Addictive Behaviors 58 (2016) 68–73
Contents lists available at ScienceDirect
Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh
The impact of motives-related feedback on drinking to cope among college students Claire E. Blevins ⁎, Robert S. Stephens Virginia Tech, Williams Hall (0436), Blacksburg, VA 24061, United States
H I G H L I G H T S • Outcomes from a brief coping motive feedback intervention are examined. • Participants were randomized to either receive motives-related or standard feedback. • The motives-based intervention showed promising treatment outcome results.
a r t i c l e
i n f o
Article history: Received 26 May 2015 Received in revised form 25 September 2015 Accepted 14 February 2016 Available online 16 February 2016 Keywords: Motives Alcohol use Brief interventions Coping Personalized feedback
a b s t r a c t Motives for alcohol use are associated with distinct antecedents and consequences. Drinking alcohol to cope with negative affect is consistently associated with the most problematic patterns of use. Interventions targeting drinking to cope are needed. This randomized controlled treatment trial is an initial attempt to evaluate the impact of a brief coping motive-specific personalized feedback intervention on motives and problematic outcomes associated with drinking. The study randomized 170 participants to receive either a brief Standard Feedback Condition (SFC; n = 83) or a Motives Feedback Condition (MFC; n = 87) that added education and feedback on drinking to cope as well as alternative coping strategies. Significant reductions in drinking to cope with anxiety and with depression were greater in the MFC at the 2-month follow-up. Significant reductions in drinking and negative consequences were observed but did not differ significantly by condition. Indirect tests showed that the MFC, relative to SFC, was associated with outcomes of drinking and negative consequences through change in drinking to cope with depression. Moderation analyses revealed that there were no differential outcomes according to baseline level of coping. This study is a promising new direction in motives research, providing support for brief personalized feedback interventions incorporating motives-related feedback. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Motives for using alcohol are psychological reasons for drinking that may provide insight into characteristics of drinkers and the consequences they experience (Kuntsche, Knibbe, Gmel, & Engels, 2005, 2006). The most commonly measured drinking motives include coping with negative affect, conformity with others, enhancement of positive affect, and social experience (Cooper, 1994). Motives for alcohol use are associated with distinct antecedents and consequences of use in a variety of populations (e.g., Cooper, Russell, Skinner, & Windle, 1992; Cooper, 1994; Kuntsche et al., 2005, 2006). Using alcohol to cope with negative affect has received the most empirical attention and has been associated with the most negative consequences (Ham & Hope, 2003; Kuntsche et al., 2005; Merrill, Wardell, & Read, 2014). The current study is an initial attempt to examine whether a brief personalized ⁎ Corresponding author. E-mail address: [email protected] (C.E. Blevins).
http://dx.doi.org/10.1016/j.addbeh.2016.02.024 0306-4603/© 2016 Elsevier Ltd. All rights reserved.
feedback intervention (PFI) targeting drinkers who use to cope augments the impact of existing interventions. Preliminary evidence indicates that targeting motives may be useful in reducing problematic use and highlights the importance of motives (Banes, Stephens, Blevins, Walker, & Roffman, 2014; Conrod, Castellanos-Ryan, & Mackie, 2011; Conrod, Stewart, Comeau, & Maclean, 2006; Conrod et al., 2000; LaBrie et al., 2008). Conrod and colleagues showed that a PFI matched to participant personality profiles reduced drinking motives consistent with the personality-targets, as well as the negative consequences associated with alcohol use (Conrod et al., 2000; Conrod et al., 2006; Conrod et al., 2011). For example, the intervention designed for individuals with anxiety sensitivity personality profile reduced levels of anxiety sensitivity, coping motives, and negative outcomes over time. Results from a brief PFI for female drinkers indicated that the intervention, which incorporated a discussion on motives, was associated with significant reductions in use and consequences as compared to the intervention without motives information (Labrie et al., 2008). A study focused on adult marijuana users
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
found a change in motives for marijuana use following participation in a motivational enhancement and cognitive-behavioral treatment (Banes et al., 2014). They reported associations between reductions in coping motives and reductions in negative outcomes of marijuana use, suggesting that motives for use may be a mechanism of change. Despite the work that has established that motives for use are associated with rates of use and consequences (c.f. Kuntsche et al., 2005, 2006), no known treatment study has tested a brief, individualized PFI that directly targets specific motives for use. The Conrod studies utilized a group format to introduce coping strategies specific to each personality profile (Conrod et al., 2000; Conrod et al., 2006; Conrod et al., 2011). LaBrie et al. (2008) utilized motives information in a brief format, but did not provide specific personalized feedback based on the motives that the individual specifically endorsed. The goal of the current study is to evaluate the effectiveness of a coping motive-specific PFI that provides information and feedback on drinking to cope. The specific hypotheses are: (1) the PFI incorporating coping motives-related feedback and information will reduce coping motives for alcohol use, quantity of alcohol consumed, and associated negative consequences compared to a standard PFI; (2) the intervention will be associated with reductions in quantity of alcohol use and consequences of use indirectly through drinking to cope; (3) coping motives measured at baseline will moderate the effect of the intervention such that those with higher levels of the coping motive will show the largest reductions in alcohol use and related problems in the coping motive intervention condition. 2. Methods 2.1. Participants Participants were undergraduates who were enrolled in psychology classes at a large southeastern university. In an effort to recruit individuals who drank regularly but not necessarily problematically, eligibility criteria included alcohol use on two or more days of a typical week. Additionally, participants were required to be at least 18 years of age. Eligibility criteria was explicitly stated in all recruitment materials. Of the 174 participants who attended a screening session, a total of 170 participants were eligible for and participated in the initial in-person intervention portion of the study. Analyses of demographic variables indicated that participants were primarily female (73.5%) and Caucasian (79.4%). The mean age of participants was 19.71 years (standard deviation = 1.42 years).
69
2.2. Design Participants were administered a baseline assessment battery and randomized to either the Motives Feedback Condition (MFC; n = 87) or the Standard Feedback Condition (SFC; n = 83). See Fig. 1 for details of participant flow through the study. Participants were re-assessed two months after baseline to determine changes in coping motives, drinking, and consequences. We anticipated effect sizes to be in the small to moderate range (d = .35–.50) based on previous studies of brief interventions (Conrod et al., 2011; LaBrie et al., 2008; Moyer, Finney, Swearingen, & Vergun, 2002; Vasilaki, Hosier, & Cox, 2006). We expected the effect of the MFC to be at the higher end of the range for measures of coping motives because the critical difference between conditions directly targeted the coping motives. Drinking and related consequences were expected to show smaller effects. Based on our sample size of 159 who completed assessments at both time points, the power to detect an effect of .50 was .88 when utilizing a mixed model GLM, with alpha set at .05 (G*Power 3; Faul, Erdfelder, Lang, & Buchner, 2007). Power for an effect size of .35 was .59. 2.3. Procedures Participants were recruited through an on-line system advertising studies available for extra credit in psychology classes. The study was described as an evaluation of different types of alcohol feedback. Procedures and measures were approved by the university Institutional Review Board. Baseline measures were obtained and the intervention was conducted in-person in a research office. Participants met with research staff to first determine eligibility. If participants were eligible and consented to participate, they completed a computerized assessment. Following the assessment, they were randomized into either the MFC or SFC condition. The PFI was delivered by a masters-level graduate student and trained undergraduate research assistants. Research assistants were observed and rated for adherence to protocol, and did not progress to administering the intervention to study subjects until they successfully administered the intervention, adhering to all aspects of the protocol, with two or more pilot participants. The intervention took approximately 15–25 min. Participants in both conditions received a personalized feedback report (PFR) that included normative information on nationwide alcohol consumption, normative information about alcohol consumption at the university, and a discussion of the individual's reported consequences associated with drinking alcohol. The PFR in the
Fig. 1. Participant flow through intervention and follow-up shows participant progress and attrition through the intervention and follow-up.
70
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
MFC condition differed only in the addition of discussion of motives for alcohol use, presentation of the participant's endorsement of coping motives in the baseline assessment, and information about research studies showing that endorsement of drinking to cope with negative affect was associated with more negative consequences relative to other reasons for drinking. They also received a list of alternative coping strategies for dealing with negative affect. The combination of coping motive-feedback, education, and alternative coping strategies was designed to draw attention to problematic drinking motives and helping develop adaptive coping strategies independent of alcohol use. As participants reviewed the PFR, research staff checked on their understanding of and interpretation of information and answered questions as needed. After reviewing the PFR, participants in both conditions completed a brief manipulation check to ensure that they received the intended information. Participants were compensated with extra credit in their psychology courses for participating in the baseline assessment and feedback intervention. Two months after completion of baseline measures and feedback interventions, participants in both conditions were contacted to complete a follow-up assessment on-line. Participants who completed the follow-up measure received additional extra credit in their psychology class and were entered to win one of twenty-three $25 gift cards. 2.4. Measures The Modified DMQ-R assessed drinking motives and included two scales assessing drinking to cope with anxiety and drinking to cope with depression (Grant, Stewart, O'Connor, Blackwell, & Conrod, 2007). Participants were asked to rate how frequently they used alcohol for the each reason specified on a scale of 1 (almost never/never) to 5 (almost always/always). For the purposes of the following analyses, the coping with anxiety and coping with depression subscales of the Modified DMQ-R were used to evaluate the effectiveness of the motives-related feedback. To determine whether the two separate coping scales should be combined for analyses, a principle components analyses was performed. Results suggested a two-factor solution. Cronbach's alphas were as follows: coping with anxiety .71 at baseline and .76 at follow-up; coping with depression was from .92 at baseline and .95 at follow-up. To assess use of alcohol, individuals were asked about their quantity and frequency of alcohol use during the past 30 days with the Daily Drinking Questionnaire-Revised (DDQ-R; Kruse, Fromme, & Corbin, 2005). The DDQ-R asks participants to indicate the number of drinks consumed on each day of the week on a typical week over the last month that they consume alcohol and the hours spent drinking. The number of drinks per week on a typical week was utilized as the measure of alcohol use. The assessment was given at baseline and follow-up. The Rutgers Alcohol Problem Index (RAPI) is a 23-item measure designed to assess a variety of consequences associated with drinking (White & LaBouvie, 1989). The RAPI was administered at baseline and follow-up sessions with an assessment window of one month. Cronbach's alpha reliability coefficients were .77 at baseline and .85 at follow-up, indicating good reliability. Participants in both conditions were asked five questions regarding the information they received in their PFRs to assess whether conditions differed as intended. Two of the questions targeted information on the coping motive and negative consequences associated with the coping motive that was specific to the MFC condition (e.g., “What motive for alcohol use was discussed as being associated with many negative outcomes, such as a higher rate of problems associated with alcohol use?”) while the other three questions were about information present in the PFRs of both conditions, including information on normative rates of alcohol use and consequences of alcohol use (e.g., “True or false: about 95% of college students nationwide drink alcohol”).
Five items inserted into the baseline and follow-up surveys were designed to test for careless reporting styles (Meade & Craig, 2012). These items were dispersed throughout the survey and directed participants to either type in a particular word or select a particular response (e.g., “Please select option 2” or “Please type END”). 2.5. Analysis plans Repeated measures General Linear Model (GLM) analyses with condition as between-subjects factor and time (baseline vs. follow-up) as the within-subjects factor, were used to test the effects of the intervention. GLM analyses were performed on the coping motives, quantity of alcohol use, and alcohol-related problems by condition. A time by condition interaction effect was predicted and post hoc comparisons of means at the follow-up were expected to show lower rates of the coping motives, use, and problems in the MFC condition. Exploratory analyses then tested differences in treatment outcomes by gender and by heavy use pattern. Interactions of time by gender/heavy use and time by condition by gender/heavy use were tested through GLM analyses. To test whether the intervention had an indirect effect on quantity of alcohol consumed or negative consequences through reductions in coping motives, Hayes and Preacher's Process SPSS macro (Preacher & Hayes, 2008) was used. This approach allows for bootstrapping of conditional indirect effects and provides higher power to detect effects. The indirect effect test was estimated through 5000 repeated samples (bootstrapping as recommended by Hayes, 2013) and is presented with 95% confidence intervals. Four separate bootstrapping analyses were performed based on two motive scores and the two alcohol variables. All analyses controlled for corresponding level of alcohol use or problems and corresponding coping motive at baseline in order to examine change in the both the coping motive and the alcohol outcomes. In order to test moderation hypotheses, the centered coping variable and the condition variable (dummy coded) were entered along with the baseline levels of the criterion (either use or problems) in order to predict change. In the second step, an interaction term was entered to test the hypothesis that level of coping motives at baseline moderates the effect of the intervention. These interaction terms were created as a product of the centered, coping and treatment variables and were expected to be significant predictors in support of moderation effects. 3. Results 3.1. Preliminary analyses Examination of the distributions of measures indicated adequate variation at both time points and no evidence of floor or ceiling effects. Data were examined for normality via Shapiro–Wilk testing and exhibited positive skew (p b .01). Thus, variables were log-transformed for GLM and indirect effects analyses. No participants answered more than one of the five careless reporting items incorrectly and, thus, no participants were excluded from analyses based on careless reporting. Comparisons of variables of interest by condition revealed no significant differences, suggesting that randomization to condition was successful. Out of the 170 participants who completed baseline assessment, 159 participants completed the on-line follow-up survey (93.5%). Data from one participant who failed to complete the entire follow-up assessment was dropped from analyses that utilized the problems scale only. Chi-square analyses revealed attrition was not significantly different by condition (p = .82). Two-way attrition by condition ANOVAs performed on baseline characteristics of the sample showed a main effect of attrition status at follow-up on quantity of alcohol consumed at baseline (p = .03) such that those who did not complete follow-up were more likely to have higher rates of use at baseline, but no other
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
main effects or significant interactions between baseline characteristics and condition. 3.2. Manipulation check Responses to the three questions regarding information in the PFRs common to both conditions showed high rates of accuracy (i.e., greater than 86%) and no significant differences between conditions (all ps N .18). Two questions specific to the MFC condition were answered differentially: 94.3% of individuals in the MFC condition were able to correctly identify the coping motive as being associated with the most negative consequences (as opposed to 15.7% in the SFC condition; p b .01) and 97.7% could identify a specific negative consequences of the coping motive (as opposed to 30.1% in the SFC condition; p b .01). These findings indicate that participants received the intended information and that the coping motives information was received uniquely in the MFC condition. 3.3. Descriptive statistics Table 1 contains means and standard deviations of variables utilized in analyses at baseline and follow-up. Notably, participants averaged 13.5 drinks per week at baseline (SD = 8.8), suggesting that participants with a variety of drinking habits were recruited. Participants on average scored a 6.7 on the RAPI (SD = 5.22). Coping with anxiety was endorsed somewhat more frequently than coping with depression (mean of 2.12 versus 1.51 out of 5). 3.4. Changes in coping motives, use, and problems Analyses revealed significant effects of time for the coping with anxiety (F(1, 157) = 9.61; p b .01) and coping with depression (F(1, 157) = 12.36; p = .01) motives (Table 1). Significant time by condition effects were found for the coping with anxiety (F(1, 157) = 4.63; p = .03) and the coping with depression motives (F(1, 157) = 13.47; p b .01),. Comparison of means at follow-up, controlling for baseline levels of use, indicated that those in the MFC condition reduced rates of the coping with anxiety (p = .03) and coping with depression motives more than those in the SFC condition (p b .01). Analyses indicated a significant effect of time on quantity of alcohol use (F(1, 157) = 33.45; p b .01) and alcohol-related problems (F(1, 156) = 37.78; p b .01). However, there were no significant time by condition interactions for either quantity of use (F(1, 157) = 1.22; p = .27) or problems (F(1, 156) = 2.37; p = .13). Between-condition effect sizes were as follows: coping with anxiety (Cohen's d = .28), coping with depression (d = .49), quantity of use (d = .28), and problems (d = .29). No
significant differences in outcomes were observed by gender, and no significant gender by condition interactions were observed (ps N .05). Additional analyses examined whether level of use at baseline moderated the effect of the treatment. A total of 48 participants were classified as lighter users based on recommended cutoffs of less than 14 drinks per week for men and less than 7 drinks per week for women (International Center for Alcohol Policies, 2003) with the remaining participants classified as heavy users. Level of use did not moderate the effect of treatment on motives or drinking outcomes. However, there was a near-significant difference in the interaction of time by heavy use in predicting a more significant reduction in use (p = .09). 3.5. Indirect tests of change in coping motives Indirect effects results are displayed in Table 2. When testing the impact of change in coping with anxiety on the intervention and follow-up quantity of alcohol use per week, the intervention did not significantly predict change in coping with anxiety, and change in coping with anxiety did not significantly predict follow-up alcohol use per week. The bootstrapped indirect effect was not significant. For the equation testing change in anxiety on the intervention and follow-up rates of alcohol problems, condition did not significantly predict change in coping with anxiety, and change in coping with anxiety did not predict follow-up problems. The bootstrapped indirect effect was not significant. In testing coping with depression, the intervention significantly predicted change in coping with depression, and change in coping with depression significantly predicted quantity of alcohol use at the follow-up. The bootstrapped indirect effect of coping with depression on use at follow-up was significant. Lastly, the intervention predicted change in coping with depression and change in coping with depression significantly predicted problems at follow-up. The bootstrapped indirect effect was significant. 3.6. Moderation by baseline level of coping Results from moderation analyses revealed no statistically significant interaction terms of the coping motives by outcome in any equation (ps N .05; not shown). Thus, there was no evidence of moderation of treatment effects by baseline level of coping.
Table 2 Tests of indirect effects of motives. Variable
Table 1 Descriptive statistics. Baseline
Drinks per typical week Alcohol-related problems Coping with anxiety motive Coping with depression motive ⁎ p b .05. ⁎⁎ p b .01.
MFC SFC Total MFC SFC Total MFC SFC Total MFC SFC Total
Follow-up
Intervention effects Time
Mean
SD
Mean
SD
12.93 14.14 13.52 6.69 6.80 6.74 2.16 2.08 2.12 1.58 1.42 1.51
7.40 10.11 8.82 5.24 5.23 5.22 0.88 0.85 0.86 0.75 0.55 0.66
9.28 11.56 10.40 4.39 5.72 5.04 1.93 2.03 1.98 1.46 1.47 1.41
6.96 9.36 8.28 4.28 5.62 5.01 0.87 0.80 0.84 0.71 0.69 0.70
Time × Cond.
⁎⁎
⁎⁎
⁎⁎
⁎
⁎
⁎⁎
71
b
Intervention — coping with anxiety — use Intervention — change in coping with anxiety −.03 Change in coping with anxiety — quantity of .36 use Intervention — quantity of use indirect −.01 Intervention — coping with anxiety — problems Intervention — change in coping with anxiety −.04 Change in coping with anxiety — problems .37 Intervention — problems indirect −.01 Intervention — coping with depression — use Intervention — change in coping with −.05 depression Change in coping with depression — quantity .40 of use Intervention — quantity of use indirect −.02 Intervention — change in coping with depression Change in coping with depression — problems −.07 Intervention — problems indirect .47 Intervention—coping with depression — −.03 problems
SE
.02 .19
p-Value
.12 .06 −.04/.01
.01 .02 .21 .01
.06 .08
.02
b.01
.19
.03
−.05/.01
−.05/−.01
.01 .03 .22 .02
95% confidence interval
b.01 .04 −.08/−.01
72
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
4. Discussion and conclusions The present study showed that providing feedback on using alcohol to cope reduced reports of using alcohol to cope with anxious and depressed affect in comparison to a more traditional personalized feedback intervention that did not incorporate motives feedback. Significant reductions in quantity of alcohol use and alcohol-related problem were also evident at follow-up, but did not differ significantly as a function of the coping motives feedback. However, there were indirect effects of treatment condition on both quantity of use and alcohol problems via changes in drinking to cope with depression, but not via changes in drinking to cope with anxiety. There was no evidence of moderation by baseline level of coping. Overall, the findings are encouraging with regard to the potential efficacy of brief interventions specifically targeting drinking to cope and suggest avenues for future research. The different findings for drinking to cope with depression versus anxiety in indirect tests were not predicted. The between-condition effect sizes of coping with anxiety and coping with depression are in the range of those reported for other brief interventions when compared to no treatment (Moyer et al., 2002; Vasilaki et al., 2006). Studies that have analyzed the impact of incremental features of brief interventions (e.g., normative feedback; Neighbors, Lewis, Bergstrom, & Larimer, 2006) typically find smaller effects. Future research should examine these questions with a larger sample. Even relatively small effects may represent meaningful addition to brief PFIs, especially considering the low cost of adding this additional feedback to typical brief interventions. In general, the brevity of the intervention and ease of implementation argue for further study. Additionally, feedback was given more generally on coping motives rather than giving feedback on drinking to cope with anxiety versus drinking to cope with depression separately. Future research could study affect-specific, targeted feedback. The relatively smaller effect on coping with anxiety in effect size and indirect testing may be related to differences in the nature and prevalence of anxious affect. Coping with anxiety was endorsed at a higher rate than coping with depression in this study, which is consistent with previous studies utilizing the Modified DMQ-R (Grant et al., 2007; Willem, Bijttebier, Claes, & Uytterhaegen, 2012). Differences in effect sizes suggest that coping with anxiety — or anxiety itself — may be a more difficult factor to affect in interventions for alcohol use. A review of the anxiety and motives literature has established the links between anxiety, anxiety sensitivity, motives and use (Demartini & Carey, 2011). Indeed, anxiety often persists after alcohol treatment and can predict less impactful treatment outcomes (Kushner et al., 2005). Thus, anxiety may be a stable or pervasive factor and a stronger predictor of substance use and use-related consequences, at least in young adult populations. Endorsement of the coping motives in this sample was consistent with other undergraduate samples (Grant et al., 2007) but lower than clinical samples (Mezquita et al., 2011). Recruitment of heavierdrinking or treatment-seeking individuals may have yielded stronger effects of the intervention (e.g., DiClemente, Bellino, & Neavins, 1999). Our sample was primarily female, Caucasian, and participants were less than 20 years of age, not of legal age to consume alcohol. Although there were no differences in changes in coping motive or outcomes by gender and no interactions between gender and condition, future research should examine these research questions in more diverse populations. Although this study employed indirect effects testing, study design precluded true meditational analyses. Thus, future studies may employ an additional time point to assess the effects on motives before the effects on drinking in order to allow for true meditational analyses (Nock, 2007). A potential limitation of the study was the self-report nature of responses. However, self-report measures have been found to be valid indicators of use (Del Boca & Darkes, 2003). Additionally, recent research utilizing transdermal alcohol sensors has found that self-reported alcohol use is highly correlated to biomarkers of use (Leffingwell et al., 2013). Another possible limitation was the evidence
of differential attrition at follow-up with heavier drinkers being more likely to be lost to follow-up. However, overall retention was good and exploratory analyses did not find that heavier users were more likely to reduce coping motives. Additionally, it is possible that participants in the motives feedback condition answered motives questions in a socially-desirable way in the follow-up, since they had received information on the problematic nature of the coping motive. Future research may clarify the impact of social desirability through study of internetbased feedback. Lastly, there was no measure included of participant response to motives feedback. Future studies may include a satisfaction rating or questions on how this information might impact their drinking. Despite limitations of the sample and the relatively brief follow-up period, the results are promising in showing the potential for coping motive feedback to augment the impact of brief PFIs. Further study of interventions specifically targeting drinking to cope with negative affect appears warranted. Designs including longer follow-ups, higher-risk samples, and perhaps additional types of motive feedback are needed. Role of funding source Funding for this study was provided by a grant through the Virginia Tech Graduate Research Development Program (GRDP); the GRDP had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Contributors CB and RS conceived, designed, and implemented the trial. CB and RS conceived the aims of this paper. CB performed the statistical analyses and wrote the initial draft of the manuscript. All authors discussed the results and contributed to and have approved the final manuscript. Conflict of interest No conflict declared. Acknowledgments The authors would like to thank Thomas Ollendick, Ph.D., Jungmeen Kim-Spoon, Ph.D., George Clum, Ph.D., and Steven Lash, Ph.D., whom were instrumental in study design.
References Banes, K.E., Stephens, R.S., Blevins, C.E., Walker, D.D., & Roffman, R.A. (2014). Changing motives for use: Outcomes from a cognitive-behavioral intervention for marijuanadependent adults. Drug and Alcohol Dependdence, 139, 41–46. Conrod, P.J., Stewart, S.H., Comeau, N., & Maclean, A.M. (2006). Efficacy of cognitive-behavioral interventions targeting personality risk factors for youth alcohol misuse. Journal of Clinical Child & Adolescent Psychology, 35(4), 550–563. Conrod, P.J., Stewart, S.H., Pihl, R.O., Côté, S., Fontaine, V., & Dongier, M. (2000). Efficacy of brief coping skills interventions that match different personality profiles of female substance abusers. Psychology of Addictive Behaviors, 14(3), 231–242. Conrod, P.J., Castellanos-Ryan, N., & Mackie, C. (2011). Long-term effects of a personalitytargeted intervention to reduce alcohol use in adolescents. Journal of Consulting and Clinical Psychology, 79, 296–306. Cooper, M.L. (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6, 117–128. Cooper, M.L., Russell, M., Skinner, J.B., & Windle, M. (1992). Development and validation of a three-dimensional measure of drinking motives. Psychological Assessment, 4, 123–132. Del Boca, F.K., & Darkes, J. (2003). The validity of self-reports of alcohol consumption: state of the science and challenges for research. Addiction, 98, 1–12. Demartini, K.S., & Carey, K.B. (2011). The role of anxiety sensitivity and drinking motives in predicting alcohol use : A critical review. Clinical Psychology Review, 31, 169–177. DiClemente, C.C., Bellino, L.E., & Neavins, T.M. (1999). Motivation for change and alcoholism treatment. Alcohol Research & Health : The Journal of the National Institute on Alcohol Abuse and Alcoholism, 23(2), 86–92. Faul, F., Erdfelder, E., Lang, A., & Buchner, A. (2007). G* Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. Grant, V.V., Stewart, S.H., O'Connor, R.M., Blackwell, E., & Conrod, P.J. (2007). Psychometric evaluation of the five-factor modified drinking motives questionnaire—Revised in undergraduates. Addictive Behaviors, 32, 2611–2632. Ham, L.S., & Hope, D.A. (2003). College students and problematic drinking: A review of the literature. Clinical Psychology Review, 23(5), 719–759. Hayes, A.F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford Publications. International Center for Alcohol Policies (2003). International drinking guidelines. Kruse, M., Fromme, K., & Corbin, W. (2005). Improving the accuracy of self-report measures of drinking: Disaggregating quantity and frequency indices of alcohol consumption. Alcoholism, Clinical and Experimental Research, 29(5).
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73 Kuntsche, E., Knibbe, R., Gmel, G., & Engels, R. (2005). Why do young people drink? A review of drinking motives. Clinical Psychology Review, 25(7), 841–861. Kuntsche, E., Knibbe, R., Gmel, G., & Engels, R. (2006). Who drinks and why? A review of socio-demographic, personality, and contextual issues behind the drinking motives in young people. Addictive Behaviors, 31(10), 1844–1857. Kushner, M.G., Abrams, K., Thuras, P., Hanson, K.L., Brekke, M., & Sletten, S. (2005). Follow-up study of anxiety disorder and alcohol dependence in comorbid alcoholism treatment patients. Alcoholism: Clinical & Experimental Research, 29(8), 1432–1443. LaBrie, J.W., Huchting, K., Tawalbeh, S., Pedersen, E.R., Thompson, A.D., Shelesky, K., ... Neighbors, C. (2008). A randomized motivational enhancement prevention group reduces drinking and alcohol consequences in first-year college women. Psychology of Addictive Behaviors, 22, 149–155. Leffingwell, T.R., Cooney, N.J., Murphy, J.G., Luczak, S., Rosen, G., Dougherty, D.M., & Barnett, N.P. (2013). Continuous objective monitoring of alcohol use: Twenty-first century measurement using transdermal sensors. Alcoholism: Clinical and Experimental Research, 37, 16–22. Meade, A.W., & Craig, S.B. (2012). Identifying careless responses in survey data. Psychological Methods, 17, 437–455. Merrill, J.E., Wardell, J.D., & Read, J.P. (2014). Drinking motives in the prospective prediction of unique alcohol-related consequences in college students. Journal of Studies on Alcohol and Drugs, 75, 93–102.
73
Mezquita, L., Stewart, S.H., Ibáñez, M.I., Ruipérez, M.A., Villa, H., Moya, J., & Ortet, G. (2011). Drinking motives in clinical and general populations. European Addiction Research, 17(5), 250–261. Moyer, A., Finney, J., Swearingen, C., & Vergun, P. (2002). Brief interventions for alcohol problems: A meta-analytic review of controlled investigations in treatmen-seeking and non-treatment-seeking populations. Addiction, 97, 279–292. Neighbors, C., Lewis, M.A., Bergstrom, R.L., & Larimer, M.E. (2006). Being controlled by normative influences: Self-determination as a moderator of a normative feedback alcohol intervention. Health Psychology, 25, 571–579. Nock, M.K. (2007). Conceptual and design essentials for evaluating mechanisms of change. Alcoholism: Clinical and Experimental Research, 31, 4S–12S. Preacher, K.J., & Hayes, A.F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891. Vasilaki, E., Hosier, S., & Cox, W. (2006). The efficacy of motivational interviewing as a brief intervention for excessive drinking: a meta-analytic review. Alcohol and Alcoholism, 328-335. White, H.R., & LaBouvie, E.W. (1989). Towards the assessment of alcohol problem drinking. Journal of Studies on Alcohol, 50, 30–37. Willem, L., Bijttebier, P., Claes, L., & Uytterhaegen, A. (2012). Temperament and problematic alcohol use in adolescence: An examination of drinking motives as mediators. Journal of Psychopathology and Behavioral Assessment, 34(2), 282–292.
Contents lists available at ScienceDirect
Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh
The impact of motives-related feedback on drinking to cope among college students Claire E. Blevins ⁎, Robert S. Stephens Virginia Tech, Williams Hall (0436), Blacksburg, VA 24061, United States
H I G H L I G H T S • Outcomes from a brief coping motive feedback intervention are examined. • Participants were randomized to either receive motives-related or standard feedback. • The motives-based intervention showed promising treatment outcome results.
a r t i c l e
i n f o
Article history: Received 26 May 2015 Received in revised form 25 September 2015 Accepted 14 February 2016 Available online 16 February 2016 Keywords: Motives Alcohol use Brief interventions Coping Personalized feedback
a b s t r a c t Motives for alcohol use are associated with distinct antecedents and consequences. Drinking alcohol to cope with negative affect is consistently associated with the most problematic patterns of use. Interventions targeting drinking to cope are needed. This randomized controlled treatment trial is an initial attempt to evaluate the impact of a brief coping motive-specific personalized feedback intervention on motives and problematic outcomes associated with drinking. The study randomized 170 participants to receive either a brief Standard Feedback Condition (SFC; n = 83) or a Motives Feedback Condition (MFC; n = 87) that added education and feedback on drinking to cope as well as alternative coping strategies. Significant reductions in drinking to cope with anxiety and with depression were greater in the MFC at the 2-month follow-up. Significant reductions in drinking and negative consequences were observed but did not differ significantly by condition. Indirect tests showed that the MFC, relative to SFC, was associated with outcomes of drinking and negative consequences through change in drinking to cope with depression. Moderation analyses revealed that there were no differential outcomes according to baseline level of coping. This study is a promising new direction in motives research, providing support for brief personalized feedback interventions incorporating motives-related feedback. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Motives for using alcohol are psychological reasons for drinking that may provide insight into characteristics of drinkers and the consequences they experience (Kuntsche, Knibbe, Gmel, & Engels, 2005, 2006). The most commonly measured drinking motives include coping with negative affect, conformity with others, enhancement of positive affect, and social experience (Cooper, 1994). Motives for alcohol use are associated with distinct antecedents and consequences of use in a variety of populations (e.g., Cooper, Russell, Skinner, & Windle, 1992; Cooper, 1994; Kuntsche et al., 2005, 2006). Using alcohol to cope with negative affect has received the most empirical attention and has been associated with the most negative consequences (Ham & Hope, 2003; Kuntsche et al., 2005; Merrill, Wardell, & Read, 2014). The current study is an initial attempt to examine whether a brief personalized ⁎ Corresponding author. E-mail address: [email protected] (C.E. Blevins).
http://dx.doi.org/10.1016/j.addbeh.2016.02.024 0306-4603/© 2016 Elsevier Ltd. All rights reserved.
feedback intervention (PFI) targeting drinkers who use to cope augments the impact of existing interventions. Preliminary evidence indicates that targeting motives may be useful in reducing problematic use and highlights the importance of motives (Banes, Stephens, Blevins, Walker, & Roffman, 2014; Conrod, Castellanos-Ryan, & Mackie, 2011; Conrod, Stewart, Comeau, & Maclean, 2006; Conrod et al., 2000; LaBrie et al., 2008). Conrod and colleagues showed that a PFI matched to participant personality profiles reduced drinking motives consistent with the personality-targets, as well as the negative consequences associated with alcohol use (Conrod et al., 2000; Conrod et al., 2006; Conrod et al., 2011). For example, the intervention designed for individuals with anxiety sensitivity personality profile reduced levels of anxiety sensitivity, coping motives, and negative outcomes over time. Results from a brief PFI for female drinkers indicated that the intervention, which incorporated a discussion on motives, was associated with significant reductions in use and consequences as compared to the intervention without motives information (Labrie et al., 2008). A study focused on adult marijuana users
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
found a change in motives for marijuana use following participation in a motivational enhancement and cognitive-behavioral treatment (Banes et al., 2014). They reported associations between reductions in coping motives and reductions in negative outcomes of marijuana use, suggesting that motives for use may be a mechanism of change. Despite the work that has established that motives for use are associated with rates of use and consequences (c.f. Kuntsche et al., 2005, 2006), no known treatment study has tested a brief, individualized PFI that directly targets specific motives for use. The Conrod studies utilized a group format to introduce coping strategies specific to each personality profile (Conrod et al., 2000; Conrod et al., 2006; Conrod et al., 2011). LaBrie et al. (2008) utilized motives information in a brief format, but did not provide specific personalized feedback based on the motives that the individual specifically endorsed. The goal of the current study is to evaluate the effectiveness of a coping motive-specific PFI that provides information and feedback on drinking to cope. The specific hypotheses are: (1) the PFI incorporating coping motives-related feedback and information will reduce coping motives for alcohol use, quantity of alcohol consumed, and associated negative consequences compared to a standard PFI; (2) the intervention will be associated with reductions in quantity of alcohol use and consequences of use indirectly through drinking to cope; (3) coping motives measured at baseline will moderate the effect of the intervention such that those with higher levels of the coping motive will show the largest reductions in alcohol use and related problems in the coping motive intervention condition. 2. Methods 2.1. Participants Participants were undergraduates who were enrolled in psychology classes at a large southeastern university. In an effort to recruit individuals who drank regularly but not necessarily problematically, eligibility criteria included alcohol use on two or more days of a typical week. Additionally, participants were required to be at least 18 years of age. Eligibility criteria was explicitly stated in all recruitment materials. Of the 174 participants who attended a screening session, a total of 170 participants were eligible for and participated in the initial in-person intervention portion of the study. Analyses of demographic variables indicated that participants were primarily female (73.5%) and Caucasian (79.4%). The mean age of participants was 19.71 years (standard deviation = 1.42 years).
69
2.2. Design Participants were administered a baseline assessment battery and randomized to either the Motives Feedback Condition (MFC; n = 87) or the Standard Feedback Condition (SFC; n = 83). See Fig. 1 for details of participant flow through the study. Participants were re-assessed two months after baseline to determine changes in coping motives, drinking, and consequences. We anticipated effect sizes to be in the small to moderate range (d = .35–.50) based on previous studies of brief interventions (Conrod et al., 2011; LaBrie et al., 2008; Moyer, Finney, Swearingen, & Vergun, 2002; Vasilaki, Hosier, & Cox, 2006). We expected the effect of the MFC to be at the higher end of the range for measures of coping motives because the critical difference between conditions directly targeted the coping motives. Drinking and related consequences were expected to show smaller effects. Based on our sample size of 159 who completed assessments at both time points, the power to detect an effect of .50 was .88 when utilizing a mixed model GLM, with alpha set at .05 (G*Power 3; Faul, Erdfelder, Lang, & Buchner, 2007). Power for an effect size of .35 was .59. 2.3. Procedures Participants were recruited through an on-line system advertising studies available for extra credit in psychology classes. The study was described as an evaluation of different types of alcohol feedback. Procedures and measures were approved by the university Institutional Review Board. Baseline measures were obtained and the intervention was conducted in-person in a research office. Participants met with research staff to first determine eligibility. If participants were eligible and consented to participate, they completed a computerized assessment. Following the assessment, they were randomized into either the MFC or SFC condition. The PFI was delivered by a masters-level graduate student and trained undergraduate research assistants. Research assistants were observed and rated for adherence to protocol, and did not progress to administering the intervention to study subjects until they successfully administered the intervention, adhering to all aspects of the protocol, with two or more pilot participants. The intervention took approximately 15–25 min. Participants in both conditions received a personalized feedback report (PFR) that included normative information on nationwide alcohol consumption, normative information about alcohol consumption at the university, and a discussion of the individual's reported consequences associated with drinking alcohol. The PFR in the
Fig. 1. Participant flow through intervention and follow-up shows participant progress and attrition through the intervention and follow-up.
70
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
MFC condition differed only in the addition of discussion of motives for alcohol use, presentation of the participant's endorsement of coping motives in the baseline assessment, and information about research studies showing that endorsement of drinking to cope with negative affect was associated with more negative consequences relative to other reasons for drinking. They also received a list of alternative coping strategies for dealing with negative affect. The combination of coping motive-feedback, education, and alternative coping strategies was designed to draw attention to problematic drinking motives and helping develop adaptive coping strategies independent of alcohol use. As participants reviewed the PFR, research staff checked on their understanding of and interpretation of information and answered questions as needed. After reviewing the PFR, participants in both conditions completed a brief manipulation check to ensure that they received the intended information. Participants were compensated with extra credit in their psychology courses for participating in the baseline assessment and feedback intervention. Two months after completion of baseline measures and feedback interventions, participants in both conditions were contacted to complete a follow-up assessment on-line. Participants who completed the follow-up measure received additional extra credit in their psychology class and were entered to win one of twenty-three $25 gift cards. 2.4. Measures The Modified DMQ-R assessed drinking motives and included two scales assessing drinking to cope with anxiety and drinking to cope with depression (Grant, Stewart, O'Connor, Blackwell, & Conrod, 2007). Participants were asked to rate how frequently they used alcohol for the each reason specified on a scale of 1 (almost never/never) to 5 (almost always/always). For the purposes of the following analyses, the coping with anxiety and coping with depression subscales of the Modified DMQ-R were used to evaluate the effectiveness of the motives-related feedback. To determine whether the two separate coping scales should be combined for analyses, a principle components analyses was performed. Results suggested a two-factor solution. Cronbach's alphas were as follows: coping with anxiety .71 at baseline and .76 at follow-up; coping with depression was from .92 at baseline and .95 at follow-up. To assess use of alcohol, individuals were asked about their quantity and frequency of alcohol use during the past 30 days with the Daily Drinking Questionnaire-Revised (DDQ-R; Kruse, Fromme, & Corbin, 2005). The DDQ-R asks participants to indicate the number of drinks consumed on each day of the week on a typical week over the last month that they consume alcohol and the hours spent drinking. The number of drinks per week on a typical week was utilized as the measure of alcohol use. The assessment was given at baseline and follow-up. The Rutgers Alcohol Problem Index (RAPI) is a 23-item measure designed to assess a variety of consequences associated with drinking (White & LaBouvie, 1989). The RAPI was administered at baseline and follow-up sessions with an assessment window of one month. Cronbach's alpha reliability coefficients were .77 at baseline and .85 at follow-up, indicating good reliability. Participants in both conditions were asked five questions regarding the information they received in their PFRs to assess whether conditions differed as intended. Two of the questions targeted information on the coping motive and negative consequences associated with the coping motive that was specific to the MFC condition (e.g., “What motive for alcohol use was discussed as being associated with many negative outcomes, such as a higher rate of problems associated with alcohol use?”) while the other three questions were about information present in the PFRs of both conditions, including information on normative rates of alcohol use and consequences of alcohol use (e.g., “True or false: about 95% of college students nationwide drink alcohol”).
Five items inserted into the baseline and follow-up surveys were designed to test for careless reporting styles (Meade & Craig, 2012). These items were dispersed throughout the survey and directed participants to either type in a particular word or select a particular response (e.g., “Please select option 2” or “Please type END”). 2.5. Analysis plans Repeated measures General Linear Model (GLM) analyses with condition as between-subjects factor and time (baseline vs. follow-up) as the within-subjects factor, were used to test the effects of the intervention. GLM analyses were performed on the coping motives, quantity of alcohol use, and alcohol-related problems by condition. A time by condition interaction effect was predicted and post hoc comparisons of means at the follow-up were expected to show lower rates of the coping motives, use, and problems in the MFC condition. Exploratory analyses then tested differences in treatment outcomes by gender and by heavy use pattern. Interactions of time by gender/heavy use and time by condition by gender/heavy use were tested through GLM analyses. To test whether the intervention had an indirect effect on quantity of alcohol consumed or negative consequences through reductions in coping motives, Hayes and Preacher's Process SPSS macro (Preacher & Hayes, 2008) was used. This approach allows for bootstrapping of conditional indirect effects and provides higher power to detect effects. The indirect effect test was estimated through 5000 repeated samples (bootstrapping as recommended by Hayes, 2013) and is presented with 95% confidence intervals. Four separate bootstrapping analyses were performed based on two motive scores and the two alcohol variables. All analyses controlled for corresponding level of alcohol use or problems and corresponding coping motive at baseline in order to examine change in the both the coping motive and the alcohol outcomes. In order to test moderation hypotheses, the centered coping variable and the condition variable (dummy coded) were entered along with the baseline levels of the criterion (either use or problems) in order to predict change. In the second step, an interaction term was entered to test the hypothesis that level of coping motives at baseline moderates the effect of the intervention. These interaction terms were created as a product of the centered, coping and treatment variables and were expected to be significant predictors in support of moderation effects. 3. Results 3.1. Preliminary analyses Examination of the distributions of measures indicated adequate variation at both time points and no evidence of floor or ceiling effects. Data were examined for normality via Shapiro–Wilk testing and exhibited positive skew (p b .01). Thus, variables were log-transformed for GLM and indirect effects analyses. No participants answered more than one of the five careless reporting items incorrectly and, thus, no participants were excluded from analyses based on careless reporting. Comparisons of variables of interest by condition revealed no significant differences, suggesting that randomization to condition was successful. Out of the 170 participants who completed baseline assessment, 159 participants completed the on-line follow-up survey (93.5%). Data from one participant who failed to complete the entire follow-up assessment was dropped from analyses that utilized the problems scale only. Chi-square analyses revealed attrition was not significantly different by condition (p = .82). Two-way attrition by condition ANOVAs performed on baseline characteristics of the sample showed a main effect of attrition status at follow-up on quantity of alcohol consumed at baseline (p = .03) such that those who did not complete follow-up were more likely to have higher rates of use at baseline, but no other
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
main effects or significant interactions between baseline characteristics and condition. 3.2. Manipulation check Responses to the three questions regarding information in the PFRs common to both conditions showed high rates of accuracy (i.e., greater than 86%) and no significant differences between conditions (all ps N .18). Two questions specific to the MFC condition were answered differentially: 94.3% of individuals in the MFC condition were able to correctly identify the coping motive as being associated with the most negative consequences (as opposed to 15.7% in the SFC condition; p b .01) and 97.7% could identify a specific negative consequences of the coping motive (as opposed to 30.1% in the SFC condition; p b .01). These findings indicate that participants received the intended information and that the coping motives information was received uniquely in the MFC condition. 3.3. Descriptive statistics Table 1 contains means and standard deviations of variables utilized in analyses at baseline and follow-up. Notably, participants averaged 13.5 drinks per week at baseline (SD = 8.8), suggesting that participants with a variety of drinking habits were recruited. Participants on average scored a 6.7 on the RAPI (SD = 5.22). Coping with anxiety was endorsed somewhat more frequently than coping with depression (mean of 2.12 versus 1.51 out of 5). 3.4. Changes in coping motives, use, and problems Analyses revealed significant effects of time for the coping with anxiety (F(1, 157) = 9.61; p b .01) and coping with depression (F(1, 157) = 12.36; p = .01) motives (Table 1). Significant time by condition effects were found for the coping with anxiety (F(1, 157) = 4.63; p = .03) and the coping with depression motives (F(1, 157) = 13.47; p b .01),. Comparison of means at follow-up, controlling for baseline levels of use, indicated that those in the MFC condition reduced rates of the coping with anxiety (p = .03) and coping with depression motives more than those in the SFC condition (p b .01). Analyses indicated a significant effect of time on quantity of alcohol use (F(1, 157) = 33.45; p b .01) and alcohol-related problems (F(1, 156) = 37.78; p b .01). However, there were no significant time by condition interactions for either quantity of use (F(1, 157) = 1.22; p = .27) or problems (F(1, 156) = 2.37; p = .13). Between-condition effect sizes were as follows: coping with anxiety (Cohen's d = .28), coping with depression (d = .49), quantity of use (d = .28), and problems (d = .29). No
significant differences in outcomes were observed by gender, and no significant gender by condition interactions were observed (ps N .05). Additional analyses examined whether level of use at baseline moderated the effect of the treatment. A total of 48 participants were classified as lighter users based on recommended cutoffs of less than 14 drinks per week for men and less than 7 drinks per week for women (International Center for Alcohol Policies, 2003) with the remaining participants classified as heavy users. Level of use did not moderate the effect of treatment on motives or drinking outcomes. However, there was a near-significant difference in the interaction of time by heavy use in predicting a more significant reduction in use (p = .09). 3.5. Indirect tests of change in coping motives Indirect effects results are displayed in Table 2. When testing the impact of change in coping with anxiety on the intervention and follow-up quantity of alcohol use per week, the intervention did not significantly predict change in coping with anxiety, and change in coping with anxiety did not significantly predict follow-up alcohol use per week. The bootstrapped indirect effect was not significant. For the equation testing change in anxiety on the intervention and follow-up rates of alcohol problems, condition did not significantly predict change in coping with anxiety, and change in coping with anxiety did not predict follow-up problems. The bootstrapped indirect effect was not significant. In testing coping with depression, the intervention significantly predicted change in coping with depression, and change in coping with depression significantly predicted quantity of alcohol use at the follow-up. The bootstrapped indirect effect of coping with depression on use at follow-up was significant. Lastly, the intervention predicted change in coping with depression and change in coping with depression significantly predicted problems at follow-up. The bootstrapped indirect effect was significant. 3.6. Moderation by baseline level of coping Results from moderation analyses revealed no statistically significant interaction terms of the coping motives by outcome in any equation (ps N .05; not shown). Thus, there was no evidence of moderation of treatment effects by baseline level of coping.
Table 2 Tests of indirect effects of motives. Variable
Table 1 Descriptive statistics. Baseline
Drinks per typical week Alcohol-related problems Coping with anxiety motive Coping with depression motive ⁎ p b .05. ⁎⁎ p b .01.
MFC SFC Total MFC SFC Total MFC SFC Total MFC SFC Total
Follow-up
Intervention effects Time
Mean
SD
Mean
SD
12.93 14.14 13.52 6.69 6.80 6.74 2.16 2.08 2.12 1.58 1.42 1.51
7.40 10.11 8.82 5.24 5.23 5.22 0.88 0.85 0.86 0.75 0.55 0.66
9.28 11.56 10.40 4.39 5.72 5.04 1.93 2.03 1.98 1.46 1.47 1.41
6.96 9.36 8.28 4.28 5.62 5.01 0.87 0.80 0.84 0.71 0.69 0.70
Time × Cond.
⁎⁎
⁎⁎
⁎⁎
⁎
⁎
⁎⁎
71
b
Intervention — coping with anxiety — use Intervention — change in coping with anxiety −.03 Change in coping with anxiety — quantity of .36 use Intervention — quantity of use indirect −.01 Intervention — coping with anxiety — problems Intervention — change in coping with anxiety −.04 Change in coping with anxiety — problems .37 Intervention — problems indirect −.01 Intervention — coping with depression — use Intervention — change in coping with −.05 depression Change in coping with depression — quantity .40 of use Intervention — quantity of use indirect −.02 Intervention — change in coping with depression Change in coping with depression — problems −.07 Intervention — problems indirect .47 Intervention—coping with depression — −.03 problems
SE
.02 .19
p-Value
.12 .06 −.04/.01
.01 .02 .21 .01
.06 .08
.02
b.01
.19
.03
−.05/.01
−.05/−.01
.01 .03 .22 .02
95% confidence interval
b.01 .04 −.08/−.01
72
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73
4. Discussion and conclusions The present study showed that providing feedback on using alcohol to cope reduced reports of using alcohol to cope with anxious and depressed affect in comparison to a more traditional personalized feedback intervention that did not incorporate motives feedback. Significant reductions in quantity of alcohol use and alcohol-related problem were also evident at follow-up, but did not differ significantly as a function of the coping motives feedback. However, there were indirect effects of treatment condition on both quantity of use and alcohol problems via changes in drinking to cope with depression, but not via changes in drinking to cope with anxiety. There was no evidence of moderation by baseline level of coping. Overall, the findings are encouraging with regard to the potential efficacy of brief interventions specifically targeting drinking to cope and suggest avenues for future research. The different findings for drinking to cope with depression versus anxiety in indirect tests were not predicted. The between-condition effect sizes of coping with anxiety and coping with depression are in the range of those reported for other brief interventions when compared to no treatment (Moyer et al., 2002; Vasilaki et al., 2006). Studies that have analyzed the impact of incremental features of brief interventions (e.g., normative feedback; Neighbors, Lewis, Bergstrom, & Larimer, 2006) typically find smaller effects. Future research should examine these questions with a larger sample. Even relatively small effects may represent meaningful addition to brief PFIs, especially considering the low cost of adding this additional feedback to typical brief interventions. In general, the brevity of the intervention and ease of implementation argue for further study. Additionally, feedback was given more generally on coping motives rather than giving feedback on drinking to cope with anxiety versus drinking to cope with depression separately. Future research could study affect-specific, targeted feedback. The relatively smaller effect on coping with anxiety in effect size and indirect testing may be related to differences in the nature and prevalence of anxious affect. Coping with anxiety was endorsed at a higher rate than coping with depression in this study, which is consistent with previous studies utilizing the Modified DMQ-R (Grant et al., 2007; Willem, Bijttebier, Claes, & Uytterhaegen, 2012). Differences in effect sizes suggest that coping with anxiety — or anxiety itself — may be a more difficult factor to affect in interventions for alcohol use. A review of the anxiety and motives literature has established the links between anxiety, anxiety sensitivity, motives and use (Demartini & Carey, 2011). Indeed, anxiety often persists after alcohol treatment and can predict less impactful treatment outcomes (Kushner et al., 2005). Thus, anxiety may be a stable or pervasive factor and a stronger predictor of substance use and use-related consequences, at least in young adult populations. Endorsement of the coping motives in this sample was consistent with other undergraduate samples (Grant et al., 2007) but lower than clinical samples (Mezquita et al., 2011). Recruitment of heavierdrinking or treatment-seeking individuals may have yielded stronger effects of the intervention (e.g., DiClemente, Bellino, & Neavins, 1999). Our sample was primarily female, Caucasian, and participants were less than 20 years of age, not of legal age to consume alcohol. Although there were no differences in changes in coping motive or outcomes by gender and no interactions between gender and condition, future research should examine these research questions in more diverse populations. Although this study employed indirect effects testing, study design precluded true meditational analyses. Thus, future studies may employ an additional time point to assess the effects on motives before the effects on drinking in order to allow for true meditational analyses (Nock, 2007). A potential limitation of the study was the self-report nature of responses. However, self-report measures have been found to be valid indicators of use (Del Boca & Darkes, 2003). Additionally, recent research utilizing transdermal alcohol sensors has found that self-reported alcohol use is highly correlated to biomarkers of use (Leffingwell et al., 2013). Another possible limitation was the evidence
of differential attrition at follow-up with heavier drinkers being more likely to be lost to follow-up. However, overall retention was good and exploratory analyses did not find that heavier users were more likely to reduce coping motives. Additionally, it is possible that participants in the motives feedback condition answered motives questions in a socially-desirable way in the follow-up, since they had received information on the problematic nature of the coping motive. Future research may clarify the impact of social desirability through study of internetbased feedback. Lastly, there was no measure included of participant response to motives feedback. Future studies may include a satisfaction rating or questions on how this information might impact their drinking. Despite limitations of the sample and the relatively brief follow-up period, the results are promising in showing the potential for coping motive feedback to augment the impact of brief PFIs. Further study of interventions specifically targeting drinking to cope with negative affect appears warranted. Designs including longer follow-ups, higher-risk samples, and perhaps additional types of motive feedback are needed. Role of funding source Funding for this study was provided by a grant through the Virginia Tech Graduate Research Development Program (GRDP); the GRDP had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Contributors CB and RS conceived, designed, and implemented the trial. CB and RS conceived the aims of this paper. CB performed the statistical analyses and wrote the initial draft of the manuscript. All authors discussed the results and contributed to and have approved the final manuscript. Conflict of interest No conflict declared. Acknowledgments The authors would like to thank Thomas Ollendick, Ph.D., Jungmeen Kim-Spoon, Ph.D., George Clum, Ph.D., and Steven Lash, Ph.D., whom were instrumental in study design.
References Banes, K.E., Stephens, R.S., Blevins, C.E., Walker, D.D., & Roffman, R.A. (2014). Changing motives for use: Outcomes from a cognitive-behavioral intervention for marijuanadependent adults. Drug and Alcohol Dependdence, 139, 41–46. Conrod, P.J., Stewart, S.H., Comeau, N., & Maclean, A.M. (2006). Efficacy of cognitive-behavioral interventions targeting personality risk factors for youth alcohol misuse. Journal of Clinical Child & Adolescent Psychology, 35(4), 550–563. Conrod, P.J., Stewart, S.H., Pihl, R.O., Côté, S., Fontaine, V., & Dongier, M. (2000). Efficacy of brief coping skills interventions that match different personality profiles of female substance abusers. Psychology of Addictive Behaviors, 14(3), 231–242. Conrod, P.J., Castellanos-Ryan, N., & Mackie, C. (2011). Long-term effects of a personalitytargeted intervention to reduce alcohol use in adolescents. Journal of Consulting and Clinical Psychology, 79, 296–306. Cooper, M.L. (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6, 117–128. Cooper, M.L., Russell, M., Skinner, J.B., & Windle, M. (1992). Development and validation of a three-dimensional measure of drinking motives. Psychological Assessment, 4, 123–132. Del Boca, F.K., & Darkes, J. (2003). The validity of self-reports of alcohol consumption: state of the science and challenges for research. Addiction, 98, 1–12. Demartini, K.S., & Carey, K.B. (2011). The role of anxiety sensitivity and drinking motives in predicting alcohol use : A critical review. Clinical Psychology Review, 31, 169–177. DiClemente, C.C., Bellino, L.E., & Neavins, T.M. (1999). Motivation for change and alcoholism treatment. Alcohol Research & Health : The Journal of the National Institute on Alcohol Abuse and Alcoholism, 23(2), 86–92. Faul, F., Erdfelder, E., Lang, A., & Buchner, A. (2007). G* Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. Grant, V.V., Stewart, S.H., O'Connor, R.M., Blackwell, E., & Conrod, P.J. (2007). Psychometric evaluation of the five-factor modified drinking motives questionnaire—Revised in undergraduates. Addictive Behaviors, 32, 2611–2632. Ham, L.S., & Hope, D.A. (2003). College students and problematic drinking: A review of the literature. Clinical Psychology Review, 23(5), 719–759. Hayes, A.F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford Publications. International Center for Alcohol Policies (2003). International drinking guidelines. Kruse, M., Fromme, K., & Corbin, W. (2005). Improving the accuracy of self-report measures of drinking: Disaggregating quantity and frequency indices of alcohol consumption. Alcoholism, Clinical and Experimental Research, 29(5).
C.E. Blevins, R.S. Stephens / Addictive Behaviors 58 (2016) 68–73 Kuntsche, E., Knibbe, R., Gmel, G., & Engels, R. (2005). Why do young people drink? A review of drinking motives. Clinical Psychology Review, 25(7), 841–861. Kuntsche, E., Knibbe, R., Gmel, G., & Engels, R. (2006). Who drinks and why? A review of socio-demographic, personality, and contextual issues behind the drinking motives in young people. Addictive Behaviors, 31(10), 1844–1857. Kushner, M.G., Abrams, K., Thuras, P., Hanson, K.L., Brekke, M., & Sletten, S. (2005). Follow-up study of anxiety disorder and alcohol dependence in comorbid alcoholism treatment patients. Alcoholism: Clinical & Experimental Research, 29(8), 1432–1443. LaBrie, J.W., Huchting, K., Tawalbeh, S., Pedersen, E.R., Thompson, A.D., Shelesky, K., ... Neighbors, C. (2008). A randomized motivational enhancement prevention group reduces drinking and alcohol consequences in first-year college women. Psychology of Addictive Behaviors, 22, 149–155. Leffingwell, T.R., Cooney, N.J., Murphy, J.G., Luczak, S., Rosen, G., Dougherty, D.M., & Barnett, N.P. (2013). Continuous objective monitoring of alcohol use: Twenty-first century measurement using transdermal sensors. Alcoholism: Clinical and Experimental Research, 37, 16–22. Meade, A.W., & Craig, S.B. (2012). Identifying careless responses in survey data. Psychological Methods, 17, 437–455. Merrill, J.E., Wardell, J.D., & Read, J.P. (2014). Drinking motives in the prospective prediction of unique alcohol-related consequences in college students. Journal of Studies on Alcohol and Drugs, 75, 93–102.
73
Mezquita, L., Stewart, S.H., Ibáñez, M.I., Ruipérez, M.A., Villa, H., Moya, J., & Ortet, G. (2011). Drinking motives in clinical and general populations. European Addiction Research, 17(5), 250–261. Moyer, A., Finney, J., Swearingen, C., & Vergun, P. (2002). Brief interventions for alcohol problems: A meta-analytic review of controlled investigations in treatmen-seeking and non-treatment-seeking populations. Addiction, 97, 279–292. Neighbors, C., Lewis, M.A., Bergstrom, R.L., & Larimer, M.E. (2006). Being controlled by normative influences: Self-determination as a moderator of a normative feedback alcohol intervention. Health Psychology, 25, 571–579. Nock, M.K. (2007). Conceptual and design essentials for evaluating mechanisms of change. Alcoholism: Clinical and Experimental Research, 31, 4S–12S. Preacher, K.J., & Hayes, A.F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891. Vasilaki, E., Hosier, S., & Cox, W. (2006). The efficacy of motivational interviewing as a brief intervention for excessive drinking: a meta-analytic review. Alcohol and Alcoholism, 328-335. White, H.R., & LaBouvie, E.W. (1989). Towards the assessment of alcohol problem drinking. Journal of Studies on Alcohol, 50, 30–37. Willem, L., Bijttebier, P., Claes, L., & Uytterhaegen, A. (2012). Temperament and problematic alcohol use in adolescence: An examination of drinking motives as mediators. Journal of Psychopathology and Behavioral Assessment, 34(2), 282–292.
