مشکلات خود و همکار مربوط به مصرف الکل در میان ACOAs و غیر ACOAs: ارتباط با نشانه های افسردگی و انگیزش برای استفاده از الکل

The present study examined whether drinking motivations and depressive symptoms would have a stronger impact on alcohol-related problems among adult children of alcoholics (ACOAs) and their dating partners as compared to non-ACOAs and their dating partners. Participants were 197 undergraduate (60 ACOAs, 137 non-ACOAs) 18 to 25 year-old female drinkers in dating relationships. Participants completed measures of ACOA screening, depressive symptoms, and drinking motives, as well as alcohol-related problems for themselves and their partner. Although no differences were found between ACOA and non-ACOA women's alcohol-related problems, ACOA women and women with greater depressive symptoms were at a higher risk of having a partner with more alcohol-related problems. In addition, we found that regardless of parental history of alcoholism, higher depressive symptoms coupled with stronger motives for drinking to cope with stressors predicted participants' own alcohol-related problems. These findings demonstrate the need for future research to examine additional factors that may moderate the effects of depressive symptoms and ACOA status on female college student drinking problems. A greater understanding of the unique and interactive effects of these variables on alcohol-related problems in both young women and their dating partners can aid in the development of prevention programs more targeted to the specific vulnerabilities of this population.

Between 25% and 30% of college students meet criteria for being an adult child of an alcoholic (ACOA; e.g., Grant, 2000). ACOAs are more likely to have alcohol use disorders (AUDs) than non-ACOAs (Guo, Hawkins, Hill, & Abbott, 2001), with heritability estimated at 50% (Goldman, Oroszi, & Ducci, 2005). Although many ACOAs have a strong genetic predisposition for alcohol misuse, other factors, such as depression (e.g., Pedrelli et al., 2011 and Weitzman, 2004) and drinking motivations (e.g., Cooper, 1994, LaBrie et al., 2011 and Read et al., 2003), are associated with alcohol-related problems in college students. For this reason, we examined the role of depressive symptoms and drinking motivations among ACOA and non-ACOA undergraduates as related to their own and their partners' alcohol-related problems. We elected to study undergraduate women because women who misuse alcohol appear to choose a partner in part based on similarity in alcohol use (e.g., Agrawal et al., 2006, Merline et al., 2008 and Tuten and Jones, 2003), women may be more susceptible to their partner's drinking (Agrawal et al., 2006 and Tuten and Jones, 2003), and among women with alcohol-related problems, drinking appears integral to their romantic relationships (Thom, 1987). 1.1. Alcohol misuse among adult children of alcoholics One of the strongest predictors of problematic alcohol use is having a parent who is alcohol dependent (e.g., Hill et al., 2011 and Kendler et al., 2008) or is thought to misuse alcohol (e.g., Braitman, Kelley, Ladage, Schroeder, Gumienny, Morrow, et al., 2009). Both adolescents (Obot, Wagner, & Anthony, 2001) and undergraduate students (Braitman et al., 2009) whose parents misuse alcohol drink more than peers with no parental history of such use. Furthermore, children of parents who misuse alcohol initiate drinking earlier (Vermeulen-Smit et al., 2012), escalate drinking more quickly (Chassin et al., 1996 and Warner et al., 2007), report more negative outcomes from alcohol use (Pollock, Schneider, Gabrielli, & Goodwin, 1987), and are at greater risk for alcohol misuse (e.g., Braitman et al., 2009, Coffelt et al., 2006, Duncan et al., 2006 and Hicks et al., 2010). In fact, King and Chassin (2007) found that parental alcoholism doubled the risk for offspring alcohol dependence. 1.2. Depression and alcohol use among ACOAs and the general population In addition to genetic propensity, internalizing symptoms have been argued as one path that may increase risk for alcohol use disorders (AUDs) among those with a parental history of alcoholism (e.g., Zucker, 2006). From a developmental ecological framework (e.g., Bronfenbrenner, 1986), the contexts created by parental alcohol misuse have broad implications for the family environment (e.g., periods in which alcohol-abusing parents are less emotionally available to children, negative parental moods, interparental conflict, family stress) that may increase offspring risk for internalizing symptoms. More globally, mood disorders are strongly correlated with alcohol use disorders (see Edwards et al., 2012 and Hasin et al., 2007). For instance, controlling for sociodemographic characteristics and other psychiatric disorders, Hasin et al. (2007) found that individuals who met DSM-IV criteria for alcohol dependence were over two times more likely to have major depression. Importantly, ACOAs are more vulnerable to a variety of internalizing disorders including depression (e.g., Burstein et al., 2006, Hussong et al., 2008, Kelley et al., 2010 and Mylant et al., 2002). Specifically, 37% of ACOAs are thought to experience lifetime depression (Anda et al., 2002 and Cuijpers et al., 1999). Comparatively, data from the U.S. National Comorbidity Survey Replication reports that 20.9% of individuals between 18 and 64 years of age meet criteria for one or more major depressive episodes in their lifetimes (Kessler & Merikangas, 2004). Women with a family history of alcoholism appear at particular risk for major depression or dysthymia compared to men (e.g., Raucher-Chéné et al., 2012). Using data from the National Epidemiological Survey on Alcohol and Related Conditions, Morgan, Desai, and Potenza (2010) found that women with parental history of alcoholism were significantly more likely to have a history of major depression or dysthymia than men with a history of parental alcoholism or men or women without a history of parental alcoholism (Morgan et al., 2010). Although depression is a risk factor for heavy drinking and alcoholism in women (Weitzman, 2004), genetic propensity combined with depressive symptoms may increase the likelihood of alcohol-related problems among female ACOAs. 1.3. Motivations for alcohol use among undergraduate drinkers Drinking motivations represent another important factor in the understanding of alcohol consumption and related problems among college students. From this vantage, individuals drink to fulfill a particular function (Cooper, 1994). Various drinking motives, including coping (i.e., drinking to reduce stressors), social (i.e., drinking to become more sociable), and enhancement (i.e., drinking to improve one's positive affect) are each uniquely associated with negative alcohol outcomes (Cooper, 1994). Coping motives are largely recognized as the strongest predictor of alcohol-related problems among the drinking motives (Cooper, 1994 and Kuntsche et al., 2005), while social motives are typically associated with light, non-problematic drinking (Cooper, 1994). Enhancement motives, on the other hand, have been found to predict alcohol use (Cooper et al., 1995, Kuntsche and Cooper, 2010 and Magid et al., 2007), heavy drinking (Kuntsche & Cooper, 2010), and indirectly, alcohol-related problems (Magid et al., 2007). Associations between drinking motives and alcohol outcomes are well established among college students who drink (e.g., Cooper, 1994, Cooper et al., 1995, LaBrie et al., 2011 and Read et al., 2003). Relatively few studies have examined drinking motivations among ACOAs. However, the enhanced reinforcement pathway posits that genetic propensity in combination with stronger beliefs transmitted by parents about alcohol's ability to heighten experiences (e.g., Gorth & Söoderpalm, 2011) or reduce negative affect (Schuckit & Smith, 1997) may increase alcohol consumption and alcohol-related consequences among ACOAs. Relatedly, the stress and negative affect pathway contends that ACOAs may be more likely to drink to cope with stressors (see Zucker, Donovan, Masten, Mattson, & Moss, 2008). In fact, Avant, Davis, and Cranston (2011) argued that ACOAs may use alcohol as a mechanism to cope with negative affective states. In one of the few studies to address drinking motives of ACOAs versus non-ACOAs, Chalder, Elgar, and Bennett (2006) found ACOA adolescents in South Wales were more likely to use alcohol to cope with problems or for enhancement, but not for social motives. Domenico and Windle's (1993) study involving adult women also revealed that ACOAs reported drinking to cope with stressful events more than non-ACOAs. However, these authors did not find significant differences between ACOAs and non-ACOAs on social or enhancement motives. These findings present conflicting evidence regarding enhancement motives for ACOAs, and relatively little research has examined whether alcohol motives may be associated with alcohol-related problems among ACOA and non-ACOA college students. 1.4. Similarity in alcohol use among women and their partners Although parental history of alcoholism, mood, and coping motives are often examined in relation to young women's own alcohol problems, fewer studies have examined whether these variables are associated with their dating partners' alcohol use. Importantly, dating partners' alcohol use may impact young women's alcohol consumption and the trajectory of alcohol use. For instance, van der Zwaluw et al. (2009) found that female adolescents who consumed alcohol at a greater frequency were significantly more likely to select a dating partner who also used alcohol frequently. This pattern was not found for male adolescents. Moreover, women in heavy drinking partnerships consume more alcohol on average than women in other drinking partnerships (Graham and Braun, 1999 and Roberts and Leonard, 1998). Similarly, in a study that tracked newly married couples over the first four years of marriage, problem-drinking partners fostered their partners' alcohol use and alcohol-related problems (Leonard & Homish, 2008). Wilsnack and Wilsnack (1993) provided an explanation for this phenomenon by speculating that when both partners misuse alcohol, alcohol becomes a shared recreational activity and partners may not discourage one another from reducing alcohol consumption. Because a partner's alcohol use appears strongly linked to women's own alcohol use and may increase women's drinking trajectory, we examined how depressive symptoms and motivations for drinking were associated with women's reports of their own, as well as their partners' alcohol-related problems. The present study specifically focused on alcohol-related consequences, rather than consumption, because the quantity or frequency of alcohol use may not be sufficient in understanding one's alcohol use severity (see Ham & Hope, 2003 for a review). Furthermore, alcohol-related problems are particularly important to examine among young women because the onset of alcohol abuse and dependence peaks at age 19 (Hasin et al., 2007) and because women inherently metabolize alcohol differently than men resulting in greater susceptibility to experiencing some types of alcohol-related problems (e.g., sexual victimization; Jersild, 2002, Perkins and Berkowitz, 1991 and Stockwell et al., 2002). We also chose to focus on women because alcohol dependent women are more likely to have substance-dependent partners (e.g., Agrawal et al., 2006, Center for Substance Abuse Treatment, 2009, Riehman et al., 2003 and Tuten and Jones, 2003), and as compared to men, women have a harder time discontinuing relationships with substance-abusing partners (CSAT, 2009). In an attempt to further understand the association between ACOAs and alcohol-related problems, we examined whether depressive symptoms and motivations for alcohol use (i.e., drinking to cope with stress, drinking to boost social interactions, drinking to enhance the effects of alcohol) had stronger effects on ACOAs' alcohol-related problems and reports of their dating partners' alcohol-related problems than those of non-ACOAs. We hypothesized a stronger negative effect for ACOAs. Specifically, we hypothesized that depressive symptoms and drinking motivations would have a stronger impact on alcohol-related problems among ACOAs and their dating partners as compared to non-ACOAs and their dating partners. These results provide important implications in the development of interventions specifically targeting college-aged females' alcohol outcomes.

3.1. Preliminary analyses Prior to hypothesis testing, data were checked for outliers and missingness. Boxplots revealed that three values were outside the third interquartile range; therefore, these scores were transformed to a value that was next to the highest score in the data. Missingness ranged from approximately 1% on the enhancement motive subscale of the DMQ-R to approximately 4% on the alcohol-related problem subscale from the AUDIT (for partner). These missing values were imputed using the expectation–maximization algorithm. Table 2 presents descriptive statistics and intercorrelations among study variables. Table 2. Descriptive statistics and intercorrelations among study variables. Variable 1 2 3 4 5 6 7 1. Alcohol-related problems (for self) (.76)a 2. Alcohol-related problems (for partner) .47⁎⁎⁎ (.78) 3. ACOA status − .21⁎⁎ − .27⁎⁎⁎ (.97) 4. Depressive symptoms .40⁎⁎⁎ .32⁎⁎⁎ − .16⁎ (.92) 5. Coping motives .47⁎⁎⁎ .28⁎⁎⁎ − .16⁎ .35⁎⁎⁎ (.88) 6. Social motives .28⁎⁎⁎ .18⁎⁎ − .09 .17⁎ .52⁎⁎⁎ (.89) 7. Enhancement motives .28⁎⁎⁎ .21⁎⁎ − .14⁎ .21⁎⁎ .59⁎⁎⁎ .76⁎⁎⁎ (.88) Mean 2.76 3.05 .70a 9.95 9.98 16.09 13.61 Standard deviation 3.38 3.80 − 5.25 4.83 5.26 5.40 Range 0–15 0–17 – 6–27 5–25 5–25 5–25 Skewness 1.50 1.46 – 1.61 .96 − .30 .25 Kurtosis 1.81 1.83 – 1.88 .10 − .75 − .64 Note. ACOA = adult child of an alcoholic. ACOA status is coded 0 = ACOA, 1 = non-ACOA, therefore, negative estimates reflect stronger effects for ACOAs. a Mean can be interpreted as proportion of non-ACOAs in sample. Coefficient alphas are presented along the diagonal. ⁎ p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001. Table options 3.2. Overview of moderation analyses To determine if depressive symptoms and drinking motives (i.e., coping, social, and enhancement) moderated the relationship between ACOA status and alcohol-related problems for self and partner, least squares regression models were estimated. Specifically, three regression models were estimated with alcohol-related problems (for the participants themselves) as the criterion and three regression models were estimated with participants' perceptions of their partners' alcohol-related problems as the criterion. In all models, the main effects were ACOA status, depressive symptoms, and drinking motives (i.e., drinking to cope, social motives, enhancement motives). ACOA status was coded as 0 = ACOA and 1 = non-ACOA; therefore, main effects of the continuous predictors (i.e., depressive symptoms and drinking motives) should be interpreted with respect to ACOAs. All continuous covariates were grand mean centered. Non-normality in the data was addressed using a non-parametric bootstrap approach through Mplus 5.2 (Muthén & Muthén, 2008). To account for multiple comparisons with each dependent variable, statistical significance was assessed using 99% bias-corrected (BC) confidence intervals (CIs) generated from 2000 bootstrap samples with replacement. Specifically, if zero is not contained within the unstandardized 99% BC CIs, then the parameter estimate is statistically significant at the .01 probability level (Efron & Tibshirani, 1993). To follow-up significant two-way interactions, simple slope analyses were conducted at different levels of the moderator variable of interest; α was set at .05 for the simple slope analyses. 3.3. Model with coping motives 3.3.1. Alcohol-related problems for self Results of the regression model predicting alcohol-related problems for self revealed that the overall model was statistically significant, F(6, 190) = 16.59, p < .001, R2 = .344 (see Table 2). Moreover, coping motives and their interaction with depressive symptoms were significant predictors of participants' alcohol-related problems. To further understand the significant interaction, the slopes of alcohol-related problems for self on depressive symptoms were examined at different levels (i.e., one standard deviation below the mean, at the mean, and one standard deviation above the mean) of coping motives. Results revealed that as endorsement of coping with stressor motives increased, the strength of the slope between alcohol-related problems for self and depressive symptoms became stronger (see Fig. 1), while controlling for the effect of ACOA status. More specifically, the simple slopes of alcohol-related problems for self on depressive symptoms were not significantly different from zero at lower levels (i.e., one standard deviation below the mean) of coping motives, B = − 0.04, 95% BC CI [− 0.24, 0.19] or at average coping motives, B = 0.09, 95% BC CI [− 0.05, 0.28]; however, the simple slope was significantly different from zero at higher levels (i.e., one standard deviation above the mean) of coping motives, B = 0.22, 95% BC CI [0.07, 0.38]. Full-size image (13 K) Fig. 1. Simple slopes of alcohol-related problems (for self) on depressive symptoms by coping motives. COP = coping motives. Figure options 3.3.2. Alcohol-related problems for partner Results of the regression model predicting alcohol-related problems for participants' dating partners revealed that the overall model was statistically significant, F(6, 190) = 8.46, p < .001, R2 = .211 (see Table 3). Two main effects (i.e., ACOA status and depressive symptoms) emerged as significant predictors of partners' alcohol-related problems. Specifically, at average depressive symptoms and coping with stressor motives, as compared to non-ACOAs, ACOAs reported that their partners had significantly more alcohol-related problems. Moreover, for ACOAs with average coping motives, higher levels of depressive symptoms were associated with a significant increase in alcohol-related problems for their partners. No two-way interactions were statistically significant. Table 3. Summary of regression analyses examining depressive symptoms and coping motives as moderators. Predictor Self — AUDIT problems Partner — AUDIT problems 99% BC CI for B 99% BC CI for B sr2 B LL UL sr2 B LL UL ACOA .01 − 0.77 − 1.98 0.36 .03 − 1.57⁎ − 2.96 − 0.11 DEP .01 0.09 − 0.10 0.35 .07 0.36⁎ 0.05 0.65 COP .06 0.35⁎ 0.10 0.66 .00 0.09 − 0.20 0.37 ACOA × DEP .00 0.02 − 0.33 0.30 .03 − 0.26 − 0.60 0.10 ACOA × COP .01 − 0.15 − 0.47 0.13 .00 0.07 − 0.24 0.41 DEP × COP .04 0.03⁎ 0.00 0.05 .02 − 0.02 − 0.05 0.02 Note. ACOA = adult child of an alcoholic (0 = ACOA, 1 = non-ACOA); DEP = depressive symptoms (as measured by the Depression subscale of the Brief Symptoms Inventory); COP = coping motives (as measured by the Coping subscale of the Drinking Motives Questionnaire). All continuous covariates were grand mean centered. BC = bias-corrected. ⁎ p < .01. Table options 3.4. Model with social motives 3.4.1. Alcohol-related problems for self Results of the regression model predicting one's own alcohol-related problems revealed that the overall model was statistically significant, F(6, 190) = 10.55, p < .001, R2 = .250 (see Table 4). However, after adjusting for multiple comparisons, results demonstrated no significant main effects or two-way interactions. Table 4. Summary of regression analyses examining depressive symptoms and social motives as moderators. Predictor Self-audit problems Partner-audit problems 99% BC CI for B 99% BC CI for B sr2 B LL UL sr2 B LL UL ACOA .02 − 1.00 − 2.28 0.30 .04 − 1.71⁎ − 3.26 − 0.28 DEP .04 0.22 − 0.04 0.50 .08 0.35⁎ 0.09 0.65 SOC .00 0.05 − 0.22 0.27 .00 0.03 − 0.30 0.31 ACOA × DEP .00 − 0.03 − 0.36 0.27 .02 − 0.24 − 0.61 0.07 ACOA × SOC .01 0.14 − 0.11 0.44 .00 0.06 − 0.23 0.44 DEP × SOC .02 0.02 − 0.01 0.04 .01 − 0.01 − 0.06 0.02 Note. ACOA = adult child of an alcoholic (0 = ACOA, 1 = non-ACOA); DEP = depressive symptoms (as measured by the Depression subscale of the Brief Symptoms Inventory); SOC = social motives (as measured by the Social subscale of the Drinking Motives Questionnaire). All continuous covariates were grand mean centered. BC = bias-corrected. ⁎ p < .01. Table options 3.4.2. Alcohol-related problems for partner Results of the regression model predicting alcohol-related problems for women's dating partners revealed that the overall model was statistically significant, F(6, 190) = 7.52, p < .001, R2 = .192 (see Table 4). Two main effects (i.e., ACOA status and depressive symptoms) were significant predictors of a partner's alcohol-related problems. No two-way interactions were statistically significant. 3.5. Model with enhancement motives 3.5.1. Alcohol-related problems for self Results of the regression model predicting alcohol-related problems (for self) revealed that the overall model was statistically significant, F(6, 190) = 9.67, p < .001, R2 = .234 (see Table 5). However, after adjusting for multiple comparisons, results demonstrated no significant main effects or two-way interactions. Table 5. Summary of regression analyses examining depressive symptoms and enhancement motives as moderators. Predictor Self-audit problems Partner-audit problems 99% BC CI for B 99% BC CI for B sr2 B LL UL sr2 B LL UL ACOA .02 − 1.03 − 2.37 0.33 .04 − 1.69⁎ − 3.20 − 0.14 DEP .06 0.26 − 0.22 0.53 .07 0.33⁎ 0.09 0.67 ENH .00 − 0.01 − 0.29 0.20 .00 0.02 − 0.21 0.28 ACOA × DEP .00 − 0.09 − 0.41 0.25 .02 − 0.24 − 0.61 0.08 ACOA × ENH .02 0.19 − 0.04 0.46 .00 0.11 − 0.19 0.38 DEP × ENH .00 0.01 − 0.02 0.04 .00 0.00 − 0.03 0.03 Note. ACOA = adult child of an alcoholic (0 = ACOA, 1 = non-ACOA); DEP = depressive symptoms (as measured by the Depression subscale of the Brief Symptoms Inventory); ENH = enhancement motives (as measured by the Enhancement subscale of the Drinking Motives Questionnaire). All continuous covariates were grand mean centered. BC = bias-corrected. ⁎ p < .01. Table options 3.5.2. Alcohol-related problems for partner Results of the regression model predicting alcohol-related problems for women's dating partners revealed that the overall model was statistically significant, F(6, 190) = 7.58, p < .001, R2 = .193 (see Table 5). In particular, results revealed that with average depressive symptoms and enhancement motives, ACOAs reported that their dating partners had significantly more alcohol-related problems than non-ACOAs. Furthermore, for ACOAs with average enhancement motives, higher depressive symptoms were associated with a significant increase in alcohol-related problems among their dating partners.