skew (.61), Journal of Psychoactive Drugs
but improved with a square-root transformation to .23. Analyses employed the square-root transformed ounces to better fit the assumptions of regression. We preferred this approach to alternatives that focus only on frequency of use; ignoring the quantity consumed can lead investigators to miss important correlates of marijuana consumption (Walden & Earleywine 2008). Combat experiences. The Combat Exposure Scale (CES; Keane et al. 1989) was created to assess stressors experienced during active duty of military personnel. The CES is a seven-item scale consisting of items such as “Were you ever under enemy fire?” Each item is rated on a fivepoint Likert scale ranging from “never experiencing the event” to “experiencing it more than 50 times.” The scoring algorithm weights more severe items, such as seeing someone get hit by enemy fire, more than other items, like going on patrols or having dangerous duties. We treated the CES as an index of combat exposure and only included participants who reported at least one potentially traumatic incident on at least one item. Total scores did not reach the maximum of the scale but ranged from 1 (light exposure) to 38 (heavy exposure). The mean score was 13.03 (SD = 9.04), which would be categorized as light-to-moderate exposure by the original developers. RESULTS We first sought to identify if participants expected greater marijuana-associated relief for some PTSD symptoms than others; the links among symptoms, expectancies, and use; the potential for expectancy-moderated links between symptoms and use; and the potential indirect effects of symptoms on use via expectancies. Given the large sample size and number of analyses, we used a modified Bonferroni approach to balance power and Type I error (Wilcox 2013). Each category of analyses (differences among average expectancies, bivariate relations, moderator effects, and indirect effects) received a family-wise error rate of p < .05. Thus, the six analyses for differences among average expectancies required p values of .0083 (.05/6) to be considered significant. The 55 bivariate correlations required a p value of .0009 (.05/55) to reach significance. The five moderator effects required a p value of .01, as did the analyses related to the five indirect effects. Differences among average expectancies. Because expectancy subscales varied in their number of items, we looked at the average single-item score for each subscale. Paired t-tests revealed that all means differed from each other, with expectancies for intrusions receiving the highest score (Mean = 1.28, SD = 0.69), followed by hyperarousal (Mean = 1.15, SD = 0.58), avoidance (Mean = 1.07, SD = 0.78), and numbing (Mean = 0.97, SD = 0.68). (T-tests (df = 650) ranged from 3.99−12.71, all ps < .001), with effect sizes ranging from d = .123 to .498. These differences among subscales suggested that veterans do 173
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
not see marijuana as equally effective for all symptoms. Nevertheless, all mean scores differed significantly from zero (ts(651) = 36. 24 to 47.23, all ps < .001), with effect sizes ranging from 1.36 to 1.98, suggesting that veterans generally expected some marijuana-related relief even for numbing, the symptoms least responsive to the plant. Bivariate relations. Correlations among PTSD symptoms, the four factors of the symptoms, expectancies, factor-specific expectancies, and use appear in Table 1. Total symptoms, each subscale of symptoms, total expectancies, and each subscale of expectancies correlated significantly with each other and with use. Moderator effects. We performed moderator analyses to see if links between symptoms and marijuana use were larger for participants with stronger expectancies for symptom relief. We regressed age and education as covariates because they are common predictors of marijuana use (DHHS 2012), then regressed PTSD total (or
subscale scores) and corresponding expectancy scores on use. Individual equations focused on each type of symptom and its related expectancies. Thus, PTSD total scores and expectancies for all symptoms appeared in one equation with their interaction to predict use. Hyperarousal symptoms and hyperarousal expectancies and their interaction predicted use in another equation (with no other symptoms or expectancies), and so on for each subscale of symptoms. Interaction terms computed as the product of the centered PTSD total (or subscale scores) and corresponding expectancy scores failed to reach significance at p < .01 in any of these equations, suggesting that the link between PTSD symptoms and use did not vary significantly with expectancies (see Table 2). Indirect effects. We hypothesized that links between symptoms and use arose via indirect paths through expectancies. We regressed age and education as covariates and PTSD total (or subscale scores) and corresponding
TABLE 1 Correlations among PTSD Symptoms, Expectancies, and Marijuana Use
1. PTSD 2. Avoidance 3. Hyperarousal 4. Intrusions 5. Numbing 6. Expectancies 7. EAvoid 8. EHyper 9. EIntrude 10. ENumb 11. Use
1
2
3
4
5
6
7
8
9
10
.86 .91 .91 .90 .45 .37 .36 .51 .30 .20
.70 .82 .70 .39 .42 .28 .46 .21 .16
.74 .78 .41 .30 .41 .41 .26 .12
.71 .40 .34 .28 .53 .21 .18
.41 .31 .30 .42 .35 .21
.83 .85 .87 .86 .27
.61 .74 .64 .20
.63 .66 .25
.58 .25
.20
All correlations are significant at p < .0009 (.05 Modified Bonferroni-corrected). PTSD = Post-traumatic stress symptoms checklist, Military version. Avoidance-Numbing = PTSD Factors. Expectancies = Anticipated marijuana-induced relief for PTSD symptoms. EAvoid-Enumb = Expectancies for each PTSD Factor. Use = Ounces of marijuana consumed per month.
TABLE 2 Predicting Marijuana Use from Symptoms and Expectancies (Beta, t-score)
Symptoms Expectancy Interaction
PCL .445 (1.84) .267 (3.10∗ ) −.365 (−1.29)
Avoidance .136 (3.54∗ ) .102 (2.84) −.022 (−0.65)
Hyperarousal .116 (3.22∗ ) .164 (4.64∗ ) −.037 (−1.13)
Intrusions .085 (2.02) .185 (4.63∗ ) .026 (9.71)
Numbing .164 (4.70∗ ) .133 (3.91∗ ) −.073 (−2.27)
∗
= p < .01. Standardized Beta (t-score) predicting ounces marijuana used per month from PTSD symptoms, associate expectancies, and their interaction. PCL = total score on the PCL. Avoidance-Numbing = relevant PTSD factor.
Journal of Psychoactive Drugs
174
Volume 46 (3), July – August 2014
Earleywine & Bolles
Marijuana and PTSD
TABLE 3 Regressions for Estimating Indirect Effects of PTSD Symptoms Predicting Marijuana Use via Expectancies
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
PTSD Total Intrusions Avoidance Hyperarousal Numbing
Symptom Beta .133 .076 .093 .089 .166
t 2.65∗∗ 1.66 2.16 2.13 4.04∗
Expect Beta .220 .216 .153 .213 .145
T 5.14∗∗ 4.76∗∗ 3.56∗∗ 5.06∗∗ 3.50∗∗
Indirect Effect .039∗∗ .023 .025∗∗ .031∗∗ .039∗∗
PTSD = Post-traumatic Stress Disorder as measured by the PCL. Symptom Beta = Beta weight for symptoms predicting marijuana use in equation with relevant expectancies. Expect Beta = Beta weight for expectancies predicting marijuana use in equation with relevant symptoms. t = t-test for beta weight of symptoms. T = t-test for beta weight of expectancies. Indirect = estimate of indirect effect based on bootstrap test with 5,000 iterations. ∗∗ P < .01.
expectancy scores on marijuana use. Thus, when total score on the PCL served as an index of PTSD symptoms, the expectancies for all symptoms served as the potential mediator. When only the numbing symptoms were used to predict marijuana use, the expectancies for numbing alone served as the potential mediator. We computed indirect effects via a bootstrapping method. Bootstrapping is a non-parametric approach that sidesteps some of the problems with power and Type I error common in other techniques. We generated 5,000 bootstrap samples by sampling with replacement to estimate the magnitude of the indirect effect, an associated standard error, and confidence intervals (Hayes 2013). All of the indirect effects but the one related to intrusions had 99% confidence intervals that did not include zero. These significant indirect paths suggest that PTSD symptoms can lead to marijuana consumption primarily through the link from symptoms to expectancies and from expectancies to use (see Table 3).
ounces of marijuana consumed per month. Correlations among symptoms and expectancies were generally in the small-to-moderate range (from approximate .2 to approximately .5; Cohen 1988). Symptoms and expectancies only showed small correlations with use (from approximately .15 to approximately .27), but seemed consistent with other work on marijuana expectancies (Vangsness et al. 2005). Moderator analyses revealed that links between PTSD symptoms and marijuana consumption did not vary significantly with expectancies. Regression analyses and tests of indirect effects showed that total PTSD symptoms linked to marijuana use through a significant (but small) indirect path via expectancies for total score and all but the intrusion-related symptoms. These data are consistent with a self-medication hypothesis of marijuana use for PTSD symptoms. As self-reported PTSD symptoms increased, so did expectations for marijuana-induced relief, which led to more marijuana consumption. Limitations of the current data suggest cautious interpretation of these results. First, the new diagnosis in the DSM-V (APA 2013) includes symptoms related to maladaptive blaming of oneself or others for the trauma or its aftermath, as well as reckless or self-destructive actions that accompany increased arousal. The current data do not address these new symptoms. In addition, self-reports of any type can prove problematic for a number of reasons. We used anonymous responding via the Internet to minimize the stigma associated with reports of symptoms or drug use. Data from participants who failed infrequency items were also omitted, potentially enhancing the validity of these reports. Nevertheless, these respondents were all affiliated in some way with marijuana legal reform. Like most practical sampling strategies, this approach to recruitment has pros and cons. Those who are willing to join the listserv might differ from other veterans and other marijuana users.
DISCUSSION A large sample of combat-exposed U.S. veterans who used marijuana at least once per week reported significant expectations of marijuana-induced relief from symptoms of PTSD. These veterans reported expecting marijuana-related improvement of all types of symptoms, but significantly more relief from symptoms of intrusions (like repeated, disturbing thoughts or dreams) than hyperarousal (trouble sleeping, irritability), followed by avoidance (avoiding thoughts or activities related to the trauma), and then numbing (feeling distant, emotionally numb). The variation in expected relief suggests that veterans do not see marijuana as a panacea for all PTSD symptoms. Correlations revealed significant links among total PTSD symptoms, each of the factors of symptoms, expectations about marijuana’s impact, and the number of
Journal of Psychoactive Drugs
175
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Despite these limitations, an Internet sample can provide more diverse and potentially more accurate results than other common methods of collecting data about substance use (Reips 2002). At least two recent studies support using the Internet for the study of illicit drug use (Rodgers et al. 2001; 2003). Marijuana prohibition might discourage users from studies that fail to provide the anonymity of the Internet. Participants also might fail to report deviant PTSD symptoms in a less anonymous setting (McClay et al. 2008: Rhodes, Bowie & Hergenrather 2003). At least one study has shown that respondents tend to report more drug use on an Internet survey than on an identical paperand-pencil questionnaire (Wang et al. 2005). This approach seemed promising for providing a large enough sample to address the hypotheses with reasonable statistical power as well. Participants were predominantly Caucasian, savvy enough with the computer to complete the questionnaire online, and exclusively male. It is unclear if similar marijuana expectancies would appear or function comparably in individuals who have experienced other traumas, in those with less experience with marijuana, or in groups showing more diversity in other ways. Extrapolating these findings to marijuana-involved, combat-exposed male veterans, however, might prove reasonable. The notion of PTSD-related expectancies for marijuana might also prove helpful to civilians who have experienced various traumas. These results have implications for marijuana-based medicine, PTSD symptoms, public policy, and substance abuse problems. Long-term, exposure-based treatment of PTSD symptoms, though far from perfect, has proven successful in multiple settings (Powers et al. 2010). People with PTSD symptoms who wish to decrease their
marijuana use might find the task easier after appropriate treatment. Challenging these marijuana expectancies might also prove useful. Training these individuals in alternative ways to handle these symptoms could lead to meaningful benefits. For example, effective cognitive behavioral strategies for improving sleep, lowering arousal, and decreasing irritability are numerous. Learning these skills could make decreasing marijuana consumption easier for those who desire to do so. These data also raise questions about marijuana-based pharmacological interventions for PTSD symptoms. The search for a pharmacological treatment for PTSD symptoms has led to trials of a host of medications (Bastien 2010; Henry, Fishman & Yougner 2007). Some of these medications might have more addictive potential than marijuana (Gore & Earleywine 2007; Nutt, King & Phillips 2010). These veteran reports of marijuana-related relief might inspire additions to the pharmacological armament against PTSD. Only continued work can answer these questions, but the potential implications of this kind of research are quite extensive. In addition, the idea of drug-related expectancies of relief for the symptoms of other disorders, both primarily physical and primarily psychiatric, could also prove useful in explaining links among symptoms, anticipated effects, and marijuana as well as with other classes of drugs. Clinical lore and case studies have linked a host of ailments to medical marijuana (Earleywine 2005), and use of the plant might covary with expected symptom relief for a number of disorders. FUNDING This work was supported by Cannabis Science Inc.
REFERENCES use coping motives: A preliminary investigation. Journal of Dual Diagnosis 6 (2): 111–122. Bonn-Miller, M.O.; Harris, A.H.S. & Trafton, J.A. 2012. Prevalence of cannabis use disorder diagnoses among veterans in 2002, 2008, and 2009. Psychological Services 9 (4): 404–416. Bonn-Miller, M.O.; Vujanovic, A.A. & Drescher, K.D. 2011. Cannabis use among military veterans after residential treatment for posttraumatic stress disorder. Psychology of Addictive Behaviors 25 (3): 485–491. Bonn-Miller, M.O.; Vujanovic, A.A.; Feldner, M.T.; Bernstein, A. & Zvolensky, M.J. 2007. Posttraumatic stress symptom severity predicts marijuana use coping motives among traumatic event-exposed marijuana users. Journal of Traumatic Stress 20 (4): 577–586. Bremner, J.D.; Southwick, S.M.; Darnell, A. & Charney, D.S. 1996. Chronic PTSD in Vietnam combat veterans: Course of illness and substance abuse. American Journal of Psychiatry 153 (3): 369–375. Broman-Fulks, J.J.; Ruggiero, K.J.; Green, B.A.; Kilpatrick, D.G.; Danielson, C.K.; Resnick, H.S. & Saunders, B.E. 2006. Taxometric investigation of PTSD: Data from two nationally representative samples. Behavior Therapy 37 (4): 364–380.
Agosti, V.; Nunes, E. & Levin, F. 2002. Rates of psychiatric comorbidity among U.S. residents with lifetime marijuana dependence. The American Journal of Drug and Alcohol Abuse 28 (4): 643–652. American Psychiatric Association. 2013. DSM 5. Washington, DC: American Psychiatric Association. Bastein, D.L. 2010. Pharmacological treatment of combat- induced PTSD: A literature review. British Journal of Nursing 19 (5): 318–321. Bliese, P.D.; Wright, K.M.; Adler, A.B.; Cabrera, O.; Castrol, C.A. & Hoge, C.W. 2008. Validating the primary care posttraumatic stress disorder screen and the posttraumatic stress disorder checklist with soldiers returning from combat. Journal of Consulting and Clinical Psychology 76 (2): 272–281. Boden, M.T.; Babson, K.A.; Vujanovic, A.A.; Short, N.A. & Bonn-Miller, M.O. 2013. Posttraumatic stress disorder and cannabis use characteristics among military veterans with cannabis dependence. The American Journal on Addictions 22 (3): 277–284. Boden, M.T.; McKay, J.R.; Long, W.R. & Bonn-Miller, M.O. 2013. The effects of cannabis use expectancies on self-initiated cannabis cessation. Addiction 108 (9): 1649–1657. Bonn-Miller, M.O.; Babson, K.A.; Vujanovic, A.A. & Feldner, M.T. 2010. Sleep problems and PTSD symptoms interact to predict marijuana
Journal of Psychoactive Drugs
176
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Metrik, J.; Rohsenow, D.J.; Monti, P.M.; McGeary, J.; Cook, T.A.R.; de Wit, H.; et al. 2009. Effectiveness of a marijuana expectancy manipulation: Piloting the balanced-placebo design for marijuana. Experimental and Clinical Psychopharmacology 17 (4): 217–225. Nutt, D.J.; King, L.A. & Phillips, L.D. 2010. Drug harms in the UK: A multicriteria decision analysis. Lancet 376 (9752): 1558–65. Onaivi, E.S. 2009. Cannabinoid receptors in brain: Pharmacogenetics, neuropharmacology, neurotoxicology, and potential therapeutic applications. International Review of Neurobiology 88: 335–369. Powers, M.B.; Halpern, J.M.; Ferenschak, M.P.; Gillihan, S.J. & Foa, E.B. 2010. A meta-analytic review of prolonged exposure for posttraumatic stress disorder. Clinical Psychology Review 30 (6): 635–641. Reips, U. 2002. Standards for internet-based experimenting. Experimental Psychology 49 (4): 243–256. Rhodes, S.D.; Bowie, D.A. & Hergenrather, K.C. 2003. Collecting behavioural data using the world wide web: Considerations for researchers. Journal of Epidemiology and Community Health 57 (4): 68–73. Rodgers, J.; Buchanan, T.; Scholey, A.B.; Heffernan, T.M.; Ling, J. & Parrott, A. 2001. Differential effects of ecstasy and cannabis on self-reports of memory ability: A web-based study. Human Psychopharmacology 16 (8): 619–625. Rodgers, J.; Buchanan, T.; Scholey, A.B.; Heffernan, T.M.; Ling, J. & Parrott, A.C. 2003. Patterns of drug use and the influence of gender on self-reports of memory ability in ecstasy users: A web-based study. Journal of Psychopharmacology 17 (4): 389–396. Simons, J.S. & Arens, A.M. 2007. Moderating effects of sensitivity to punishment and sensitivity to reward on associations between marijuana effect expectancies and use. Psychology of Addictive Behaviors 21 (3): 409–414. Skenderian, J.J.; Siegel, J.T.; Crano, W.D.; Alvaro, E.E. & Lac, A. 2008. Expectancy change and adolescents’ intention to use marijuana. Psychology of Addictive Behaviors 22 (4): 563–569. Sundin, J.; Fear, N.T.; Iversen, A.; Rona, R.J. & Wessely, S. 2010. PTSD after deployment to Iraq: Conflicting rates, conflicting claims. Psychological Medicine 40 (3): 367–382. Vangsness, L.; Bry, B.H. & LaBouvie, E.W. 2005. Impulsivity, negative expectancies, and marijuana use: A test of the acquired preparedness model. Addictive Behaviors 30 (5): 1071–1076. Walden, N. & Earleywine, M. 2008. How high? Quantity as a predictor of marijuana-related problems. Harm Reduction Journal 5: 20. Available at: http://www.harmreductionjournal.com/content/5/1/20. Wang, Y.; Lee, C.; Lew-Ting, C.; Hsiao, C.K.; Chen, D. & Chen, W.J. 2005. Survey of substance use among high school students in Taipei: Web-based questionnaire versus paper-and-pencil questionnaire. Journal of Adolescent Health 37 (4): 289–295. Weathers, F.W.; Litz, B.T.; Herman, D.S.; Huska, J.A. & Keane, T.M. 1993. The PTSD Checklist PCL: Reliability, validity, and diagnostic utility. Paper presented at the Annual Meeting of the International Society of Traumatic Stress Studies, San Antonio, TX, June 19–21, 1993. Wilcox, R.R. 2013. New Statistical Procedures for the Social Sciences: Modern Solutions to Basic Problems. Florence, KY: Psychology Press.
Buckner, J.D. & Schmidt, N.B. 2008. Marijuana effect expectancies: Relations to social anxiety and marijuana use problems. Addictive Behaviors 33 (11): 1477–1483. Calhoun, P.S.; Sampson, W.S.; Bosworth, H.B.; Feldman, M.E.; Kirby, A.C.; Hertzberg, M.A.; Wampler, T.P.; Tate-Williams, F.; Moore, S.D. & Beckham, J.C. 2000. Drug use and validity of substance use self-reports in veterans seeking help for posttraumatic stress disorder. Journal of Consulting and Clinical Psychology 68 (5): 923–927. Chilcoat, H.D. & Breslau, N. 1998. Posttraumatic stress disorder and drug disorders: Testing causal pathways. Archives of General Psychiatry 55 (10): 913–917. Cougle, J.R.; Bonn-Miller, M.O.; Vujanovic, A.A.; Zvolensky, M.J. & Hawkins, K.E. 2011. Posttraumatic stress disorder and cannabis use in a nationally representative sample. Psychology of Addictive Behaviors 25 (3): 554–558. Department of Health and Human Services. 2012. Results from the 2009 National Survey on Drug Use and Health: Volume I. Substance Abuse and Mental Health Services Administration, Office of Applied Studies, Washington, DC. Earleywine, M. 2005. Understanding Marijuana: A New Look at the Scientific Evidence. New York: Oxford University Press. Earleywine, M. 2011. The elephant in the room with the grass ceiling. Addiction 106 (2): 245. Gore, R. & Earleywine, M. 2007. Marijuana’s perceived addictiveness: A survey of clinicians and researchers. In M. Earleywine (Ed.) Pot Politics: Marijuana and the Costs of Prohibition, pp. 176–186. New York: Oxford University Press. Hawkins, L. 2007. The impact of personality and affect on college students’ motives for marijuana use. Dissertation Abstracts International: Section B: The Sciences and Engineering 68: 1306. Henry, M.; Fishman, J.R.; Youngner, S. J. 2007. Propranolol and the prevention of post-traumatic stress disorder: Is it wrong to erase the “sting” of bad memories? The American Journal of Bioethics: AJOB 7 (9): 12–20. Keane, T.M.; Fairbank, J.A.; Caddell, J.M.; Zimering, R.T.; Taylor, K.L. & Mora, C.A. 1989. Clinical evaluation of a measure to assess combat exposure. Journal of Consulting and Clinical Psychology 1 (1): 53–55. Ling, J. & Parrott, A.C. 2003. Patterns of drug use and the influence of gender on self-reports of memory ability in ecstasy users: A webbased study. Journal of Psychopharmacology 17 (4): 389–396. Mansfield, A.J.; Williams, J.; Hourani, L.L. & Babeu, L.A. 2010. Measurement invariance of posttraumatic stress disorder symptoms among U.S. military personnel. Journal of Traumatic Stress 23 (1): 91–99. McClay, R.N.; Deal, W.E.; Murphy, J.A.; Center, K.B.; Kolkow, T.T. & Grieger, T.A. 2008. On-the-record screenings versus anonymous surveys in reporting PTSD. American Journal of Psychiatry 165 (6): 775–776. McFall, M.E.; Mackay, P.W. & Donovan, D.M. 1992. Combat-related posttraumatic stress disorder and severity of substance abuse in Vietnam veterans. Journal of Studies on Alcohol and Drugs 53 (4): 357–363.
Journal of Psychoactive Drugs
177
Volume 46 (3), July – August 2014
Journal of Psychoactive Drugs Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ujpd20
Marijuana, Expectancies, and Post-Traumatic Stress Symptoms: A Preliminary Investigation a
Mitch Earleywine Ph.D. & Jamie R. Bolles B.A.
b
a
Professor of Psychology and Director of Clinical Training, University at Albany, State University of New York, Albany, NY b
Graduate Associate Researcher in Psychology, University at Albany, State University of New York, Albany, NY Published online: 14 Jul 2014.
To cite this article: Mitch Earleywine Ph.D. & Jamie R. Bolles B.A. (2014) Marijuana, Expectancies, and Post-Traumatic Stress Symptoms: A Preliminary Investigation, Journal of Psychoactive Drugs, 46:3, 171-177, DOI: 10.1080/02791072.2014.920118 To link to this article: http://dx.doi.org/10.1080/02791072.2014.920118
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Journal of Psychoactive Drugs, 46 (3), 171–177, 2014 Copyright © Taylor & Francis Group, LLC ISSN: 0279-1072 print / 2159-9777 online DOI: 10.1080/02791072.2014.920118
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Marijuana, Expectancies, and Post-Traumatic Stress Symptoms: A Preliminary Investigation Mitch Earleywine, Ph.D.a & Jamie R. Bolles, B.A.b
Abstract — Previous work suggests that people might turn to marijuana to alleviate the symptoms of post-traumatic stress disorder (PTSD) and associated distress. Expectancy theories emphasize that the use of drugs correlates with their anticipated effects. The current study examined multivariate links among marijuana use, PTSD symptoms, and expectancies for marijuana-induced changes in those symptoms. Over 650 combat-exposed, male veterans who used marijuana at least once per week completed measures of PTSD symptoms, marijuana expectancies, and marijuana use in an Internet survey. Participants generally expected marijuana to relieve PTSD symptoms, especially those related to intrusions and arousal. Symptoms, expectancies for relief of symptoms, and marijuana consumption correlated significantly. Regressions revealed significant indirect effects of symptoms on use via expectancies, but no significant interactions of expectancies and symptoms. Combat-exposed veterans who use marijuana appear to use more as the magnitude of PTSD symptoms and their expectations of marijuana-induced relief of those symptoms increase. These results emphasize the importance of PTSD treatments in an effort to keep potential negative effects of marijuana to a minimum. They also underscore the import of expectancies in predicting marijuana use. Keywords — cannabis, expectancies, marijuana, PTSD symptoms
As Operation Iraqi Freedom approaches its second decade, a clear understanding of post-traumatic stress disorder (PTSD) is increasingly important. One estimate of the prevalence of PTSD among soldiers who have served in Iraq exceeds 30% (Sundin et al. 2010), suggesting that the correlates, causes, and treatments of PTSD
symptoms are worthy of investigation. Many who experience symptoms of the disorder also use marijuana (Calhoun et al. 2000; McFall, Mackay & Donovan 1992), and the National Comorbidity Study revealed that 18.5% of dependent users also have a PTSD diagnosis (Agosti, Nunes & Levin 2002). In addition, many people with PTSD have marijuana receptor genes that differ from those found in people without the disorder, suggesting that the cannabinoid system may be involved in intriguing ways (Onaivi 2009). Previous work suggests that marijuana has the potential to alleviate some of the symptoms of PTSD (Bonn-Miller et al. 2007; Bremner et al. 1996; Chilcoat & Breslau 1998), which might lead those who expect the plant to decrease their suffering to consume more marijuana. Even among the general population, expectancies about marijuana’s effects consistently covary with its use.
We thank Elana B. Gordis for comments on a previous draft. of Psychology and Director of Clinical Training, University at Albany, State University of New York, Albany, NY. b Graduate Associate Researcher in Psychology, University at Albany, State University of New York, Albany, NY. Please address correspondence to Mitch Earleywine, Ph.D., Habits and Lifestyles Laboratory, Department of Psychology, University at Albany, State University of New York, Albany, NY 12222; phone: +1-518-442-4836; fax: +1-518-442-4867; email: [email protected] a Professor
Journal of Psychoactive Drugs
171
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Those who anticipate that the plant will create a positive outcome tend to use more than those who expect otherwise (Hawkins 2007; Metrik et al. 2009; Skenderian et al. 2008; Vangsness, Bry & LaBouvie 2005). These results suggest that standard expectancy theories of marijuana use might apply to a self-medication hypothesis of marijuana consumption for PTSD symptoms. This self-medication could manifest multiple ways. Expectancies might moderate the link between PTSD symptoms and marijuana use. Thus, PTSD symptoms may covary with marijuana use primarily in those who hold expectancies that the plant will alleviate symptoms, but less so in those who do not hold these expectancies. Alternatively, the impact of PTSD symptoms on marijuana consumption might arise via an indirect effect through expectancies. Those with worse symptoms might develop stronger expectancies about marijuana-induced relief through personal experience, modeling, media, or word of mouth. These expectancies would then lead to more consumption of marijuana. Despite effusive clinical lore, few large-sample studies address expectancies about marijuana’s palliative effects for symptoms related to trauma. We sought to examine expectancies about marijuana that were specific to the relief of PTSD symptoms in combat-exposed veterans. We also examined these expectancies as a potential moderator, or as an intervening variable, for the link between PTSD symptoms and marijuana use. Finally, we sought to explore which clusters of symptoms could be the ones that veterans anticipate that the plant would help. Previous work has yet to address which symptoms of PTSD might be most related to marijuana in the minds of veterans. Expectations about marijuana’s effects on some symptoms first appeared millennia ago (e.g., insomnia, irritability; Earleywine 2005), but the plant’s impact on other symptoms has not been examined extensively (e.g., re-experiencing, numbing). The diagnosis has changed with the advent of the Diagnostic and Statistical Manual’s fifth edition (APA 2013), but many symptoms remain the same. The Posttraumatic Stress Disorder Checklist-Military Version (PCL-M; Weathers et al. 1993) remains one of the most frequently investigated assessments of symptoms. Recent confirmatory factor analyses on the scale in over 15,000 U.S. veterans revealed a four-factor structure of Intrusions (with items addressing disturbed memories or sudden reliving of traumatic events), Hyperarousal (trouble sleeping, irritability), Avoidance (avoiding thoughts or activities related to the trauma), and Numbing (feeling distant, emotionally numb; Mansfield et al. 2010). In an effort to see if marijuana, PTSD symptoms, and expectancies might covary differently for some symptoms than others, we examined the role of these individual factors and their associated expectancies on marijuana use.
Journal of Psychoactive Drugs
METHOD Procedure To identify veterans who also used marijuana, we recruited participants from a listserv devoted to altering marijuana prohibition, the one maintained by the National Organization for Reform of Marijuana Laws (NORML). An e-mail that included a link to the survey was sent to all listserv subscribers with a request for military veterans to participate for a chance to receive a prize.
Participants We sought to identify veterans who had combat experience and who used marijuana at least once per week. A total of 1,337 people completed the survey. Responses revealed that 29 of these were not veterans. An additional 344 veterans had no combat experience based on scores on the Combat Experiences Scale (described below). Unfortunately, only 68 veterans were women, making gender comparisons problematic. The small number of women led us to drop them from the analyses. Two men failed to provide consent (n = 2). Others gave inconsistent reports of marijuana use (n = 15), used marijuana less than once per week (n = 189), or failed infrequency items (n = 66). (Participants were asked to respond with a specific number or answer “T” explicitly on three items. Those who failed any of these simple questions were dropped.) We also screened for duplicate cases; none were evident. This left a sample of 653 combat-exposed, male veterans. Compared to those who were not selected, this selected sample had a younger mean age (39.85 (14.26) vs. 44.28 (13.63), t(1247) = 5.61), had higher mean scores on the combat experiences scale (13.70 vs. 3.29, t(1266) = 22.12, p < .001), and used slightly less marijuana per month (.84 oz/month (.57) vs. .96 oz/month (.57), t(1266) = 2.95, p < .05). They were also more likely to be single (Chisquared) = 17.27, p < .05). The groups did not differ on any other variables. Participants ranged in age from 18 to 87 (M = 40.11, SD = 14.27). Ethnic backgrounds included Caucasian (85%), African or Caribbean Descent (2%), Asian or Pacific Islander (1%), Hispanic or Latino (3%), Native American (1%), and mixed or multiple ethnic groups (8%). Marital status included 24% who were never married, 3% separated, 21% divorced, and 52% married. Fortyeight percent were junior enlisted men, with 45% noncommissioned officers and 7% officers. Forty-two percent were Army, 25% Navy, 16% Air Force, 15% Marine, and 2% Coast Guard. Years in the military ranged from 1 to 44 (M = 6.10 SD = 4.60). Education ranged from some high school through advanced degree, with mode of some college (45%).
172
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Measures PTSD symptoms. The Posttraumatic Stress Disorder Checklist-Military Version (PCL-M; Weathers et al. 1993) assessed PTSD symptoms associated with military experiences. Respondents were asked to refer to their military experiences in general rather than focusing on one specific traumatic event. The PCL-M consists of 17 items rated on a five-point Likert scale ranging from “1 = Not at all” to “5 = Extremely.” We treated scores as continuous rather than focusing on cut-offs, an approach consistent with taxometric work on PTSD (Broman-Fulks et al. 2006). Cronbach’s alpha was .95 in this sample. Average score was 45.07 (SD = 16.36), which exceeds recommended cutoffs for screening for the diagnosis in soldiers returning from combat (Bliese et al. 2008). Recent confirmatory factor analyses on PCL scores in over 15,000 U.S. veterans revealed a four-factor structure (Mansfield et al. 2010). We computed factor scores for each of these subscales. Average scores were 12.22 (SD = 5.33) for Intrusions, 14. 50 (SD = 5.06) for Hyperarousal, 4.92 (SD = 2.44) for Avoidance, and 13.42 (SD = 5.26) for Numbing. Internal consistency was high for all despite a smaller number of items (Cronbach’s alpha = .92 for Intrusions, .84 for Hyperarousal, .83 for Avoidance, and .84 for numbing). PTSD expectancies. Marijuana expectancy effects were assessed for each PTSD symptom by following every PCL item with a question stating “Marijuana makes this. . . .” Expectancy items were rated on a five-point Likert scale ranging from “−2 = Much Worse” to “2 = Much better.” Each item had scores reflecting the full range. Observed total scores ranged from −19 (suggesting that marijuana made symptoms worse) to +34 (suggesting marijuana improved symptoms dramatically), out of a possible range of −34 to +34 for the 17 items. The mean score was 18.43 (SD = 10.26), suggesting that, on average, these veterans expected improvement of PTSD symptoms from marijuana. Cronbach’s alpha was .93. Average scores for the expectancy subscales were 6.19 (SD = 3.59) for Intrusions Expectancies (EIntrude), 5.57 (SD = 3.03) for Hyperarousal Expectancies (EHyper), 2.10 (SD = 1.59) for Avoidance Expectancies (EAvoid), and 4.69 (SD = 3.51) for Numbing Expectancies (ENumb). Internal consistencies for the subscale expectancies were acceptable despite the smaller number of items (Cronbach’s alpha = .89 for EIntrude, .70 for EHyper, .77 for EAvoid, and .83 for ENumb). Marijuana use. Marijuana use was assessed by responses to the question “Approximately how much marijuana do you use in a month?” (1 = less than 1/4 ounce to 14 = more than 3 ounces, each response option increased by one-fourth of an ounce). Amounts ranged from one-fourth of an ounce to over three ounces per month. The modal response was three-fourths of an ounce per month. The distribution had a positive skew (.61), Journal of Psychoactive Drugs
but improved with a square-root transformation to .23. Analyses employed the square-root transformed ounces to better fit the assumptions of regression. We preferred this approach to alternatives that focus only on frequency of use; ignoring the quantity consumed can lead investigators to miss important correlates of marijuana consumption (Walden & Earleywine 2008). Combat experiences. The Combat Exposure Scale (CES; Keane et al. 1989) was created to assess stressors experienced during active duty of military personnel. The CES is a seven-item scale consisting of items such as “Were you ever under enemy fire?” Each item is rated on a fivepoint Likert scale ranging from “never experiencing the event” to “experiencing it more than 50 times.” The scoring algorithm weights more severe items, such as seeing someone get hit by enemy fire, more than other items, like going on patrols or having dangerous duties. We treated the CES as an index of combat exposure and only included participants who reported at least one potentially traumatic incident on at least one item. Total scores did not reach the maximum of the scale but ranged from 1 (light exposure) to 38 (heavy exposure). The mean score was 13.03 (SD = 9.04), which would be categorized as light-to-moderate exposure by the original developers. RESULTS We first sought to identify if participants expected greater marijuana-associated relief for some PTSD symptoms than others; the links among symptoms, expectancies, and use; the potential for expectancy-moderated links between symptoms and use; and the potential indirect effects of symptoms on use via expectancies. Given the large sample size and number of analyses, we used a modified Bonferroni approach to balance power and Type I error (Wilcox 2013). Each category of analyses (differences among average expectancies, bivariate relations, moderator effects, and indirect effects) received a family-wise error rate of p < .05. Thus, the six analyses for differences among average expectancies required p values of .0083 (.05/6) to be considered significant. The 55 bivariate correlations required a p value of .0009 (.05/55) to reach significance. The five moderator effects required a p value of .01, as did the analyses related to the five indirect effects. Differences among average expectancies. Because expectancy subscales varied in their number of items, we looked at the average single-item score for each subscale. Paired t-tests revealed that all means differed from each other, with expectancies for intrusions receiving the highest score (Mean = 1.28, SD = 0.69), followed by hyperarousal (Mean = 1.15, SD = 0.58), avoidance (Mean = 1.07, SD = 0.78), and numbing (Mean = 0.97, SD = 0.68). (T-tests (df = 650) ranged from 3.99−12.71, all ps < .001), with effect sizes ranging from d = .123 to .498. These differences among subscales suggested that veterans do 173
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
not see marijuana as equally effective for all symptoms. Nevertheless, all mean scores differed significantly from zero (ts(651) = 36. 24 to 47.23, all ps < .001), with effect sizes ranging from 1.36 to 1.98, suggesting that veterans generally expected some marijuana-related relief even for numbing, the symptoms least responsive to the plant. Bivariate relations. Correlations among PTSD symptoms, the four factors of the symptoms, expectancies, factor-specific expectancies, and use appear in Table 1. Total symptoms, each subscale of symptoms, total expectancies, and each subscale of expectancies correlated significantly with each other and with use. Moderator effects. We performed moderator analyses to see if links between symptoms and marijuana use were larger for participants with stronger expectancies for symptom relief. We regressed age and education as covariates because they are common predictors of marijuana use (DHHS 2012), then regressed PTSD total (or
subscale scores) and corresponding expectancy scores on use. Individual equations focused on each type of symptom and its related expectancies. Thus, PTSD total scores and expectancies for all symptoms appeared in one equation with their interaction to predict use. Hyperarousal symptoms and hyperarousal expectancies and their interaction predicted use in another equation (with no other symptoms or expectancies), and so on for each subscale of symptoms. Interaction terms computed as the product of the centered PTSD total (or subscale scores) and corresponding expectancy scores failed to reach significance at p < .01 in any of these equations, suggesting that the link between PTSD symptoms and use did not vary significantly with expectancies (see Table 2). Indirect effects. We hypothesized that links between symptoms and use arose via indirect paths through expectancies. We regressed age and education as covariates and PTSD total (or subscale scores) and corresponding
TABLE 1 Correlations among PTSD Symptoms, Expectancies, and Marijuana Use
1. PTSD 2. Avoidance 3. Hyperarousal 4. Intrusions 5. Numbing 6. Expectancies 7. EAvoid 8. EHyper 9. EIntrude 10. ENumb 11. Use
1
2
3
4
5
6
7
8
9
10
.86 .91 .91 .90 .45 .37 .36 .51 .30 .20
.70 .82 .70 .39 .42 .28 .46 .21 .16
.74 .78 .41 .30 .41 .41 .26 .12
.71 .40 .34 .28 .53 .21 .18
.41 .31 .30 .42 .35 .21
.83 .85 .87 .86 .27
.61 .74 .64 .20
.63 .66 .25
.58 .25
.20
All correlations are significant at p < .0009 (.05 Modified Bonferroni-corrected). PTSD = Post-traumatic stress symptoms checklist, Military version. Avoidance-Numbing = PTSD Factors. Expectancies = Anticipated marijuana-induced relief for PTSD symptoms. EAvoid-Enumb = Expectancies for each PTSD Factor. Use = Ounces of marijuana consumed per month.
TABLE 2 Predicting Marijuana Use from Symptoms and Expectancies (Beta, t-score)
Symptoms Expectancy Interaction
PCL .445 (1.84) .267 (3.10∗ ) −.365 (−1.29)
Avoidance .136 (3.54∗ ) .102 (2.84) −.022 (−0.65)
Hyperarousal .116 (3.22∗ ) .164 (4.64∗ ) −.037 (−1.13)
Intrusions .085 (2.02) .185 (4.63∗ ) .026 (9.71)
Numbing .164 (4.70∗ ) .133 (3.91∗ ) −.073 (−2.27)
∗
= p < .01. Standardized Beta (t-score) predicting ounces marijuana used per month from PTSD symptoms, associate expectancies, and their interaction. PCL = total score on the PCL. Avoidance-Numbing = relevant PTSD factor.
Journal of Psychoactive Drugs
174
Volume 46 (3), July – August 2014
Earleywine & Bolles
Marijuana and PTSD
TABLE 3 Regressions for Estimating Indirect Effects of PTSD Symptoms Predicting Marijuana Use via Expectancies
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
PTSD Total Intrusions Avoidance Hyperarousal Numbing
Symptom Beta .133 .076 .093 .089 .166
t 2.65∗∗ 1.66 2.16 2.13 4.04∗
Expect Beta .220 .216 .153 .213 .145
T 5.14∗∗ 4.76∗∗ 3.56∗∗ 5.06∗∗ 3.50∗∗
Indirect Effect .039∗∗ .023 .025∗∗ .031∗∗ .039∗∗
PTSD = Post-traumatic Stress Disorder as measured by the PCL. Symptom Beta = Beta weight for symptoms predicting marijuana use in equation with relevant expectancies. Expect Beta = Beta weight for expectancies predicting marijuana use in equation with relevant symptoms. t = t-test for beta weight of symptoms. T = t-test for beta weight of expectancies. Indirect = estimate of indirect effect based on bootstrap test with 5,000 iterations. ∗∗ P < .01.
expectancy scores on marijuana use. Thus, when total score on the PCL served as an index of PTSD symptoms, the expectancies for all symptoms served as the potential mediator. When only the numbing symptoms were used to predict marijuana use, the expectancies for numbing alone served as the potential mediator. We computed indirect effects via a bootstrapping method. Bootstrapping is a non-parametric approach that sidesteps some of the problems with power and Type I error common in other techniques. We generated 5,000 bootstrap samples by sampling with replacement to estimate the magnitude of the indirect effect, an associated standard error, and confidence intervals (Hayes 2013). All of the indirect effects but the one related to intrusions had 99% confidence intervals that did not include zero. These significant indirect paths suggest that PTSD symptoms can lead to marijuana consumption primarily through the link from symptoms to expectancies and from expectancies to use (see Table 3).
ounces of marijuana consumed per month. Correlations among symptoms and expectancies were generally in the small-to-moderate range (from approximate .2 to approximately .5; Cohen 1988). Symptoms and expectancies only showed small correlations with use (from approximately .15 to approximately .27), but seemed consistent with other work on marijuana expectancies (Vangsness et al. 2005). Moderator analyses revealed that links between PTSD symptoms and marijuana consumption did not vary significantly with expectancies. Regression analyses and tests of indirect effects showed that total PTSD symptoms linked to marijuana use through a significant (but small) indirect path via expectancies for total score and all but the intrusion-related symptoms. These data are consistent with a self-medication hypothesis of marijuana use for PTSD symptoms. As self-reported PTSD symptoms increased, so did expectations for marijuana-induced relief, which led to more marijuana consumption. Limitations of the current data suggest cautious interpretation of these results. First, the new diagnosis in the DSM-V (APA 2013) includes symptoms related to maladaptive blaming of oneself or others for the trauma or its aftermath, as well as reckless or self-destructive actions that accompany increased arousal. The current data do not address these new symptoms. In addition, self-reports of any type can prove problematic for a number of reasons. We used anonymous responding via the Internet to minimize the stigma associated with reports of symptoms or drug use. Data from participants who failed infrequency items were also omitted, potentially enhancing the validity of these reports. Nevertheless, these respondents were all affiliated in some way with marijuana legal reform. Like most practical sampling strategies, this approach to recruitment has pros and cons. Those who are willing to join the listserv might differ from other veterans and other marijuana users.
DISCUSSION A large sample of combat-exposed U.S. veterans who used marijuana at least once per week reported significant expectations of marijuana-induced relief from symptoms of PTSD. These veterans reported expecting marijuana-related improvement of all types of symptoms, but significantly more relief from symptoms of intrusions (like repeated, disturbing thoughts or dreams) than hyperarousal (trouble sleeping, irritability), followed by avoidance (avoiding thoughts or activities related to the trauma), and then numbing (feeling distant, emotionally numb). The variation in expected relief suggests that veterans do not see marijuana as a panacea for all PTSD symptoms. Correlations revealed significant links among total PTSD symptoms, each of the factors of symptoms, expectations about marijuana’s impact, and the number of
Journal of Psychoactive Drugs
175
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Despite these limitations, an Internet sample can provide more diverse and potentially more accurate results than other common methods of collecting data about substance use (Reips 2002). At least two recent studies support using the Internet for the study of illicit drug use (Rodgers et al. 2001; 2003). Marijuana prohibition might discourage users from studies that fail to provide the anonymity of the Internet. Participants also might fail to report deviant PTSD symptoms in a less anonymous setting (McClay et al. 2008: Rhodes, Bowie & Hergenrather 2003). At least one study has shown that respondents tend to report more drug use on an Internet survey than on an identical paperand-pencil questionnaire (Wang et al. 2005). This approach seemed promising for providing a large enough sample to address the hypotheses with reasonable statistical power as well. Participants were predominantly Caucasian, savvy enough with the computer to complete the questionnaire online, and exclusively male. It is unclear if similar marijuana expectancies would appear or function comparably in individuals who have experienced other traumas, in those with less experience with marijuana, or in groups showing more diversity in other ways. Extrapolating these findings to marijuana-involved, combat-exposed male veterans, however, might prove reasonable. The notion of PTSD-related expectancies for marijuana might also prove helpful to civilians who have experienced various traumas. These results have implications for marijuana-based medicine, PTSD symptoms, public policy, and substance abuse problems. Long-term, exposure-based treatment of PTSD symptoms, though far from perfect, has proven successful in multiple settings (Powers et al. 2010). People with PTSD symptoms who wish to decrease their
marijuana use might find the task easier after appropriate treatment. Challenging these marijuana expectancies might also prove useful. Training these individuals in alternative ways to handle these symptoms could lead to meaningful benefits. For example, effective cognitive behavioral strategies for improving sleep, lowering arousal, and decreasing irritability are numerous. Learning these skills could make decreasing marijuana consumption easier for those who desire to do so. These data also raise questions about marijuana-based pharmacological interventions for PTSD symptoms. The search for a pharmacological treatment for PTSD symptoms has led to trials of a host of medications (Bastien 2010; Henry, Fishman & Yougner 2007). Some of these medications might have more addictive potential than marijuana (Gore & Earleywine 2007; Nutt, King & Phillips 2010). These veteran reports of marijuana-related relief might inspire additions to the pharmacological armament against PTSD. Only continued work can answer these questions, but the potential implications of this kind of research are quite extensive. In addition, the idea of drug-related expectancies of relief for the symptoms of other disorders, both primarily physical and primarily psychiatric, could also prove useful in explaining links among symptoms, anticipated effects, and marijuana as well as with other classes of drugs. Clinical lore and case studies have linked a host of ailments to medical marijuana (Earleywine 2005), and use of the plant might covary with expected symptom relief for a number of disorders. FUNDING This work was supported by Cannabis Science Inc.
REFERENCES use coping motives: A preliminary investigation. Journal of Dual Diagnosis 6 (2): 111–122. Bonn-Miller, M.O.; Harris, A.H.S. & Trafton, J.A. 2012. Prevalence of cannabis use disorder diagnoses among veterans in 2002, 2008, and 2009. Psychological Services 9 (4): 404–416. Bonn-Miller, M.O.; Vujanovic, A.A. & Drescher, K.D. 2011. Cannabis use among military veterans after residential treatment for posttraumatic stress disorder. Psychology of Addictive Behaviors 25 (3): 485–491. Bonn-Miller, M.O.; Vujanovic, A.A.; Feldner, M.T.; Bernstein, A. & Zvolensky, M.J. 2007. Posttraumatic stress symptom severity predicts marijuana use coping motives among traumatic event-exposed marijuana users. Journal of Traumatic Stress 20 (4): 577–586. Bremner, J.D.; Southwick, S.M.; Darnell, A. & Charney, D.S. 1996. Chronic PTSD in Vietnam combat veterans: Course of illness and substance abuse. American Journal of Psychiatry 153 (3): 369–375. Broman-Fulks, J.J.; Ruggiero, K.J.; Green, B.A.; Kilpatrick, D.G.; Danielson, C.K.; Resnick, H.S. & Saunders, B.E. 2006. Taxometric investigation of PTSD: Data from two nationally representative samples. Behavior Therapy 37 (4): 364–380.
Agosti, V.; Nunes, E. & Levin, F. 2002. Rates of psychiatric comorbidity among U.S. residents with lifetime marijuana dependence. The American Journal of Drug and Alcohol Abuse 28 (4): 643–652. American Psychiatric Association. 2013. DSM 5. Washington, DC: American Psychiatric Association. Bastein, D.L. 2010. Pharmacological treatment of combat- induced PTSD: A literature review. British Journal of Nursing 19 (5): 318–321. Bliese, P.D.; Wright, K.M.; Adler, A.B.; Cabrera, O.; Castrol, C.A. & Hoge, C.W. 2008. Validating the primary care posttraumatic stress disorder screen and the posttraumatic stress disorder checklist with soldiers returning from combat. Journal of Consulting and Clinical Psychology 76 (2): 272–281. Boden, M.T.; Babson, K.A.; Vujanovic, A.A.; Short, N.A. & Bonn-Miller, M.O. 2013. Posttraumatic stress disorder and cannabis use characteristics among military veterans with cannabis dependence. The American Journal on Addictions 22 (3): 277–284. Boden, M.T.; McKay, J.R.; Long, W.R. & Bonn-Miller, M.O. 2013. The effects of cannabis use expectancies on self-initiated cannabis cessation. Addiction 108 (9): 1649–1657. Bonn-Miller, M.O.; Babson, K.A.; Vujanovic, A.A. & Feldner, M.T. 2010. Sleep problems and PTSD symptoms interact to predict marijuana
Journal of Psychoactive Drugs
176
Volume 46 (3), July – August 2014
Downloaded by [University of Texas Libraries] at 13:54 26 November 2014
Earleywine & Bolles
Marijuana and PTSD
Metrik, J.; Rohsenow, D.J.; Monti, P.M.; McGeary, J.; Cook, T.A.R.; de Wit, H.; et al. 2009. Effectiveness of a marijuana expectancy manipulation: Piloting the balanced-placebo design for marijuana. Experimental and Clinical Psychopharmacology 17 (4): 217–225. Nutt, D.J.; King, L.A. & Phillips, L.D. 2010. Drug harms in the UK: A multicriteria decision analysis. Lancet 376 (9752): 1558–65. Onaivi, E.S. 2009. Cannabinoid receptors in brain: Pharmacogenetics, neuropharmacology, neurotoxicology, and potential therapeutic applications. International Review of Neurobiology 88: 335–369. Powers, M.B.; Halpern, J.M.; Ferenschak, M.P.; Gillihan, S.J. & Foa, E.B. 2010. A meta-analytic review of prolonged exposure for posttraumatic stress disorder. Clinical Psychology Review 30 (6): 635–641. Reips, U. 2002. Standards for internet-based experimenting. Experimental Psychology 49 (4): 243–256. Rhodes, S.D.; Bowie, D.A. & Hergenrather, K.C. 2003. Collecting behavioural data using the world wide web: Considerations for researchers. Journal of Epidemiology and Community Health 57 (4): 68–73. Rodgers, J.; Buchanan, T.; Scholey, A.B.; Heffernan, T.M.; Ling, J. & Parrott, A. 2001. Differential effects of ecstasy and cannabis on self-reports of memory ability: A web-based study. Human Psychopharmacology 16 (8): 619–625. Rodgers, J.; Buchanan, T.; Scholey, A.B.; Heffernan, T.M.; Ling, J. & Parrott, A.C. 2003. Patterns of drug use and the influence of gender on self-reports of memory ability in ecstasy users: A web-based study. Journal of Psychopharmacology 17 (4): 389–396. Simons, J.S. & Arens, A.M. 2007. Moderating effects of sensitivity to punishment and sensitivity to reward on associations between marijuana effect expectancies and use. Psychology of Addictive Behaviors 21 (3): 409–414. Skenderian, J.J.; Siegel, J.T.; Crano, W.D.; Alvaro, E.E. & Lac, A. 2008. Expectancy change and adolescents’ intention to use marijuana. Psychology of Addictive Behaviors 22 (4): 563–569. Sundin, J.; Fear, N.T.; Iversen, A.; Rona, R.J. & Wessely, S. 2010. PTSD after deployment to Iraq: Conflicting rates, conflicting claims. Psychological Medicine 40 (3): 367–382. Vangsness, L.; Bry, B.H. & LaBouvie, E.W. 2005. Impulsivity, negative expectancies, and marijuana use: A test of the acquired preparedness model. Addictive Behaviors 30 (5): 1071–1076. Walden, N. & Earleywine, M. 2008. How high? Quantity as a predictor of marijuana-related problems. Harm Reduction Journal 5: 20. Available at: http://www.harmreductionjournal.com/content/5/1/20. Wang, Y.; Lee, C.; Lew-Ting, C.; Hsiao, C.K.; Chen, D. & Chen, W.J. 2005. Survey of substance use among high school students in Taipei: Web-based questionnaire versus paper-and-pencil questionnaire. Journal of Adolescent Health 37 (4): 289–295. Weathers, F.W.; Litz, B.T.; Herman, D.S.; Huska, J.A. & Keane, T.M. 1993. The PTSD Checklist PCL: Reliability, validity, and diagnostic utility. Paper presented at the Annual Meeting of the International Society of Traumatic Stress Studies, San Antonio, TX, June 19–21, 1993. Wilcox, R.R. 2013. New Statistical Procedures for the Social Sciences: Modern Solutions to Basic Problems. Florence, KY: Psychology Press.
Buckner, J.D. & Schmidt, N.B. 2008. Marijuana effect expectancies: Relations to social anxiety and marijuana use problems. Addictive Behaviors 33 (11): 1477–1483. Calhoun, P.S.; Sampson, W.S.; Bosworth, H.B.; Feldman, M.E.; Kirby, A.C.; Hertzberg, M.A.; Wampler, T.P.; Tate-Williams, F.; Moore, S.D. & Beckham, J.C. 2000. Drug use and validity of substance use self-reports in veterans seeking help for posttraumatic stress disorder. Journal of Consulting and Clinical Psychology 68 (5): 923–927. Chilcoat, H.D. & Breslau, N. 1998. Posttraumatic stress disorder and drug disorders: Testing causal pathways. Archives of General Psychiatry 55 (10): 913–917. Cougle, J.R.; Bonn-Miller, M.O.; Vujanovic, A.A.; Zvolensky, M.J. & Hawkins, K.E. 2011. Posttraumatic stress disorder and cannabis use in a nationally representative sample. Psychology of Addictive Behaviors 25 (3): 554–558. Department of Health and Human Services. 2012. Results from the 2009 National Survey on Drug Use and Health: Volume I. Substance Abuse and Mental Health Services Administration, Office of Applied Studies, Washington, DC. Earleywine, M. 2005. Understanding Marijuana: A New Look at the Scientific Evidence. New York: Oxford University Press. Earleywine, M. 2011. The elephant in the room with the grass ceiling. Addiction 106 (2): 245. Gore, R. & Earleywine, M. 2007. Marijuana’s perceived addictiveness: A survey of clinicians and researchers. In M. Earleywine (Ed.) Pot Politics: Marijuana and the Costs of Prohibition, pp. 176–186. New York: Oxford University Press. Hawkins, L. 2007. The impact of personality and affect on college students’ motives for marijuana use. Dissertation Abstracts International: Section B: The Sciences and Engineering 68: 1306. Henry, M.; Fishman, J.R.; Youngner, S. J. 2007. Propranolol and the prevention of post-traumatic stress disorder: Is it wrong to erase the “sting” of bad memories? The American Journal of Bioethics: AJOB 7 (9): 12–20. Keane, T.M.; Fairbank, J.A.; Caddell, J.M.; Zimering, R.T.; Taylor, K.L. & Mora, C.A. 1989. Clinical evaluation of a measure to assess combat exposure. Journal of Consulting and Clinical Psychology 1 (1): 53–55. Ling, J. & Parrott, A.C. 2003. Patterns of drug use and the influence of gender on self-reports of memory ability in ecstasy users: A webbased study. Journal of Psychopharmacology 17 (4): 389–396. Mansfield, A.J.; Williams, J.; Hourani, L.L. & Babeu, L.A. 2010. Measurement invariance of posttraumatic stress disorder symptoms among U.S. military personnel. Journal of Traumatic Stress 23 (1): 91–99. McClay, R.N.; Deal, W.E.; Murphy, J.A.; Center, K.B.; Kolkow, T.T. & Grieger, T.A. 2008. On-the-record screenings versus anonymous surveys in reporting PTSD. American Journal of Psychiatry 165 (6): 775–776. McFall, M.E.; Mackay, P.W. & Donovan, D.M. 1992. Combat-related posttraumatic stress disorder and severity of substance abuse in Vietnam veterans. Journal of Studies on Alcohol and Drugs 53 (4): 357–363.
Journal of Psychoactive Drugs
177
Volume 46 (3), July – August 2014
