Available online at www.sciencedirect.com

ScienceDirect Behavior Therapy 45 (2014) 745 – 759

www.elsevier.com/locate/bt

Brief Interventions to Reduce Ecstasy Use: A Multi-Site Randomized Controlled Trial Melissa M. Norberg Centre for Emotional Health, Department of Psychology, Macquarie University National Cannabis Prevention and Information Centre, UNSW Medicine, Sydney National Drug and Alcohol Research Centre, UNSW Medicine, Sydney Leanne Hides Centre for Youth Substance Abuse Research, Institute of Health & Biomedical Innovation, School of Psychology and Counselling, Queensland University of Technology Jake Olivier School of Mathematics and Statistics, UNSW, Sydney Laila Khawar National Drug and Alcohol Research Centre, UNSW Medicine, Sydney Rebecca McKetin National Drug and Alcohol Research Centre, UNSW Medicine, Sydney Centre for Research on Ageing, Health and Well-being, the Australian National University, Canberra Jan Copeland National Cannabis Prevention and Information Centre, UNSW Medicine, Sydney

Studies examining the ability of motivational enhancement therapy (MET) to augment education provision among ecstasy users have produced mixed results and none have examined whether treatment fidelity was related to ecstasy use outcomes. The primary objectives of this multi-site, parallel, two-group randomized controlled trial were to determine if a single-session of MET could instill greater commitment to change and reduce ecstasy use and related problems more so than an education-only intervention and whether MET sessions delivered with higher treatment fidelity are associated with better outcomes. The secondary objective was to assess participants’ satisfaction with their assigned interventions. Participants (N = 174; Mage = 23.62) at two Australian universities were allocated randomly to receive a 15-minute educational session on ecstasy use (n = 85) or a 50-minute session of MET that

included an educational component (n = 89). Primary outcomes were assessed at baseline, and then at 4-, 16-, and 24-weeks postbaseline, while the secondary outcome measure was assessed 4-weeks postbaseline by researchers blind to treatment allocation. Overall, the treatment fidelity was acceptable to good in the MET condition. There were no statistical differences at follow-up between the groups on the primary outcomes of ecstasy use, ecstasy-related problems, and commitment to change. Both intervention groups reported a 50% reduction in their ecstasy use and a 20% reduction in the severity of their ecstasy-related problems at the 24-week follow up. Commitment to change slightly improved for both groups (9%–17%). Despite the lack of between-group statistical differences on primary outcomes, participants who received a single session of MET were slightly more satisfied with their intervention than those who received education only. MI fidelity was not

746

norberg et al.

associated with ecstasy use outcomes. Given these findings, future research should focus on examining mechanisms of change. Such work may suggest new methods for enhancing outcomes. Australia and New Zealand Clinical Trial Registry: ACTRN12611000136909

Keywords: MDMA; randomized controlled trial; motivational enhancement therapy; ecstacy; education

AN ESTIMATED 19 MILLION PEOPLE WORLDWIDE use ecstasy annually (United Nations Office on Drugs and Crime, 2013). Australia and New Zealand have the highest prevalence of ecstasy use in the world (United Nations Office on Drugs and Crime, 2013), with 3% of Australians reporting ecstasy use in 2010 (Australian Institute of Health and Welfare, 2011). Even the typical person who uses only a few times a year (Australian Institute of Health and Welfare, 2011) may experience serotonergic neurotoxicity (de Win et al., 2008; Di lorio et al., 2012). More frequent use is associated with greater impairments in sleep (McCann, Sgambati, Schwartz, & Ricaurte, 2009), neuroendocrine functioning (Parrott, 2009), prospective memory (Rendell, Gray, Henry, & Tolan, 2007), visual information processing (Dickson, Bruno, & Brown, 2009), and higher-order cognitive functioning (Fox, Parrott, & Turner, 2001; Hanson & Luciana, 2004), as well as depression (Rogers et al., 2009). About one in six who use ecstasy do so at least monthly (Australian Institute of Health and Welfare, 2011), and about a third of these individuals may meet diagnostic criteria for ecstasy dependence, with the most common symptoms being use despite harms, tolerance, and spending a significant amount of time obtaining, using, and recovering from ecstasy use (McKetin et al., 2014). Ecstasy users need to be informed about these wide-ranging effects and encouraged to decrease their use. Providing individuals with educational materials about ecstasy use within a motivational framework might be the optimal method for achieving this goal. Most ecstasy users report wanting more information about ecstasy, particularly its shortand long-term effects and how to decrease risks This study was funded by the National Health & Medical Research Council/ Project Grant (630570). This funding source had no role in the conduct of this study or in its publication. Address correspondence to Melissa M. Norberg, Ph.D., Centre for Emotional Health, Department of Psychology, Macquarie University, Sydney, NSW AUSTRALIA 2109; e-mail: [email protected]. 0005-7894/45/745-759/$1.00/0 © 2014 Association for Behavioral and Cognitive Therapies. Published by Elsevier Ltd. All rights reserved.

when using it (Carlson, Falck, McCaughan, & Siegal, 2004; Topp, Hando, Dillon, Roche, & Solowij, 1999), and less than 15% report being interested in formal treatment (Sindicich & Burns, 2013; Topp et al., 1999). As such, ecstasy users may be open to receiving support that is largely educational, brief, and not marketed as treatment. This educational material may need to be provided in a nonjudgmental and nonadversarial manner in order to increase the likelihood that ecstasy users will use the information to change their behavior (Carlson et al., 2004). These assumptions led to three published comparisons of motivational enhancement therapy (MET) to an educational control condition for ecstasy users (Huang, Tang, Lin, & Yen, 2011; Marsden et al., 2006; Martin & Copeland, 2010). In the first trial, 342 young people (Mage = 18) in the United Kingdom who primarily used cocaine or ecstasy were randomly allocated to receive one session of MET or a pamphlet on the health risks of stimulants and alcohol (Marsden et al.). Participants’ preferences determined whether caseworkers focused the manualized MET intervention on ecstasy, cocaine, or alcohol use. At baseline, ecstasy users (77%) had used ecstasy at an average of two pills per occasion on 18 of the past 90 days. At the 6-month follow-up, there were no statistical differences between groups. Both groups reported decreasing their ecstasy use to 1.5 pills per occasion on 8 of the past 90 days, a 60% reduction. In the second trial, 50 ecstasy users (Mage = 29) in Australia were assigned randomly to receive one session of MET or educational materials (Martin & Copeland). At baseline, participants reported using an average of 1.5 pills per occasion on 10 of the last 90 days. At the 3-month follow-up, participants in the MET group had decreased their ecstasy use by 32%, while the control condition had not made a change. This disparity was not statistically different; however, groups statistically differed in their level of ecstasy-related problems. At the 3-month follow-up, the MET group reported one dependence symptom on average, while the control group reported two. In the third trial, 200 ecstasy users (Mage = 17) were recruited from a juvenile abstinence center in Taiwan (Huang et al., 2011). This study differed from the previous two, in that ecstasy use was not assessed as presumably abstinent participants were randomly assigned to receive three sessions of MET delivered over 1 week or educational materials about ecstasy use and methods for resisting its use (Huang et al.). At the only follow-up (1-week postbaseline), groups did not statistically differ on how committed they were to changing. Taken together, the results of the three clinical trials

brief interventions to reduce ecstasy use

747

Completed telephone screen (n=444) Excluded (n=257) ♦ Ineligible (n=92) ♦ Declined to participate (n=47) ♦ Unable to contact (n=54) ♦ Did not attend baseline assessment (n=64)

Enrollment

Completed baseline assessment (n=187)

Ineligible (n=13) ♦ Infrequent ecstasy use (n=8) ♦ Moderate to severe alcohol dependence (n=4) ♦ Moderate to severe stimulant dependence (n=1)

Randomized (n=174)

Allocation Allocated to E Check-up (n=89)

Allocated to Education-only (n=85)

Follow-Up 4-week Follow-up Fully Completed (n=70) ♦ Partial assessment only (n=11) ♦ Lost to follow-up (n=7) ♦ Dropped out (n=1)

4-week Follow-up Fully Completed (n=79) ♦ Partial assessment only (n=4) ♦ Lost to follow-up (n=2)

16-week Follow-up Fully Completed (n=68) ♦ Partial assessment only (n=10) ♦ Lost to follow-up (n=10) ♦ Dropped out (n=1)

16-week Follow-up Fully Completed (n=70) ♦ Partial assessment only (n=6) ♦ Lost to follow-up (n=9)

24-week Follow-up Fully Completed (n=66) ♦ Partial assessment only (n=5) ♦ Lost to follow-up (n=16) ♦ Dropped out (n=2)

24-week Follow-up Fully Completed (n=68) ♦ Partial assessment only (n=7) ♦ Lost to follow-up (n=11)

FIGURE 1

CONSORT flow chart. This figure illustrates study recruitment, intervention allocation, and retention.

are conflicting. While the larger-scale studies suggest that MET may not increase motivation to change or reduce ecstasy use over and above the provision of educational materials, the smaller study suggests that it may be able to reduce problems to a greater degree. Therapist fidelity to motivational interviewing (MI) principles was not monitored in any of these studies, indicating that the lack of substantial evidence for MET may be due to poor therapist adherence, a predictor of poor treatment outcomes (Apodaca & Longabaugh, 2009). Research using client-rated and practitioner-rated fidelity measures have shown that greater use and quality of MI components are associated with better substance use outcomes (McNally, Palfai, & Kahler, 2005; Strang & McCambridge, 2004). Despite a developer of MI (Miller, 2001) advising that all studies should assess

fidelity through direct monitoring of the intervention, few have done so. Given the mixed findings, methodological differences, and lack of reporting on therapist fidelity among prior studies, we conducted the current study to determine whether MET delivered with a high level of treatment fidelity was more efficacious than an MI-informed educational intervention among ecstasy users. We chose to deliver the educational intervention within an MI-framework to ensure therapists did not deliver the psychoeducation in an argumentative or prescriptive manner, which is inconsistent with the needs of ecstasy users (Carlson et al., 2004) and has been shown to predict poor outcomes (Gibbons et al., 2010). MI is a way of interacting with clients, whereas MET analyzes and feedbacks assessment information to the client within the spirit of MI

748

norberg et al.

Table 1

Baseline Characteristics of Participants (N = 174) Variable

M [95% CI]

Mean Age a Mean Years of Education a Proportion Male Proportion Australian Born Proportion Heterosexual b Proportion In a Relationship Proportion FT Employed b Proportion Drinker d Proportion Opiate User d Proportion Cannabis User d Proportion Cocaine User d Proportion Stimulant User d Proportion Sedative User d Proportion Tobacco User d Ecstasy Outcomes No. of pills in 90 days Days of use in 90 days URICA Readiness score c URICA Action score c SDS score b Proportion SDS score ≥ 4 b

23.62 [22.85, 24.39] 14.11 [13.79, 14.43]

P [95% CI]

.65 [.58, .68 [.58, .75 [.68, .57 [.49, .21 [.15, .98 [.96, .14 [.08, .79 [.73, .52 [.45, .52 [.45, .27 [.20, .68 [.61,

.72] .77] .81] .64] .27] 1.00] .18] .85] .60] .60] .34] .75]

14.08 [12.18, 15.99] 6.77 [6.14, 7.40] 5.36 [4.97, 5.76] 2.70 [2.57, 2.82] 2.46 [2.16, 2.77] .30 [.23, .36]

Note. FT = Full-time; No. = Number; URICA = University of Rhode Island Change Assessment; SDS = Severity of Dependence Scale. Participants were classified as a drinker, opiate, cannabis, cocaine, stimulant, sedative, or tobacco user if they reported any use of the substances in the past 90 days on the Brief Treatment Outcome Measure. aTwo participants missing data. bOne participant missing data. c Five participants missing data. dThree participants missing data.

(Miller & Rollnick, 2002; Project MATCH Research Group, 1997). A recent meta-analysis found MET produces statistically significantly better outcomes than MI alone (Lundahl, Kunz, Brownell, Tollefson, & Burke, 2010). Based on this meta-analysis and previous ecstasy use treatment studies, we hypothesized that individuals who received one session of MET (the E Check-up) would have better outcomes (ecstasy use, ecstasy-related problems, and commitment to change) at follow-up than individuals who received an MI-informed educational intervention. Additionally, we predicted that MET sessions in which therapists demonstrated better adherence and competence to MI principles would be related to better outcomes. Our primary outcome variables were assessed at 4-weeks, 16-weeks, and 24-weeks postbaseline. Our secondary outcome variable, satisfaction with the assigned intervention, was assessed only at the 4-week follow-up. We predicted that both groups would report high levels of treatment satisfaction.

Method participants We enrolled 174 individuals (see Figure 1 for CONSORT flow chart; see Table 1 for descriptive baseline characteristics). To be included in the study,

participants were required to be fluent in English, over 16 years of age, and to have used ecstasy on at least three different occasions during the past 90 days. We originally included only those individuals who had used ecstasy at least six times in the past 90 days. However, we updated this inclusion criterion 7 months into recruitment, as nearly 70% of ineligible individuals had not used ecstasy twice a month. Nine participants who were initially ineligible based on their frequency of use enrolled in the study after this change. Participants recruited before the change (n = 11) reported using ecstasy on an average of 9.18 days (SD = 3.03) in the 90 days preceding enrollment, whereas participants recruited after the change reported using ecstasy on an average of 6.61 days (SD = 4.24). Individuals were excluded if they met criteria for moderate to severe substance dependence for another drug (excluding cannabis and tobacco) as determined by the Patient Edition of the Structured Clinical Interview for DSM-IV-TR Axis I Disorders: Research Version (SCID-I-RV; First, Spitzer, Gibbon, & Williams, 2007). Additionally, we excluded individuals if they had received substance use treatment in the last 90 days or showed evidence of obvious medical, cognitive, or psychological impairment that would interfere with participation.

brief interventions to reduce ecstasy use measures Primary Outcomes We chose the following measures based on their psychometric properties and for their ability to allow for comparisons to previously published trials. The Timeline Followback (TLFB; Sobell & Sobell, 1996) was used to assess quantity and frequency of ecstasy use on a month-to-month basis. The TLFB has fair to good test-retest reliability for the assessment of hallucinogens and amphetamines (Fals-Stewart, O'Farrell, Freitas, McFarlin, & Rutigliano, 2000) and telephone administrations obtain similar findings to face-to-face interviews (Sobell, Brown, Leo, & Sobell, 1996). Due to the absence of accurate dosage data, TLFB quantity estimates were based on the number of pills taken. When ecstasy was not taken in pill form, we assumed the following quantities were equivalent to one pill: 1 capsule, .25 grams of powder, 1.25 lines, and 1 pinch (Sindicich & Burns, 2013). Severity of ecstasy-related problems was measured using the 5-item Severity of Dependence Scale (SDS; Gossop, Griffiths, Powis, & Strang, 1992). The SDS has satisfactory test-retest reliability (Gossop, Best, Marsden, & Strang, 1997) and scores of 4 and higher suggest the presence of significant ecstasy-related problems (Bruno et al., 2009). Across baseline to the 24-week follow-up, the internal consistency of the SDS ranged from .66 to .79. The 32-item University of Rhode Island Change Assessment Scale (URICA; McConnaughy, Prochaska, & Velicer, 1983) was used to gauge motivation for changing ecstasy use. Readiness to Change scores range from − 2 to + 14; scores lower than 8 indicate precontemplation, scores of 8 to 11 indicate contemplation, and scores above 12 indicate preparation (DiClemente, Schlundt, & Gemmell, 2004). Findings from Project MATCH suggest that Readiness to Change scores should be used prior to treatment, whereas Action scores should be used to measure change after treatment (DiClemente, Carbonari, Zweben, Morrel, & Lee, 2001). Action scores range from one to five, with higher scores indicating greater commitment to changing drug use. The URICA Action subscale has good internal consistency and construct validity (Field, Adinoff, Harris, Ball, & Carroll, 2009). In the current study, the internal consistency of the URICA Action subscale ranged from .90 to .92. Secondary Outcome The 8-item Client Satisfaction Questionnaire (CSQ-8; Attkisson & Greenfield, 2004) assessed how much participants valued their assigned interventions. Scores range from 8 to 32, with higher scores indicating greater satisfaction. The measure has demonstrated excellent reliability and

749

moderate predictive validity (Attkisson & Greenfield). At the 4-week follow-up, the CSQ achieved an α = .84. Therapy Process The 4-item version of the Credibility and Expectancy Questionnaire (CEQ; Devilly & Borkovec, 2000) was used to measure participants’ acceptance of the intervention rationale after their randomization was unveiled. Three items are rated on an 11-point scale from 1 (not at all) to 10 (very) and assess how credible the intervention is, how successful the intervention will be at reducing ecstasy use, and how confident participants would be in recommending the intervention to a friend. The fourth item assesses how much of a reduction in ecstasy use will occur as a result of an intervention and is rated from 0% to 100%. The CEQ has adequate test-retest reliability (Devilly & Borkovec). The Motivational Interviewing section of the second edition of the Yale Adherence and Competence Scale (YACS-II; Nuro et al., 2005) was used to rate therapists’ skill and the extent of MI delivered during the E Check-up. As such, E Check-up sessions were audiotaped and rated. Scoring involved an independent rater listening to each session recording twice, once to score nine MI process variables (e.g., affirmation of strengths and self-efficacy and fostering a collaborative atmosphere) for adherence (frequency and extensiveness) and a second time to score these same process variables for competence (skill level). Adherence and Competence scores range from 1 (not at all/ very poor) to 7 (extensive/excellent), with scores of 4 indicating somewhat and adequate. In a largescale study assessing therapist fidelity across 37 trained therapists working at five different community treatment centers, YACS-II Motivational Interviewing Adherence scores were 3.8 (SD = .90) and Competence scores were 4.6 (SD = .90), on average (Carroll et al., 2006). In a smaller-scale therapist training effectiveness study with four therapists, therapists achieved posttraining competence subscale scores ranging from 2.33 (SD = 0.58) to 4.3 (SD = 0.56; Davis, Devitt, O'Neill, Kaiser, & Mueser, 2014). These studies show that MI fidelity with substance abuse clients has ranged from poor to adequate. Demographics and Diagnoses Additional items queried participants’ age, gender, education, sexual orientation, relationship status, country of birth, employment status, and alcohol and drug use in the last 90 days. Further, the substance use module of the SCID-I-RV (First et al., 2007) was used to assess substance dependence on other drugs for exclusionary purposes. The substance use module

750

norberg et al.

has excellent interrater reliability and fair test-retest reliability (Zanarini et al., 2000).

procedures (see appendix a) Study Design The study was a multi-site, parallel, two-group randomized controlled trial conducted at the University of New South Wales in Sydney, Australia, and the Queensland University of Technology in Brisbane, Australia. Using simple randomization for each site, we allocated participants to either the E Check-up or the education control group. The first author used the http://www.randomization. com/ website to configure randomization. She then concealed each participants’ allocation using sealed opaque envelopes. Envelopes were not opened by a therapist until a participant had completed the baseline assessment and was deemed eligible. Participants completed all primary outcome measures at baseline, and again at 4-weeks, 16-weeks, and 24-weeks postbaseline via post, email, or face‐to‐ face interview, dependent on participant preference. The CSQ-8 was completed at the 4-week follow-up only. If participants reported they were unable to complete a follow-up appointment, research assistants encouraged them to complete the TLFB over the phone. Research assistants were blind to treatment allocation to avoid ascertainment bias. Both therapists and participants were instructed not to disclose allocation assignments to the research assistants and data indicative of allocations (e.g., session notes, therapist logs) were not made available to the research assistants. Participants received $25, $35, $40, and $40 for the baseline, 4-, 16-, and 24-week follow-up assessments, respectively. The last participant completed the final follow-up assessment in April 2012. The Human Research Ethics Committees at the University of New South Wales, the Queensland University of Technology, and at relevant New South Wales Area Health Services (for recruitment purposes) approved the conduct of this study. The study was registered (ACTRN12611000136909) in the Australia and New Zealand Clinical Trial Registry (https://www.anzctr.org.au/Trial/ Registration/TrialReview.aspx?id=336495) in February 2011 and funded by National Health & Medical Research Council/ Project Grant (630570). Recruitment and Enrollment Recruitment began in January 2010 and terminated in October 2011. Participants were recruited using print and online media advertisements in magazines and on help-seeking and social networking websites. Additionally, we posted flyers and brochures within drug, health, and mental health organiza-

tions and we distributed flyers on university campuses, at pubs and bars, and at festivals and music venues. Ads and flyers asked individuals who used ecstasy to “come talk to us.” Additionally, they instructed individuals that participation was confidential and involved completing an E Check-up, followed by three follow-up assessments, to which individuals would be reimbursed $25 to $40 per assessment. Due to slower than anticipated recruitment, we implemented respondent-driven sampling in July 2010. Enrolled participants were asked to recruit up to three friends and acquaintances, receiving $25 for each referral who completed a baseline assessment. Interested individuals made initial contact with research staff via telephone or email. They were informed that the study was examining the effectiveness of an E Check-up, a nonjudgmental program for learning about ecstasy use and how it might affect them. Individuals were instructed they would never be pressured into quitting. Individuals who provided verbal consent to participate were then telephone screened for initial eligibility. Potentially eligible individuals were posted or emailed baseline self‐report measures. Once these measures were completed, research assistants scheduled the baseline assessment. At the baseline assessment, therapists explained the study in detail and obtained written informed consent. Therapists then administered the SCIDI-RV (First et al., 2007) and the TLFB (Sobell & Sobell, 1996) and obtained contact details for two individuals who could locate a participant if the researchers were unable to do so. Therapists then determined if participants met inclusion/exclusion criteria. Therapists thanked ineligible participants for their time and excused them from the study. Eligible participants had their randomization unveiled and their allocated intervention briefly explained to them (semistandardized) before therapists asked them to complete the CEQ. The interventions then started after a short break. The break allowed therapists time to complete the Personalized Feedback Report for those assigned to the E Check-up. Education Participants randomized to the MI-informed educational condition were given the 15-page ecstasy booklet, Ecstasy: Facts and Fiction (2nd ed., Silins, Bleeker, & Martin, 2008), to review in conjunction with their therapist. The booklet covers a variety of issues related to ecstasy use, including its history, its consequences, and methods of harm reduction. Therapists reviewed the booklet and answered any questions posed by participants within 15 minutes

brief interventions to reduce ecstasy use in an MI-consistent manner. Specifically, therapists were instructed to use core interviewing skills, such as asking open-ended questions (e.g., “What are you hoping to learn from this educational booklet?”) and using reflection (e.g., “You're just interested in hearing about the consequences of ecstasy use and you would rather not discuss the history of its use”). In this way, therapists were instructed to develop a strong therapeutic alliance with their participants by listening to their concerns and by avoiding arguments and prescribing change to clients when delivering the psychoeducational material. Therapists were not encouraged to evoke change talk or plan for change in this condition. Participants were allowed to keep the booklet. MET The goal of the E Check-up was to motivate participants to reduce their ecstasy use within a 50-minute motivational session. When MI is combined with personalized feedback and education it is referred to as MET (Miller & Rollnick, 2002; Project MATCH Research Group, 1997). Accordingly, therapists reviewed the Ecstasy: Facts and Fiction booklet (Silins et al., 2008) and provided participants with structured feedback to their baseline assessment results using a Personal Feedback Report (see Appendix B). We used scores from a series of measures administered at the baseline assessment to construct the Personal Feedback Report. The SCID-I-RV (First et al., 2007) was used for diagnostic information, the SDS was used for problem severity, the TLFB was used for ecstasy use patterns, and the URICA was used to discuss how motivated participants’ were to reduce their use. In addition, the 23-item MDMA‐Beliefs Questionnaire (Businelle, Kendzor, Rash, Patterson, & Copeland, 2009) was used to discuss risk perception, the 50-item Inventory of Drug Taking Situations (Annis & Martin, 1985) was used for acknowledging high-risk situations, and the 8-item Drug Taking Confidence Questionnaire (Sklar & Turner, 1999) was used to discuss confidence for resisting ecstasy use in high-risk situations. A shortened version of the Important People Initial Interview (Longabaugh, Wirtz, Zywiak, & O'malley, 2010) was used to discuss options for social support for resisting ecstasy use. The 21-item Depression Anxiety Stress Scale (Lovibond & Lovibond, 1995) was used to discuss psychological distress, while the 10-item Acceptance and Action Questionnaire-II (Bond et al., 2011) was used to discuss willingness to experience emotional distress. Finally, the Valued Living Questionnaire (Wilson, Sandoz, Kitchens, & Roberts, 2010) was used to assess participants' commitment and action towards 12 life domains.

751

At the end of the session, therapists created change plans in collaboration with participants who reported interest in reducing their ecstasy use (which were few). Participants who remained uninterested in reducing their use were encouraged to monitor their ecstasy use to avoid any increases. All participants were provided with a self-monitoring diary to track their use, and were given the booklet and feedback form to take home. Therapists and Intervention Integrity Seven individuals provided the interventions. The first and second authors, doctoral-level clinical psychologists with over 10 years of experience delivering substance use treatment, conducted 7% (n = 13) of the sessions. Three recently registered psychologists provided 77% (n = 134) of the sessions. Lastly, two clinical psychology students in training provided 16% (n = 27) of the sessions. Assignment to therapist was based on availability. The first and second authors provided 2 days of training (14 hours) to all therapists and provided fortnightly clinical supervision throughout the project. The first author also trained two clinical psychology students to rate the E Check-up sessions using the YACS-II. Each rater read the YACS-II manual and rating system and practiced rating the items until they matched with the first author. Sixty-six (74%) of the E Check-up sessions were coded by at least one rater. The remaining sessions were either not taped, taped only in part, or inaudible. A reliability sample of 18 tapes evaluated by the two raters indicated an excellent level of interrater reliability for both the Adherence (ICC = .98) and Competence scales (ICC = .99).

sample size and data analytic strategy A power analysis using the G*Power 3 computer program (Faul, Erdfelder, Lang, & Buchner, 2007) indicated that a total sample of 140 people would be needed to detect small to medium between-group effects (ƒ = .23; Martin & Copeland, 2010) with 80% power and alpha set at .01 to control for multiple comparisons, Fcrit (138) = 6.82, λ = 11.99. Planning for 20% dropout, we enrolled 174 individuals. We used descriptive statistics to examine participants’ characteristics. In order to assess for differences at baseline between group assignments and between study sites, chi-square analyses were used for dichotomous variables, while t-tests were used for quantitative variables. We also examined estimated differences and their 95% CIs due to the unreliability of p-values (Cumming, 2013b). We followed this same procedure to assess for differences between those who completed all their follow-up assessments

752

norberg et al.

Table 2

Means and Confidence Intervals for Ecstasy Use Across Time No. of Pills

Days of Use

E Check-up

Education-only

E Check-up

Education-only

Month

M [95% CI]

M [95% CI]

M [95% CI]

M [95% CI]

Month − 3 Month − 2 Month − 1 Month 1 Month 2 Month 3 Month 4 Month 5 Month 6

4.29 4.97 3.96 2.75 1.25 1.68 2.34 1.69 1.79

4.66 5.51 4.88 3.52 1.79 2.40 2.21 2.58 2.39

2.10 [1.74, 2.46] 2.44 [1.95, 2.93] 2.08 [1.77, 2.39] 1.28 [1.00, 1.57] .59 [.36, .82] .83 [.53, 1.13] 1.18 [.84, 1.52] 1.03 [.53, 1.53] .99 [.68, 1.29]

2.25 [1.88, 2.62] 2.58 [2.11, 3.05] 2.29 [1.91, 2.68] 1.76 [1.29, 2.22] .79 [.51, 1.06] 1.08 [.76, 1.40] 1.01 [.62, 1.41] 1.18 [.71, 1.64] 1.18 [.76, 1.59]

[3.26, 5.33] [3.69, 6.25] [3.20, 4.71] [1.75, 3.75] [.71, 1.79] [.93, 2.44] [1.46, 3.23] [.74, 2.63] [1.01, 2.58]

[3.74, [4.17, [3.77, [2.41, [1.16, [1.64, [1.28, [1.52, [1.32,

5.59] 6.86] 5.90] 4.63] 2.42] 3.15] 3.14] 3.64] 3.46]

Note. No. of Pills = Total number of pills per month. TLFB data for months − 3, − 2, and −1 were collected at the baseline assessment. TLFB data for month 1 were collected at the 4-week follow-up, months 2, 3, 4, were collected at the 16-week follow-up, and months 5 and 6 were collected at the 24-week follow-up.

and those who did not complete at least one follow-up assessment. The education control group had a greater proportion of Australian-born and full-time employed individuals than the MET group (see Appendix C; Table C1). Participants recruited at the Brisbane site were younger on average and more likely to be Australian born (see Appendix C; Table C2). Participants who were follow-up noncompliant were younger, less educated, and more likely to be Australian born (see Appendix C; Table C3). Little’s MCAR test (Little, 1988) suggested that the missing follow-up data (132 data points out of 1566; 8.4%) was missing completely at random, χ 2 (373) = 352.535, p = .77. A generalized estimating equations (GEE) approach was used to test our primary hypothesis that the E Check-up group would have better outcomes at follow-up than the education group. All models assessed if participants changed differentially over time as a function of group (Time × Treatment interaction) and included all available data (N =

174). Poisson models with log link functions were used to examine quantity and frequency of ecstasy use data (count data) obtained using the TLFB. Normal models with identity link functions were used to examine URICA Action and SDS total scores, while a binomial logit function model was used to examine SDS scores dichotomized to indicate clinically significant ecstasy use (scores 0–3 vs. scores 4–15). Data from the TLFB were recorded month by month; thus, hypothesis testing using the TLFB data included nine assessment points (three prior to baseline and six after baseline). We examined all other primary outcomes at four time assessment points (baseline, 4-week follow-up, 16-week follow-up, and 24-week follow-up). Our secondary hypothesis, that the E Check-up group would report greater satisfaction with their assigned intervention at the 4-week follow-up than the education group, was tested using a between-groups ANOVA. In all hypothesis tests, baseline characteristics that differed across interventions, study sites, or follow-up compliant

Table 3

Means and Confidence Intervals for URICA and Severity of Dependence Variables Across Time

Variable by Condition

URICA Action score E Check-Up Education-Only SDS score E Check-Up Education-Only Proportion of SDS scores ≥ 4 E Check-Up Education-Only

Baseline

4-Week FU

16-Week FU

24-Week FU

M or P [95% CI]

M or P [95% CI]

M or P [95% CI]

M or P [95% CI]

2.75 [2.58, 2.93] 2.64 [2.46, 2.83]

3.12 [2.92, 3.32] 2.94 [2.73, 3.14]

3.12 [2.92, 3.32] 3.06 [2.83, 3.28]

2.99 [2.79, 3.19] 3.09 [2.87, 3.32]

2.46 [ 2.04, 2.89] 2.46 [2.02, 2.91]

2.10 [1.62, 2.59] 2.17 [1.71, 2.62]

2.00 [1.53, 2.47] 2.26 [1.69, 2.83]

1.95 [1.39, 2.52] 1.92 [1.37, 2.47]

.27 [.18, .39] .23 [.15, .34]

.25 [.16, .36] .23 [.15, .34]

.24 [.15, .36] .20 [.12, 32]

.34 [.25, .44] .25 [.17, .35]

Note. FU = Follow-up; URICA = University of Rhode Island Change Scale; SDS = Severity of Dependence Scale.

753

brief interventions to reduce ecstasy use Table 4

Generalized Estimating Equation Analyses for Ecstasy Outcomes by Group Assignment

No. of Pills Time Study Arm Time by Study Arm Days of Use Time Study Arm Time by Study Arm URICA Action score Time Study Arm Time by Study Arm SDS score Time Study Arm Time by Study Arm Proportion of SDS scores ≥ 4 Time Study Arm Time by Study Arm

Β

SE

95% CI

Z

p

Cohen’s d [95% CI]

-.13 .14 .02

.02 .15 .03

-.17, − .08 -.15, .43 -.04, .09

− 5.45 .96 .70

b .0001 .33 .70

.41 [.26, .56] .21 [− .09, .51] .15 [− .15, .45]

-.11 .09 .008

.02 .10 .03

-.16, − .07 -.11, .29 -.05, .07

− 5.45 .86 .25

b .0001 .39 .80

.41 [.26, .56] .18 [− .12, .48] .05 [ − .25, .35]

.04 -.08 .03

.02 .12 .02

.008, .07 -.32, .15 -.02, .08

2.41 -.70 1.23

.02 .48 .22

.18 [.03, .33] .15 [− .15, .45] .26 [− .09, .61]

-.07 -.02 .003

.04 .29 .06

-.15, .003 -.58, .54 -.11, .11

− 1.90 -.07 .06

.06 .94 .96

.14 [− .10, .29] .02 [− .32, .36] .01 [ − .33, .35]

.09 .44 -.03

.04 .34 .07

-.001, .17 -.22, 1.09 -.16, .10

1.94 1.31 -.44

.05 .19 .66

.15 [.00, .30] .28 [ − .07, .63] .09 [− .26, .43]

Note. URICA = University of Rhode Island Change Scale; SDS = Severity of Dependence Scale.

status, were included as covariates in order to obtain a more precise treatment effect. These variables were birth country, employment status, age, and years of education. Years of education statistically significantly contributed to the models for the quantity, frequency, and severity of ecstasy-related problems (all p's b . 009). Birth country statistically significantly contributed to the URICA action model (p = .04). Additionally, we calculated Cohen’s d using the

qffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2 n1 þ n2) and the pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi within-group t-test value (d ¼ t = n1 þ n2). Confidence intervals for d were calculated using the Exploratory Software for Confidence Intervals (Cumming, 2013a). We then used descriptive and inferential statistics to examine therapy process variables. We examined participants’ expectations for their assigned

between-group t-test value (d ¼ t

Table 5

Descriptive Statistics for Process Variables

Variable

E Check-up

Education-Onlya

M [95% CI]

M [95% CI]

Cohen’s d [95% CI] t

7.56 [7.15, 7.97] 3.90 [3.43, 4.38] 7.82 [7.39, 8.24] 34.68 [27.27, 42.09]

.09 [− .21, .39] .18 [− .12, .48] .19 [− .11, .49] .07 [.07, − .23]

-.86, p = .39 − 1.69, p = .09 1.73, p = .09 -.61, p = .55

— —

— —

— —

a

CEQ Credibility of the intervention 7.80 [7.44, 8.16] Predicted success of the intervention 4.45 [4.02, 4.88] Confidence in recommending the intervention 7.24 [6.72, 7.75] Predicted % reduction in ecstasy use 37.87 [30.52, 45.21] YACS-II b Adherence score 4.54 [4.42, 4.68] Competence score 5.03 [4.84, 5.23] Adherence score (intermediate only, n = 56) 4.53 [4.38, 4.67] Competence score (intermediate only, n = 56) 5.04 [4.81, 5.27] Adherence score (in-training only, n = 8) 4.75 [4.47, 5.03] Competence score (in-training only, n = 8) 5.08 [4.73, 5.43]

Note. CEQ = Credibility and Expectancy Questionnaire. YACS-II = Yale Adherence and Competence Scale. Intermediate only = recently registered psychologist. In-training only = clinical psychology students in training. Descriptive data not provided specifically for experienced therapists as only 2 of their tapes were rated, making CIs inaccurate; however, their ratings are factored into the combined scores. aSeven participants missing data. b66 tapes were coded. cTwenty-five participants missing data.

754

norberg et al.

Table 6

Generalized Estimating Equation Analyses for Motivational Interviewing Fidelity and Ecstasy Outcomes

No. of Pills Time YACS-II Adherence YACS-II Competence Days of Use Time YACS-II Adherence YACS-II Competence URICA Action score Time YACS-II Adherence YACS-II Competence SDS score Time YACS-II Adherence YACS-II Confidence Proportion of SDS scores ≥ 4 Time YACS-II Adherence YACS-II Confidence

Β

SE

-.16 -.006 -.08

.03 .20 .18

-.13 .05 -.05

95% CI

Z

p

Cohen’s d [95% CI]

.21, − .11 -.39, .38 -.44, .28

− 5.89 -.03 -.43

b .0001 .98 .66

.73 [.46, 1.00] .004 [− .24, .25] .05 [− .19, .29]

.02 .15 .16

-.17, − .09 -.25, .34 -.36, .26

− 5.67 .29 -.31

b .0001 .76 .75

.70 [.43, .97] .04 [− .20, .28] .04 [− .20, .28]

.05 .11 .16

.02 .25 .18

-.05, .03 -.38, .61 -.20, .52

2.57 .65 .86

.01 .65 .39

.32 [.07, .57] .06 [− .18, .30] .11 [− .13, .35]

-.05 .40 .18

.05 .60 .42

-.15, .04 -.79, 1.58 -.64, 1.00

− 1.11 .66 .44

.27 .51 .66

.14 [− .10, .38] .08 [− .16, .32] .05 [− .19, .29]

.06 -.27 -.09

.06 .74 .48

-.04, .16 − 1.72, 1.17 -.86, 1.04

1.12 -.37 .19

.26 .71 .85

.14 [− 10, .38] .05 [− .19, .29] .02 [− .22, .26]

Note. YACS-II = Second Edition of the Yale Adherence and Competence Scale; URICA = University of Rhode Island Change Scale; SDS = Severity of Dependence Scale.

interventions using the CEQ. Two independent sample t-tests assessed for potential statistical differences in credibility and expectations between the E Check-up and education groups. Additionally, we examined how well therapists adhered to MI during the E Check-up as measured by the Adherence and Competence scales of the YACS-II. The GEE approach described above was used to test the hypothesis that greater MI fidelity would be associated with better treatment outcomes.

Results ecstasy use outcomes Means and confidence intervals for ecstasy use outcomes are presented in Tables 2 and 3. Contrary to study hypotheses, effect sizes and p-values suggested there were no notable differences between groups on any primary outcome measure (see Table 4). Examination of effect sizes and p-values for time effects showed that both groups experienced statistically different and moderate reductions in their quantity and frequency of ecstasy of use from baseline (52% to 56% reductions on average), and small improvements in their problems (21% to 22% reductions on average) and commitment to change (9% to 17% increases on average). Additionally, slightly fewer participants endorsed a clinically significant ecstasy use problem at follow-up than at baseline, as indicated by SDS scores greater than four (5% to 10% fewer participants per condition; see Table 6).

client satisfaction As predicted, participants who received the E Check-up (M = 26.33, 95% CI = [25.42, 27.25]) reported being more satisfied with their assigned intervention than participants who received the education intervention (M = 24.45, 95% CI = [23.60, 25.31]; F(1, 140) = 8.66, p = .004, d = .50, 95% CI = [.17, .83]). process measures The E Check-up and education groups did not statistically differ in their ratings of credibility and expectancy for their assigned interventions. On average, participants thought the program interventions were logical and that they would recommend them to their friends. However, participants only felt the interventions would be modestly successful and that they would help them decrease their ecstasy use by one-third (see Table 5). Based on YACS-II ratings, therapists used MI principles to a moderate degree during the E Checkup, and in a manner slightly better than acceptable/ average. Examination of box-and-whisker plots showed that intermediate therapists had substantially more variability in their adherence and competence ratings than did experienced therapists or therapists in training. Hypothesis testing was not conducted on this data due to a violation of statistical assumptions (e.g., inequality of variances, unbalanced sample sizes, non-random assignment to

brief interventions to reduce ecstasy use therapist), but means and confidence intervals are reported in Table 5. Means and CIs are not reported for highly experienced therapists as only two MET sessions were scored for these therapists. CIs can be calculated for as few as two data points; however, they generate wider intervals and are not reflective of the true variability. We examined if YACS-II Adherence and Competence scores were related to ecstasy use outcomes in the 66 E Check-up sessions rated for MI fidelity. When controlling for birth country, employment status, age, and years of education, MI adherence and MI competence were not related to changes in ecstasy use outcomes over time (see Table 6).

Discussion The current study aimed to determine if a single session of MET delivered with a high level of treatment fidelity was more efficacious than an MI-informed educational control. Contrary to expectations, we did not find between-group differences on commitment to change, ecstasy use, or ecstasy-related problems at the 4-, 16-, or 24-week follow-up. Yet, in support of our secondary hypothesis, participants who received the E Check-up were more satisfied with their assigned intervention at the 4-week follow-up than were participants assigned to the education condition. Since group differences in satisfaction were small (2-point difference on a 24-point measure) and greater MI fidelity was not associated with better outcomes, the additional resources involved in providing the MET intervention may not be warranted. Our between-group null findings are consistent with the results of two previously published largescale studies comparing MET with an education control (Huang et al., 2011; Marsden et al., 2006), but our finding that MET and MI-informed educational control participants reduced their ecstasy-related problems at a similar rate over time differs from the smaller-scale published study. Given that fidelity to MI was not related to treatment outcomes, inadequate treatment delivery is unlikely to be responsible for the current results. It is more likely that the effect size estimate from the smaller study was not an accurate representation of the true effect size, given that three large-scale studies now suggest that MET does not enhance outcomes over the provision of self-administered or therapist-delivered education. Effect size estimates from small samples are inherently imprecise (Kraemer, Mintz, Noda, Tinklenberg, & Yesavage, 2006). This study has a number of strengths. First, only this study has monitored therapist fidelity to MI principles when delivering MET for ecstasy use, finding acceptable to good levels of treatment

755

fidelity ratings on the 74% of E Check-up sessions that were rated. While levels were not excellent, they were higher than scores reported in previously published substance abuse studies (Carroll et al., 2006; Davis et al., 2014). Second, the current sample’s baseline ecstasy use pattern was similar to that of regular ecstasy users sampled at the population-level in Australia (Australian Institute of Health and Welfare, 2011). Marsden et al. (2006) reported similar findings when assessing a sample that used about two-thirds as much ecstasy as the current sample; and thus, current findings may apply to both light and heavy ecstasy users. Third, only 5.7% to 16.7% of participants failed to complete all outcome measures at any specific follow-up point, and GEE were able to use all available data and account for within-subject variability. This study’s most important limitation is its sole comparison condition. As education was provided as part of both interventions and as we did not include a no-contact control or wait-list control, we are unable to determine whether education provision was the mechanism responsible for change. Although our design also precludes knowing to what extent MI was delivered in the educational condition, adherence to MI fidelity is unlikely to be responsible for the substantial gains made in that condition due to MI fidelity not being related to outcomes for MET participants. Participants in both interventions experienced moderate reductions in ecstasy use (50%) and slightly less impressive improvements in ecstasy-related problems (20%) and commitment to change ecstasy use (9%–17%). Some of this change may have been due to regression to the mean; however, we incorporated multiple baseline measurements for the TLFB which reduces measurement variability and therefore regression to the mean effects (Barnett, van der Pols, & Dobson, 2005). Accumulating research suggests that the assessments conducted as part of our research process may be responsible, at least to some extent, for our treatment outcomes (Kypri, Langley, Saunders, & Cashell-Smith, 2007; McCambridge et al., 2013). For example, McCambridge et al. found university students who completed 10 assessment items via email reported drinking less alcohol at a 3-month follow-up than no-contact control students (did not receive a baseline assessment or feedback), and that assessment-only students (received a baseline assessment, but no feedback) did not statistically differ on drinking outcomes compared to students who received both assessment and feedback. While research provides evidence for the positive influence of brief assessment on drinking levels, the associated benefits have been small (5%–17% reductions

756

norberg et al.

in drinking). Potentially, the current study’s comprehensive assessment may have had more powerful effects. Future research comparing varying types and lengths of assessment in comparison to education and no-contact control groups is required to determine the contributing effects of education and assessment. Inclusion of a no-contact control group in future studies also would help shed light on the current study’s discrepant findings between small improvements in motivation to change and large changes in actual use. These divergent findings suggest that natural recovery or drug unavailability, rather than conscious decisions, might be partially responsible for outcomes. Although drug availability data have not been collected, Australian population-based data demonstrates a decline in ecstasy use between 2007 (3.5%) and 2010 (3.0%) (Australian Institute of Health and Welfare, 2011). As we recruited continuously between 2010 and 2011, populationbased changes in ecstasy use may not be responsible for study results. In regards to natural recovery, there is cross-sectional evidence from an Australian study that almost all users who attempt to reduce their ecstasy use without formal assistance are successful (91% were successful, 41% tried; Topp et al., 1999). Parrott (2013) posited that serotonin neurotoxicity may be an important factor in natural recovery. Reductions in neurotransmitter functioning reduce the subjective pleasurable effects of ecstasy, which encourages the use of higher doses, but these higher doses are not able to instill the desired pleasurable effects due to an increasingly impaired neurotransmitter system. Increasing levels of psychobiological problems are experienced at these higher doses, leading the cost-benefit ratio for ecstasy use to deteriorate over time (Parrott, 2013). Three studies have provided preliminary evidence that the positive effects of ecstasy use decline over time, while its negative effects increase (Parrott, 2010; Verheyden, Henry, & Curran, 2003; Verheyden, Maidment, & Curran, 2003). While regular use of many psychoactive drugs leads to the development of tolerance, ecstasy differs in that tolerance appears to develop quite quickly, even after one dose (Parrott, 2005). Integrated psychobiological and psychosocial research is required to determine what mechanisms underlie recovery and how to accelerate them. Other noteworthy limitations of this study include its inability to assess for particular MI-associated techniques (e.g., use of a decisional balance or change plans) delivered in the E Check-up group or whether the extent of MI delivered between the MI-informed educational control and E Check-up groups differed. The YACS-II does not provide a checklist of MI-associated techniques. Inclusion of

content checklists in future studies would help to address this issue. While therapists were strongly encouraged to adhere to MI-principles when delivering the E Check-up, therapists also were encouraged not to be MI-inconsistent when delivering the educational control. Research has found that therapist engagement in MI-inconsistent behavior (e.g., confrontation, advising without permission) is related to poor therapeutic outcomes (Apodaca & Longabaugh, 2009; Gaume, Gmel, Faouzi, & Dappen, 2009). Thus, therapists were instructed to use basic clinical skills when delivering the educational information. These basic clinical skills (reflection, asking open-ended questions, avoiding argumentation) are also considered core motivational interviewing skills (Miller & Rollnick, 2002). Future studies should consider monitoring MI adherence in their control conditions, although as this study demonstrated, MI fidelity may not be a putative mechanism underlying change. As such, it may be more important to monitor the extent of MI-inconsistent behavior using a rating scale such as the Motivational Interviewing Skill Code (Miller, Moyers, Ernst, & Amrhein, 2008) or the Independent Tape Rater Scale (Martino, Ball, Nich, Frankforter, & Carroll, 2008). In summary, our results suggest that education about ecstasy use, provided in the context of comprehensive assessment and basic clinical skills, is as effective as a single session of MET for individuals who use ecstasy approximately twice a month. Fidelity to MI-consistent principles may not improve outcomes, at least for low-level users in the precontemplation and contemplation stage of change. As the MET group reported slightly higher levels of posttreatment satisfaction, the face validity of the MI-informed educational intervention may need to be increased. Future research is required to determine what mechanisms are responsible for change and how change can be enhanced. Although participants decreased their use by 50% by the 24-week follow-up, most were using ecstasy about once per month, a level that is associated with serotonergic neurotoxicity (de Win et al., 2008; Di lorio et al., 2012). With just slight additions to education provision, based on a better understanding of the causal mechanisms responsible for change, extremely impressive results may be achieved. Conflict of Interest Statement The authors declare that there are no conflicts of interest.

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.beth.2014.05. 006.

brief interventions to reduce ecstasy use References Annis, H. M., & Martin, G. (1985). Inventory of Drug-Taking Situations. Toronto: Addiction Research Foundation. Apodaca, T. R., & Longabaugh, R. (2009). Mechanisms of change in motivational interviewing: A review and preliminary evaluation of the evidence. Addiction, 104, 705–715. Attkisson, C. C., & Greenfield, T. K. (2004). The UCSF client satisfaction scales: I. The Client Satisfaction Questionnaire-8. In M. Maruish (Ed.), The use of psychological testing for treatment planning and outcomes assessment (3rd ed., pp. 799-811). Mahwah, NJ: Lawrence Erlbaum Associates. Australian Institute of Health and Welfare. (2011). 2010 National Drug Strategy Household Survey Report. Canberra: AIHW. Barnett, A. G., van der Pols, J. C., & Dobson, A. J. (2005). Regression to the mean: What it is and how to deal with it. International Journal of Epidemiology, 34, 215–220. Bond, F. W., Hayes, S. C., Baer, R. A., Carpenter, K. M., Guenole, N., Orcutt, H. K., … Zettle, R. D. (2011). Preliminary psychometric properties of the Acceptance and Action Questionnaire–II: A revised measure of psychological inflexibility and experiential avoidance. Behavior Therapy, 42, 676–688. http://dx.doi.org/10.1016/j.beth.2011.03.007 Bruno, R., Matthews, A. J., Topp, L., Degenhardt, L., Gomez, R., & Dunn, M. (2009). Can the Severity of Dependence Scale be usefully applied to ecstasy? Neuropsychobiology, 60, 137–147. http://dx.doi.org/10.1159/000253550 Businelle, M. S., Kendzor, D. E., Rash, C. J., Patterson, S., & Copeland, A. L. (2009). The development and validation of the methylenedioxymethamphetamine (MDMA) beliefs questionnaire (MDMA-BQ) in college students. Addiction Research and Theory, 17, 432–445. http://dx.doi.org/10. 1080/16066350801902442 Carlson, R. G., Falck, R. S., McCaughan, J. A., & Siegal, H. A. (2004). MDMA/Ecstasy use among young people in Ohio: Perceived risk and barriers to intervention. Journal of Psychoactive Drugs, 36, 181–189. http://dx.doi.org/10. 1080/02791072.2004.10399728 Carroll, K. M., Ball, S. A., Nich, C., Martino, S., Frankforter, T. L., Farentinos, C., … Woody, G. E. (2006). Motivational interviewing to improve treatment engagement and outcome in individuals seeking treatment for substance abuse: A multisite effectiveness study. Drug and Alcohol Dependence, 81, 301–312. Cumming, G. (2013a). ESCI (Exploratory Software for Confidence Intervals) - Software from The New Statistics: Estimation for Better Research. Retrieved on 14/FEB/2014 from http://www. latrobe.edu.au/psy/research/cognitive-and-developmentalpsychology/esci. Cumming, G. (2013b). The new statistics: Why and how. Psychological Science, 25, 7–29. http://dx.doi.org/10.1177/ 0956797613504966 Davis, K. E., Devitt, T., O'Neill, S. O., Kaiser, S. M., & Mueser, K. T. (2014). Targeting consumers in the early stages of substance use treatment: A pilot study. Psychiatric Rehabilitation Journal, 37, 37–42. de Win, M. M., Jager, G., Booij, J., Reneman, L., Schilt, T., Lavini, C., … van den Brink, W. (2008). Sustained effects of ecstasy on the human brain: A prospective neuroimaging study in novel users. Brain, 131, 2936–2945. http://dx.doi.org/ 10.1093/brain/awn255 Devilly, G. J., & Borkovec, T. D. (2000). Psychometric properties of the credibility/expectancy questionnaire. Journal of Behavior Therapy and Experimental Psychiatry, 31, 73–86. http://dx.doi.org/10.1016/s0005-7916(00)00012-4 Di lorio, C. R., Watkins, T. J., Dietrich, M. S., Cao, A., Blackford, J. U., Rogers, B., … Cowan, R. L. (2012). Evidence

757

for chronically altered serotonin function in the cerebral cortex of female 3, 4-methylenedioxymethamphetamine polydrug users. Archives of General Psychiatry, 69, 399–409. http://dx.doi.org/10.1001/archgenpsychiatry. 2011.156 Dickson, C., Bruno, R., & Brown, J. (2009). Investigating the role of serotonin in visual orientation processing using an 'ecstasy' (MDMA)-based research model. Neuropsychobiology, 60, 204–212. http://dx.doi.org/10.1159/000253556 DiClemente, C. C., Carbonari, J., Zweben, A., Morrel, T., & Lee, R. E. (2001). Motivation hypothesis casual chain analysis. In R. Longabaugh, & P. W. Wirtz (Eds.), Project MATCH: A priori matching hypotheses, results, and mediating mechanisms (Vol. 8, pp. 206-222). Rockville, MD: National Institute on Alcohol Abuse and Alcoholism. DiClemente, C. C., Schlundt, D., & Gemmell, L. (2004). Readiness and stages of change in addiction treatment. American Journal on Addictions, 13, 103–119. http://dx.doi.org/10. 1080/10550490490435777103 Fals-Stewart, W., O'Farrell, T. J., Freitas, T. T., McFarlin, S. K., & Rutigliano, P. (2000). The Timeline Followback reports of psychoactive substance use by drug-abusing patients: Psychometric properties. Journal of Consulting and Clinical Psychology, 68, 134–144. Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191. http://dx.doi.org/10. 3758/BF03193146 Field, C. A., Adinoff, B., Harris, T. R., Ball, S. A., & Carroll, K. M. (2009). Construct, concurrent and predictive validty of the URICA: Data from two multi-site clinical trials. Drug and Alcohol Dependence, 101, 115–123. http://dx.doi.org/ 10.1016/j.drugalcdep.2008.12.003 First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2007). Structured Clinical Interview for DSM-IV-TR Axis I Disorders-Patient Edition (SCID-I/P, 1/2007 revision). New York: Biometrics Research Department. Fox, H. C., Parrott, A. C., & Turner, J. J. D. (2001). Ecstasy use: Cognitive deficits related to dosage rather than self-reported problematic use of the drug. Journal of Psychopharmacology, 15, 273–281. http://dx.doi.org/ 10.1177/026988110101500406 Gaume, J., Gmel, G., Faouzi, M., & Dappen, J.-B. (2009). Counselor skill influences outcomes of brief motivational interventions. Journal of Substance Abuse Treatment, 37, 151–159. Gibbons, C. J., Nich, C., Steinberg, K., Roffman, R. A., Corvino, J., Babor, T., & Carroll, K. M. (2010). Treatment process, alliance and outcome in brief versus extended treatments for marijuana dependence. Addiction, 105, 1799–1808. Gossop, M., Best, D., Marsden, J., & Strang, J. (1997). Test-retest reliability of the Severity of Dependence Scale. Addiction, 92, 353. http://dx.doi.org/10.1111/j.13600443.1997.tb03205.x Gossop, M., Griffiths, P., Powis, B., & Strang, J. (1992). Severity of dependence and route of administration of heroin, cocaine, and amphetamines. British Journal of Addiction, 87, 527–1536. http://dx.doi.org/10.1111/j. 1360-0443.1992.tb02660.x Hanson, K. L., & Luciana, M. (2004). Neurocognitive function in users of MDMA: The importance of clinically significant patterns of use. Psychological Medicine, 34, 229–246. http://dx.doi.org/10.1017/S0033291703001132 Huang, Y., Tang, T., Lin, C., & Yen, C. (2011). Effects of motivational enhancement therapy on readiness to change

758

norberg et al.

MDMA and methamphetamine use behaviors in Taiwanese adolescents. Substance Use and Misuse, 46, 411–416. Kraemer, H. C., Mintz, J., Noda, A., Tinklenberg, J., & Yesavage, J. A. (2006). Caution regarding the use of pilot studies to guide power calculations for study proposals. Archives of General Psychiatry, 63, 484–489. http://dx.doi.org/10. 1001/archpsyc.63.5.484 Kypri, K., Langley, J. D., Saunders, J. B., & Cashell-Smith, M. L. (2007). Assessment may conceal therapeutic benefit: Findings from a randomized controlled trial for hazardous drinking. Addiction, 102, 62–70. http://dx.doi.org/10.1111/j. 1360-0443.2006.01632.x Little, R. J. A. (1988). A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Society, 83, 1198–1202. Longabaugh, R., Wirtz, P. W., Zywiak, W. H., & O'malley, S. S. (2010). Network support as a prognostic indicator of drinking outcomes: The COMBINE study. Journal of Studies on Alcohol and Drugs, 71, 837. Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the depression anxiety stress scales. (2nd ed.). Sydney: Psychology Foundation. Lundahl, B. W., Kunz, C., Brownell, C., Tollefson, D., & Burke, B. L. (2010). A meta-analyis of motivational interviewing: Twenty-five years of empirical studies. Research on Social Work Practice, 20, 137–160. Marsden, J., Stillwell, G., Barlow, H., Boys, A., Taylor, C., Hunt, N., & Farrell, M. (2006). An evaluation of a brief motivational intervention among young ecstasy and cocaine users: No effect on substance and alcohol use outcomes. Addiction, 101, 1014–1026. Martin, G., & Copeland, J. (2010). Brief intervention for regular ecstasy (MDMA) users: Pilot randomized trial of a Check-up model. Journal of Substance Use, 15, 131–142. Martino, S., Ball, S. A., Nich, C., Frankforter, T. L., & Carroll, K. M. (2008). Community program therapist adherence and competence in motivational enhancement therapy. Drug and Alcohol Dependence, 96, 37–48. McCambridge, J., Bendtsen, M., Karlsson, N., White, I. R., Nilsen, P., & Bendtsen, P. (2013). Alcohol assessment and feedback by email for university students: Main findings from a randomised controlled trial. BJPsych, 203, 334–340. http://dx.doi.org/10.1192/bjp.bp.113.128660 McCann, U. D., Sgambati, F. P., Schwartz, A. R., & Ricaurte, G. A. (2009). Sleep apnea in young abstinent recreational MDMA (ecstasy) consumers. Neurology, 73, 2011–2017. http://dx.doi.org/10.1212/WNL.0b013e3181c51a62 McConnaughy, E. A., Prochaska, J. O., & Velicer, W. F. (1983). Stages of change in psychotherapy: Measurement and sample profiles. Psychotherapy: Theory, Research and Practice, 20, 368–375. McKetin, R., Copeland, J., Norberg, M. M., Bruno, R., Hides, L., & Khawar, L. (2014). The effect of the ecstasy 'come-down' on the diagnosis of ecstasy dependence. Drug and Alcohol Dependence, 139, 26–32. McNally, A. M., Palfai, T. P., & Kahler, C. W. (2005). Motivational interventions for heavy drinking college students: Examining the role of discrepancy-related psychological processes. Psychology of Addictive Behaviors, 19, 79–87. Miller, W. R. (2001). When is it motivational interviewing? Addiction, 96, 1770–1771. Miller, W. R., Moyers, T. B., Ernst, D., & Amrhein, P. (2008). Manual for the Motivational Interviewing Skill Code (MISC). Center on Alcoholism, Substance Abuse, and Addictions. : The University of New Mexico. Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Preparing people for change (2nd ed.). New York: Guilford.

Nuro, K. F., Maccarelli, L., Ball, S. A., Martino, S., Baker, S. M., Rounsaville, B. J., & Carroll, K. M. (2005). Yale Adherence and Competence Scale (YACSII) guidelines (2nd ed.). West Haven, CT: Yale University Psychotherapy Development Center. Parrott, A. C. (2005). Chronic tolerance to recreational MDMA (3, 4-methylenedioxymethamphetamine) or ecstasy. Journal of Psychopharmacology, 19, 75–87. Parrott, A. C. (2009). Cortisol and MDMA (3,4-methylenedioxymethamphetamine): Neurohormonal aspects of bioenergetic-stress in Ecstasy users. Neuropsychobiology, 60, 148–158. Parrott, A. C. (2010). Conscious awareness versus optimistic beliefs in recreational Ecstasy/MDMA users. In E. Perry, D. Collerton, F. LeBeau, & H. Ashton (Eds.), New horizons in the neuroscience of consciousness. Amsterdam: John Benjamins Publishing Company. Parrott, A. C. (2013). MDMA, serotongeric neurotoxicity, and the diverse functional deficits of recreational “Ecstasy” users. Neuroscience and Biobehavioral Reviews, 37, 1466–1484. http://dx.doi.org/10.1016/j.neubiorev. 2013.04.016 Project MATCH Research Group. (1997). Matching alcoholishm treatment to client heterogeneity: Project MATCH posttreatment drinking outcomes. Journal of Studies on Alcohol, 58, 7–29. Rendell, P. G., Gray, T. J., Henry, J. D., & Tolan, A. (2007). Prospective memory impairment in ecstasy (MDMA) users. Psychopharmacology, 194, 497–504. http://dx.doi.org/ 10.1007/s00213-007-0859-z Rogers, G., Elston, J., Garside, R., Roome, C., Taylor, R., Younger, P., … Somerville, M. (2009). The harmful health effects of recreational ecstasy: A systematic review of observational evidence. Health Technology Assessment, 13, 1–315. http://dx.doi.org/10.3310/hta13060 Silins, E., Bleeker, A., & Martin, G. (2008). Ecstasy: Facts and Fiction (2nd ed.). Syndey: National Drug and Alcohol Research Centre. Sindicich, N., & Burns, L. (2013). Australian trends in ecstasy and related drug markets 2012. Findings from the Ecstasy and Related Drugs Reporting System (EDRS). Sydney: National Drug and Alcohol Research Centre, University of New South Wales. Sklar, S. M., & Turner, N. E. (1999). A brief measure for the assessment of coping self-efficacy among alcohol and other drug users. Addiction, 94, 723–729. http://dx.doi.org/ 10.1046/j.1360-0443.1999.94572310.x Sobell, L. C., Brown, J., Leo, G. I., & Sobell, L. C. (1996). The reliability of the Alcohol Timeline Followback when adminsitered by telephone and by computer. Drug and Alcohol Dependence, 42, 49–54. Sobell, L. C., & Sobell, M. C. (1996). Timeline Followback: A calendar method for assessing alcohol and drug use. Toronto, Ontario: Addiction Research Foundation. Strang, J., & McCambridge, J. (2004). Can the practitioner correctly predict outcome in motivational interviewing? Journal of Substance Abuse Treatment, 27, 83–88. Topp, L., Hando, J., Dillon, P., Roche, A., & Solowij, N. (1999). Ecstasy use in Australia: Patterns of use and associated harm. Drug and Alcohol Dependence, 55, 105–115. United Nations Office on Drugs and Crime. (2013). World Drug Report 2013 Retrieved from http://www.unodc.org/ unodc/secured/wdr/wdr2013/World_Drug_Report_2013. pdf on 14/FEB/2014. Verheyden, S. L., Henry, J. A., & Curran, G. (2003). Acute, sub-acute and long-term subjective consequences of “ecstasy” (MDMA) consumption in 430 regular users. Human

brief interventions to reduce ecstasy use Psychopharmacology, 18, 507–517. http://dx.doi.org/10. 1002/hup.529 Verheyden, S. L., Maidment, R., & Curran, H. V. (2003). Quitting ecstasy: An investigation of why people stop taking the drug and their subsequent mental health. Journal of Psychopharmacology, 17, 371–378. http://dx.doi.org/ 10.1177/0269881103174014 Wilson, K. G., Sandoz, E. K., Kitchens, J., & Roberts, M. (2010). The Valued Living Questionnaire: Defining and measuring valued action within a behavioral framework. The Psychological Record, 60, 249–272.

759

Zanarini, M. C., Skodol, A. E., Bender, D., Dolan, R., Sanislow, C., Schaefer, E., … Gunderson, J. G. (2000). The Collaborative Longitudinal Personality Disorders Study: Reliability of axis I and II diagnoses. Journal of Personality Disorders, 14, 291–299. http://dx.doi.org/10.1521/pedi.2000. 14.4.291

R E C E I V E D : February 14, 2014 A C C E P T E D : May 30, 2014 Available online 10 July 2014