Understanding the Association Between Childhood Maltreatment
and Alcohol Misuse in a Student Population
Rina Apriliani Sugiarti S2226790
Master Thesis Clinical Psychology Supervisor: Dr. M. S. Tollenaar Institute of Psychology
Universiteit Leiden 2020
Table of Contents
Abstract ... 3
1. Introduction ... 4
2. Methods ... 8
2.1. Research Design and Participants ... 8
2.2. Measures ... 8
2.2.1 Childhood Trauma Questionnaire (CTQ) ... 8
2.2.2 Alcohol Use Disorders Identification Test (AUDIT) ... 9
2.2.3 N2-back test ... 9 2.3. Procedure ... 9 2.4. Statistical Analysis ... 10 3. Results ... 11 3.1. Descriptive Statistics ... 11 3.2. Assumptions ... 12
3.3. Multiple Linear Regression Analysis ... 13
3.4. Mediation Analysis ... 14
4. Discussion ... 15
4.1. Strengths of the Study ... 18
4.2. Limitations and Further Research ... 19
4.3. Implications and Conclusions ... 20
ABSTRACT
Background: Alcohol misuse continues to be one of the global health burdens. Understanding the underlying mechanism of alcohol misuse is considered necessary to advance the current interventions. Here we explored the role of childhood maltreatment and working memory performance in drinking problems among student population.
Objective: The first aim of the study is to investigate the association between exposure to childhood maltreatment and alcohol misuse. We then explored if the relationship between childhood maltreatment and alcohol misuse was mediated by working memory performance. Methods: A total of 77 participants (63.2% female, age ranged from 18 to 21 years old) were recruited at the Radboud University in Nijmegen to complete a computer task and several self-report questionnaires. Childhood maltreatment was measured with the Childhood Trauma Questionnaire (CTQ), alcohol misuse was assessed by the Alcohol Use Disorders Identification Test (AUDIT), and an N2-back test was used to measure working memory performance. Multiple regression analysis and mediation analysis were performed to test the hypotheses. Results: Experiencing childhood maltreatment predicted increased risk of alcohol misuse (B = .223, t (72) = 3.514, p = .001). However, the association was not significantly mediated by working memory performance.
Conclusions: The findings in this study corroborate the evidence that childhood adversities associated with more alcohol problems. Screening tests and tailored treatment for alcohol misuse that considers the comorbidity of childhood trauma are expected to result in better treatment outcomes. However, further studies are needed to explain the underlying mechanisms between childhood maltreatment and alcohol misuse.
Keywords: childhood maltreatment, alcohol misuse, working memory, cross-sectional, mediation
1. INTRODUCTION
Despite the negative consequences of its persistent consumption (Sher et al., 2012), alcohol continues to be one of the most commonly used substances worldwide. In 2016, based on drinking pattern in the last 12 months, 43% of the global population were reported as current drinkers (World Health Organization, 2018). The number exceeded half of the population in the Americas, Europe, and the Western Pacific. Among the current drinkers worldwide, more than a quarter of them were 15-19-year-olds, which accounted for approximately 155 million adolescents. WHO (2018) from the same report conveyed that harmful use of alcohol (any drinking patterns that cause health and social consequences) had caused three million deaths worldwide. The rate of mortality was even higher than tuberculosis, HIV/AIDS, and diabetes. It also cost 132.6 million Disability-Adjusted Life Years (DALYs) – 5.1% of all DALYs in that year – which came in the forms of injuries, digestive diseases, and alcohol use disorders. Because the devastating global consequences remain high, alcohol is one of the health priorities in the 2030 Agenda of Sustainable Development Goals (SDGs). Understanding the factors that influence the development of alcohol misuse is necessary to advance the current treatment and prevention programs.
Both genetic and environmental factors are known to contribute to the development of substance dependence. The heritability of alcoholism is established at 50%, which indicates that environmental factors contribute similarly (Enoch, 2011, Magnusson et al., 2012). From environmental determinants, one of the possible factors is experiencing childhood trauma (Shin et al., 2009). Experiences of early life stress can produce a profound and lasting influence that affects individual’s mental and physical health during the life course (McEwen, 2008). Alterations in brain development that occur in response to childhood stress is proposed to explain the mechanism, as the associated cognitive impairments cause changes in cognitive functioning that is found similar to those observed in many psychiatric disorders (Anda et al., 2006, Philip et al., 2016). Childhood maltreatment (physical, sexual, emotional and abuse; physical and emotional neglect) is shown to associate with higher vulnerability to various psychiatric comorbidities (Huang et al., 2012) including alcohol problems.
A number of studies that has been conducted found an increased risk of alcohol dependence for individuals with childhood maltreatment experience (Clark et al., 1997, Evren et al., 2006, Huang et al., 2012, Mirsal et al., 2004, Schwandt et al., 2013, Young-Wolff, et al.,
2010). In a study of twin males, Young Wolff et al. (2010) found that compared to a control group, exposure to childhood maltreatment associated with 1.7 times higher risk of alcohol dependence. In another study by Schwandt et al., (2013), emotional abuse was ten times more likely to be found in alcohol dependent subjects compared to the nondependent group. They also found that the association was mediated by impulsiveness which reflects in difficulty to overcome urge. However, most of these investigations only included subjects who met diagnostic criteria of alcohol dependence. Research focusing on the correlation between childhood maltreatment and alcohol misuse within non-clinical samples is scarce, with some exceptions (Dube et al., 2002, Goldstein et al., 2010, Schwandt et al, 2013, Shin et al., 2009), and the findings were mixed. In one study by Schwandt et al., (2013), for example, childhood trauma was found not associated with alcohol misuse in non-clinical individuals, while in another study by Shin et al. (2009), the association was only found on individuals exposed to multiple types of maltreatment. Other than that, most of the research excluded other patterns of drinking that are actually also considered as harmful, such as hazardous drinking, single episode of harmful use of alcohol, and harmful pattern of drinking (WHO, 2018). Heavy episodic drinkers, for example, accounted for 39.5% of alcohol drinkers worldwide, with half of them were coming from young people population at the age of 20-24 years old (WHO, 2018). This population is known as less likely to have received treatment for their alcoholism although the group is at high vulnerability of drinking problems (DeMartini & Carey, 2012). Thus, considering the possible adverse consequences, more studies examining drinking problems in this non-clinical population is needed. This study will be conducted to address this gap.
To analyse early life adversities, examining the effects of various types of childhood trauma cumulatively is recently suggested, as it often shows a more harmful impact than analysing of a single type of maltreatment (Goldstein et al., 2010). The foci of the studies that had been conducted to assess the relationship between childhood maltreatment and alcohol problems tend to aim at a specific type of abuse such as physical abuse or sexual abuse (Dube et al., 2002, Goldstein et al., 2010, Jasinski et al., 2000, Young-Wolff et al., 2011), and left out the other types of childhood maltreatment, such as emotional abuse and emotional neglect. The possible explanation is because physical and sexual abuse are presumably thought to be more prevalent and dangerous than other forms of maltreatment (Schwandt et al., 2013). Meanwhile, data showed that one in four adults reported experiencing abuse as children. Thirty-six percent of them were emotionally abused, and the prevalence was higher than sexual abuse (WHO,
2014). Moreover, experiencing multiple forms of maltreatment was found to be common; polyvictimization might be the norm that makes it necessary to measure different types of trauma simultaneously (Schwandt, 2013). Considering the high prevalence rate of childhood maltreatment in the community, cumulatively using continuous scores that include various types of maltreatment may be a better predictor to measure an outcome, which will be carried out in this study.
Most of the investigations on childhood maltreatment and alcohol abuse in adult samples stem from observational studies (Banerjee et al., 2018, Dube et al., 2002, Goodman et al., 2019, Rothman et al., 2008). The underlying processes of its association with psychopathology development such as alcohol misuse are still unclear and need to be furtherly investigated (Goldstein et al., 2010, Trautmann et al., 2018). Researchers have previously suggested a mediating role of coping skills, post-traumatic stress disorder, and antisocial behaviour in the relationship between child maltreatment and alcohol abuse (Widom, 2001). From a more recent analysis, there is a growing interest in examining possible explanations from a cognitive perspective. Early life stress is found to have lasting effects on the developing brain that alters its cognitive function (McEwen, 2008). Childhood trauma indicates detrimental effects on brain regions that are associated with addiction susceptibility (Andersen, 2019, Enoch, 2011). The impairment on brain function is found in higher order cognitive functions including working memory which is considered important for self-control (Cowell, 2015).
Several systematic reviews (Goodman et al., 2019, Pechtel & Pizzagalli, 2010) have shown associations specifically between early life stress and deficits in working memory performance. It implied that adverse childhood experiences lead to a change in cognitive capabilities, including cognitive performance, memory, and executive functioning. Exposure to early life stress disrupts development in brain regions which associate with higher-order cognitive functions. These brain areas undergo prolonged developmental trajectories that make it more vulnerable to impairment when individuals experience childhood trauma (Pechtel & Pizzagalli, 2010). As an elementary component of executive cognitive functioning, working memory is responsible for temporary storage and manipulation of information during complex cognitive processing (Baddeley et al., 2011). Working memory contributes to the prediction of cognitive capabilities, such as reading comprehension, language acquisition, nonverbal problem solving, and reasoning skills (Morrison & Chein, 2011). Recently, studies have found working memory’s influence on self-regulation, which indicated that deficits in working
memory were linked to difficulty in behaviour control that exists in substance use problem such as alcohol misuse (Ellingson et al., 2014, Tahaney & Palfai, 2018, Day et al., 2013, Peeters et al., 2015). Working memory deficits increase individual impulsivity in responding to stimuli. The impulsivity may occur due to lower ability in maintaining task-relevant information during complex cognitive tasks, which is an important cognitive element of self-control abilities (Day et al., 2013, Tahaney et al., 2018). Thus, lower working memory likely leads individuals to fulfilling immediate satisfaction such as drinking rather than following a controlled behaviour (Peeters et al., 2015).
Taken together, experiencing childhood maltreatment affects brain development which in turn has an impact on one’s cognitive capabilities such as executive functioning. Working memory, as a core component of executive functioning, has been found in latest studies to associate with problems of self-control. Lower performance of working memory correlates with higher level of impulsivity and difficulty in behaviour control which is often found in individual with substance use problems. It implies the possibility of working memory performance as a mechanism in the association between childhood maltreatment and alcohol misuse. Research on these associations could lead to a possibility of explanation one of the mechanism between early life stress and drinking problems. Confirming such a mechanism may help to optimize treatment for alcohol use disorder by targeting components that are crucial for recovery. Working memory training, for example, has been suggested as a treatment option with promising results for substance use (Gunn et al., 2018, Hendershot et al., 2018, Houben et al., 2011). Working memory training created a possibility of improving executive cognitive function that would help persons with drinking problems increasing control over their automatic impulses to consume alcohol (Houben et al., 2011). Thus, this study will examine the possible role of working memory capacity in mediating the association between childhood maltreatment and alcohol misuse.
In summary, there were two objectives in this study. The first one was to identify the relationship between childhood maltreatment and alcohol misuse in a non-clinical sample from the student population. The hypothesis was that childhood trauma experience would positively correlate with alcohol misuse. The second objective was to examine working memory capacity as a mediator of the relationship between childhood maltreatment and alcohol misuse. The hypothesis was that lower levels of working memory performance would manifest in
individuals with higher severity of childhood maltreatment, thereby increasing the level of alcohol misuse.
2. METHODS
2.1. Research Design and Participants
The design of this research was a cross-sectional study, using previously collected data. Participants completed computer tasks and several self-report questionnaires. The sample in this study consisted of 77 students. Participants were college students recruited at Radboud University Nijmegen who responded to recruitment by flyers or advertisement on the university website. Inclusion criteria were female or male students, with an age range from 17-35 years old who were fluent in speaking Dutch. There were no exclusion criteria.
2.2. Measures
2.2.1 Childhood Trauma Questionnaire (CTQ)
Childhood maltreatment was retrospectively measured using the Childhood Trauma Questionnaire (Bernstein et al., 2003). The CTQ consists of 28 items and is a self-report questionnaire measuring five categories of childhood trauma, i.e. physical, sexual, and emotional abuse, and physical and emotional neglect. An example of an item to measure emotional abuse is: “People in my family said hurtful or insulting things to me.” Each type of maltreatment has five questions to measure frequency with the score ranges from 1 (never true) to 5 (very often true), yielding dimensional scales that enhance reliability and maximize statistical power. In this study, the overall CTQ score indicated level of maltreatment severity and was used in the analysis as continuous data. Cut-offs were used to identify exposure of at least low severity of maltreatment, i.e. emotional abuse ³ 9, physical abuse ³ 8, sexual abuse ³ 6, emotional neglect ³ 10, and physical neglect ³ 8 (Schwandt et al., 2013). The scale has shown good reliability and validity in clinical and nonclinical samples (Bernstein et al., 2003, Thombs, et al., 2009). Cronbach’s alpha for the CTQ subscales were previously reported .91 for physical abuse, .89 for emotional abuse, .95 for sexual abuse, .63 for physical neglect, and .91 for emotional neglect (Thombs et al., 2009).
2.2.2 Alcohol Use Disorders Identification Test (AUDIT)
Alcohol misuse was measured using the AUDIT (Saunders et al., 1993). The instrument is a self-report questionnaire for assessing alcohol use during the last year, which includes alcohol consumption, drinking behaviour, and alcohol related problems. It consists of 10 questions with response option ranges from 0 (never) to 4 (daily or almost daily). An example of an item is: “How often during the last year have you found that you were not able to stop drinking once you had started?” In this study, the overall score will be used to measure the hazardous level of the alcohol misuse in continuous data. Cut-off score of 8 with maximum score of 40 is suggested to identify alcohol consumption at-risk for physical and/or psychological harm (DeMartini & Carey, 2012, Reinert & Allen, 2007). The AUDIT has shown good reliability and validity with Cronbach’s alpha .83 (Bradley et al., 2003, Reinert & Allen, 2007).
2.2.3 N2-back test
To measure working memory performance, a computerized N2-back test using E-Prime program was used (Jaeggi et al., 2010). In the N-back test, the participants view a stream of stimuli. Their task is to decide whether each stimulus that appears is similar to the stimulus N positions before. In this study, verbal material (small and capital letters) was used with the load level of N = 2. Each stimulus presented for 450 ms with interstimulus interval of 750 ms, after which the next trial appeared. If the stimulus matched with the stimulus from 2 positions before, the participant should hit the left mouse button. There were 197 trials in total, which in 64 trials the letters were matched. The participant could hit them or miss them. They could also hit accidentally (false alarm). Each participant had the chance to practice with 15 trials before the test was started. The outcome of working memory in this study was measured from the correct response rate [(hits - false alarms)/targets] x 100.
2.3. Procedure
The procedures in this study were approved by the Psychology Ethic Committee of the Radboud University. All procedures of measurement were conducted at the laboratory of Radboud University Nijmegen, executed by a trained Master student. Participants first received information about the study and then signed an informed consent as a form of agreement. A word recall test then conducted (not described here), followed by the N2-back test and several psychological self-reported questionnaires, including the CTQ and AUDIT. Thirty-five minutes
after the first test, a second recall task was performed. Participants could continue completing the questionnaires after that. The duration of the test session in total was approximately one hour per participant. All of the measurements were computerized using E-Prime programme. At the end of the session the participants were given a debriefing form of the study and were paid study credits or 6,50 euros.
2.4. Statistical Analysis
All of the analyses were conducted using IBM SPSS 26. Data were first examined for descriptive statistics and were checked for assumptions (linearity, normality, homoscedasticity, and independence). In addition, data were also screened for outliers. Correlations between variables were then calculated using Spearman correlation.
For the first hypothesis, ordinary least squares multiple linear regression was performed to examine whether childhood maltreatment could predict alcohol misuse. Multiple linear regression could help to rule out possible explanation from other variables which were not of interest in the study. In this analysis, CTQ total score was the independent variable and AUDIT score was the dependent variable. Demographic variables of age and gender were included as the covariates.
To analyse the second hypothesis, the mediation analysis in this study was conducted using the methods recommended by Preacher and Hayes (2013). The PROCESS macro for SPSS was used to estimate the regression coefficient of the indirect path from childhood maltreatment severity to harmful level of alcohol misuse through working memory performance, with 10000 bootstrapped samples. The coefficient was the result of calculating the OLS regression coefficients from childhood maltreatment to working memory performance (a path), and from working memory performance to harmful level of alcohol misuse (b path). Mediation was established if the 95% bias-corrected confidence interval for the product of the a and b path did not include zero (Hayes, 2013), indicating a significant indirect effect of childhood trauma on
alcohol misuse through working memory performance. Statistical significance was set at p < .05 for all analyses.
3. RESULTS
3.1. Descriptive Statistics
Data from one subject in the sample could not be evaluated because of missing CTQ and AUDIT data, which resulted in a final sample of 76 subjects. The participants were primarily female (63.2%, n = 48). The sample’s age ranged from 18 to 21 years old (M = 19.30, SD = 1.108), and males (M = 19.68, SD = 1.156) were significantly older than females (M = 19.08, SD = 1.028), t (75) = 2.265, p < .05. The majority of the participants were Dutch (79%, n = 60) and the rest (21%, n = 16) were German. Based on the CTQ cut-off scores to identify at least low maltreatment severity, the prevalence of childhood maltreatment in the sample was found high (68.42%). Of the participants who experienced maltreatment, 36.54% of them were exposed to two or more types of childhood trauma. Among the five types of maltreatment, emotional abuse and neglect were most frequent with 27.6% and 44.7% respectively.
Table 1 shows descriptive statistics of the main variables. Although childhood maltreatment in the sample was found highly prevalent, the mean scores for all types of trauma were in low level of severity. However, in each subscale there were also subjects who reported experiencing severe child abuse or neglect. As a result, these high scores were found as outliers (> 2.58 SD). Similar with the CTQ score, the AUDIT mean score was relatively low (M = 7.49, SD = 4.59), classified by WHO guidelines (2001) as low risk alcohol problems. However, the prevalence of alcohol misuse based on the cut-off score in the sample was 44.7% (n = 34). Three outliers were detected for subjects who scored very high levels of alcohol misuse (AUDIT score ³ 20). Because there were more low than high scores in the sample, childhood maltreatment and alcohol misuse variables were skewed to the right with skewness = 1.11 (SD = .276) and .869 (SD = .276) respectively. On the other hand, working memory performance variable was negatively skewed (-.868, SD = .274), with an average accuracy score of 49.33 (SD = 18.40). Two subjects showed distinct low scores (-9.38 and -18.75) and were considered as outliers. Nevertheless, after close examination of all outliers from the three variables, it was decided that the values were still acceptable to represent observations in the population and there was no compelling reason to exclude them from analyses.
Table 2 presents bivariate correlations results between all main and demographic variables. Spearman’s correlation was used because data were not normally distributed. Cumulative experience of childhood maltreatment (CTQ total score) was positively correlated with alcohol misuse (r = .30, p < .01). More specific, drinking alcohol was found positively
Table 1. Descriptive statistics of the main measures
Notes. CTQ = Childhood Trauma Questionnaire; AUDIT = Alcohol Use Disorders Identification Test; WM Performance = Working Memory Performance; All values are raw, unstandardized scores.
correlated with emotional neglect (r = .34, p < .01). Working memory performance was negatively correlated with sexual abuse (r = -.25, p < .05), but not with alcohol use (r = .01, p > .05). Finally, men reported more emotional neglect (r = .29, p < .05) and more alcohol use (r = .63, p < .01) than women, while older respondents reported higher levels of emotional neglect (r = .26, p < .05).
3.2. Assumptions
There were several assumptions checked before the regression analyses. Linearity was examined using scatterplots. From visual inspection, linearity was not violated. Assumption of normality of residuals was also checked using the Kolmogorov-Smirnov test and was met. Levene’s test result for homoscedasticity was non-significant so this assumption was satisfied. Lastly, the independence assumption was also met with a Durbin-Watson test score of 1.886 which was considered acceptable within the range of 1.5 to 2.5.
Yes (%) No (%) Mean SD Minimum Maximum Emotional Abuse 21 (27.6) 55 (72.4) 7.64 2.86 5.00 17.00 Physical Abuse 3 (3.9) 73 (96.1) 5.36 1.17 5.00 12.00 Sexual Abuse 12 (15.8) 64 (84.2) 5.46 1.46 4.00 13.00 Emotional Neglect 34 (44.7) 42 (55.3) 9.43 3.12 5.00 19.00 Physical Neglect 9 (11.8) 67 (88.2) 6.20 2.12 5.00 14.00 Total CTQ Score 34.09 6.44 25.00 52.00 AUDIT 7.49 4.59 0.00 22.00 WM Performance 49.33 18.40 -18.75 90.63
Table 2. Correlations between main variables and demographic variables
Notes. * p < .05 (2-tailed); ** p < .01 (2-tailed); Emotional abuse, physical abuse, sexual abuse, emotional neglect and physical neglect are subscales in the Childhood Trauma Questionnaire (CTQ); AUDIT = Alcohol Use Disorders Identification Test; WM Performance = Working Memory Performance; Gender coded as 0 = male, 1 = female; Values are unstandardized scores. 3.3. Multiple Linear Regression Analysis
Multiple regression was used to test childhood maltreatment as predictor of alcohol misuse. Table 3 shows the final results of the regression analysis. Consistent with the hypothesis, more experience of early life adversities predicted higher risk of alcohol misuse. Holding age and gender constant, there was a statistically significant and positive partial association between childhood maltreatment and drinking problems, with B = .223, t (72) = 3.514, p = .001. Thus, the null hypothesis could be rejected. Furthermore, partial correlation showed a medium effect with r = .383, which means of the variance in alcohol misuse that was not explained by age and gender, 100(.3832)% = 14.67% of it was explained by childhood maltreatment. From the analysis, gender was found as a significant predictor of alcohol misuse (coded as 0 = male, 1 = female), with B = -5.487, t (72) = -6.342, p = .000. 1 2 3 4 5 6 7 8 9 1 Age 2 Gender -.25* 3 Emotional Abuse -.09 .05 4 Physical Abuse .01 .02 .24* 5 Sexual Abuse -.09 .10 .04 .28* 6 Emotional Neglect .26* -.29* .49** .01 -.05 7 Physical Neglect .17 .12 .30** .06 -.04 .30** 8 CTQ Total Score .15 -.07 .73** .27* .20 .78** .58** 9 AUDIT .18 -.63** .15 -.02 .07 .34** .11 .30** 10 WM Performance -.09 -.10 -.01 -.08 -.25* -.11 -.02 -.08 .01
Notes. CTQ = Childhood Trauma Questionnaire; CI = Confidence Interval; Values are unstandardized scores.
3.4. Mediation Analysis
Simple mediation analysis was performed to examine the effect of childhood maltreatment on alcohol misuse through working memory performance. Results are summarized in Figure 1. Table 3. Multiple regression of the correlation between childhood trauma and alcohol misuse
B Std. Error t p 95% CI Lower Upper (Constant) 8.239 7.604 1.083 .282 -6.920 23.397 CTQ Total Score .223 .063 3.514 .001 .097 .350 Age -.253 .382 -.664 .509 -1.015 .508 Gender -5.487 .865 -6.342 .000 -7.212 -3.763 X Y M C2 C1 Childhood Maltreatment Working Memory Alcohol Misuse Age Gender a = -.296 (.336) b = -.019 (.022) c' = .218 (.064)*
Figure 1. Simple mediation model for working memory influence on the relationship between childhood maltreatment and alcohol misuse; Values show coefficients and standard error of unstandardized Betas; * p < .05
Contrary to the hypothesis, the indirect effect in this model was not statistically different from zero, with 95% bootstrap confidence interval [-.012, .063], ab = .006, SE = .019. Thus, working memory performance did not mediate the association between child maltreatment and alcohol
misuse. However, a direct effect was found for child maltreatment on drinking problem, c’= .218, t = 3.402, p = .001, 95% CI [.090, .345], which means that individuals who experience
more childhood maltreatment but are equal in working memory capacity are estimated to have a higher risk of abusing alcohol by .218 units. Meanwhile, analysis did not show any significant effect of working memory performance, either as consequence of maltreatment (path a, B = -.296, p = .382) or antecedent of alcohol abuse (path b, B = -.019, p = .407) in the mediation model.
4. DISCUSSION
The aim of the present study was twofold. First, it was examined if childhood maltreatment associated with alcohol misuse in a nonclinical, student population. Next, we investigated a possible mechanism of this association by examining working memory performance as a mediator. We hypothesized that experiencing childhood trauma would positively correlate with drinking problems. We also presumed that working memory capacity would mediate the relationship between childhood maltreatment and alcohol misuse, i.e., experiencing early life adversities would predict lower levels of working memory performance, whereby increasing the level of harmful drinking.
The findings in this study supported one of the two hypotheses. We found that exposure to childhood maltreatment was linked to a higher level of at-risk drinking after demographic covariates were controlled for, with a medium effect size (r = .383). However, the results of the mediation analysis revealed that working memory performance did not mediate the relationship between early life trauma and alcohol misuse. Thus, contrary to the hypothesis, in this healthy, highly educated sample, working memory performance was not found to play a role in the underlying mechanism of the significant association between childhood maltreatment and hazardous drinking.
Unlike studies in clinical populations, the findings from previous studies about the effect of childhood trauma on alcohol misuse using healthy samples were mixed and inconclusive (Schwandt et al., 2013, Shin et al., 2009). The result of the present study, for example, contradicted with the outcome of a study by Schwandt et al. (2013) which did not find an
association between maltreatment and drinking problem in their healthy control group. The difference could possibly be the result of different sampling method. The healthy sample characteristics in the study by Schwandt et al., (2013) were more diverse in races, education, and age (ranged from 20-36 years old), to compare with current study of a student sample which were more homogeneous demographically. However, the fact that the current study discovered that early life adversities predicted alcohol misuse supports the general notion of the negative health consequences following childhood trauma, which include alcohol problems (Huang et al., 2012). More general, childhood maltreatment has been associated with various problems in individual’s psychological and physical development which persist into adulthood (Schwandt et al., 2013), making them more vulnerable to developing psychiatric disorders as well as alcohol problems (Fuge et al., 2014, Huang et al., 2012). The current study contributes to the existing literature by demonstrating that the effect of childhood maltreatment on increasing the level of alcohol misuse is present not only in clinical populations, but also among healthy and highly educated individuals with relatively low levels of maltreatment severity.
When studying the types of trauma separately, bivariate correlation analysis only found a significant association between alcohol misuse and one type of maltreatment, i.e., emotional neglect. This could possibly be because emotional neglect was the most prevalent type of maltreatment in the sample. The result was consistent with a previous finding that revealed emotional maltreatment as the most frequent childhood trauma found in alcohol dependence subjects (Schwandt et al., 2013). On the other hand, correlations between alcohol misuse and the other single types of trauma were not found, except when the maltreatment was measured cumulatively (total CTQ score). This finding confirmed prior studies that suggested assessing the cumulative effects from various types of maltreatment rather than single trauma as a better predictor to measure adverse outcomes (Goldstein et al., 2010, Schwandt et al., 2013). Besides, numerous studies found that the majority of maltreated children were exposed to multiple forms of maltreatment (Bernstein et al., 2003, Huang et al., 2012), and there was a dose-response relationship observed between the number of maltreatment and the risk of psychiatric comorbidities (Huang et al., 2012). Thus, multiple types of abuse often co-occur and yield more severe effects than a single type. Likewise, polyvictimization was also found in the current study, with approximately four out of every ten victims reported experiencing more than one type of maltreatment. Thus, it appeared that the association between childhood maltreatment and alcohol misuse was found significant in this study because the childhood trauma severity
was evaluated cumulatively from various types of maltreatment. Without assessing the total score of all maltreatment categories, it is possible we would have been missing important data. Earlier studies have proposed theories to explain the mechanism underlying the association between childhood trauma and its adverse consequences. One of the possible mechanisms is related to alterations in the brain’s structure and function. As described in the theoretical framework, early life stress could produce changes in the individual’s cognitive functioning, which is often found in many psychiatric disorders, including addiction (Anda et al., 2006, Andersen, 2019, Enoch, 2011). Problems in working memory, which is a primary component of executive cognitive function, has been found associated with childhood trauma (Goodman et al., 2019, Pechtel & Pizzagalli, 2010), and also with difficulty in behaviour control which often occurs in addiction (Ellingson et al., 2014, Tahaney & Palfai, 2018, Day et al., 2013). However, the result in the current study did not support this notion. Working memory performance did not mediate the effect of childhood adversities on alcohol misuse. In contrast with previous research, we did not find an association between experiencing childhood maltreatment and lower working memory performance, nor that working memory capacity predicted the risk of alcohol misuse.
The non-result of the mediation analysis could be due to the inclusion criteria with respect to childhood maltreatment. Prior research showing a significant effect of abuse on working memory impairments only included participants who met at least a moderate severity level of childhood trauma (Fuge et al., 2014, Philip et al., 2016). In this study, we evaluated every participant including those with none to a minimal level of maltreatment, and it was shown that the majority of the sample only experienced a low level of maltreatment. In addition, in a study by Fuge et al. (2014), they found that working memory impairment was only found in individuals with moderate to severe early life stress, whereas lower levels of severity showed no difference in working memory performance. Thus, it was likely that the result of the current study are in line with those studies, and the null finding was present because there were not many individuals in the sample who were severely maltreated, whereas those not severely abused were not affected cognitively. However, since the participants were university students, it was also possible that we had restrictions of range with only included high IQ subjects, and it might have influenced the working memory performance outcome. Nevertheless, as the aim of this study was to get an overview of the conditions in a healthy population, it is possible that
the low severity level of trauma in this sample provided reliable population estimates of healthy individuals, and working memory deficits’ effect did not exist in this healthy sample.
We also did not find an association between working memory performance and alcohol misuse. The result was opposed to a body of evidence which demonstrated the effect of working memory deficits on substance use, including alcohol misuse (Ellingson et al., 2014, Tahaney & Palfai, 2018, Day et al., 2013). However, the current outcome was in line with one prospective study by Peeters at al. (2014) using an adolescent sample. In this study, they found that although working memory performance could predict alcohol misuse prospectively, the relation did not exist when the variables were studied concurrently (Peeters et al., 2014). The possible explanation could be that the effect of working memory performance on alcohol misuse was not direct, but was mediated by other variables such as impulsivity, as revealed from a study by Khurana et al., (2012). Hence, we still cannot rule out the possibility that the effect of working memory performance on alcohol misuse simply did not exist. Thus, although a significant association was found between childhood maltreatment and alcohol misuse, another mechanism that is not related to the individual’s working memory capacity may account for that association. Previous studies of mediators have revealed other variables that could explain the underlying mechanism such as negative post-traumatic cognitions (Banerjee et al., 2018), impulsiveness subfacet of neuroticism (Schwandt et al., 2013), and post-traumatic stress disorder (Cross et al., 2015) to name a few, but these were not the focus in the present study. 4. 1. Strengths of the study
The present study was among the first, to our knowledge, to examine working memory performance as a mediator in the association between childhood maltreatment and alcohol misuse. The results did not support the hypothesis, as we found that in this healthy sample, working memory performance did not act as an underlying mechanism of predicting at-risk drinking after childhood maltreatment. It is possible that the association was moderated by the level of maltreatment severity (Fuge et al., 2014; Pechtel & Pizzagalli, 2010). As such, current research helped to contribute to the existing literature by giving a better understanding of the role of early life adversities in predicting working memory impairments.
Furthermore, we used the Childhood Trauma Questionnaire to measure maltreatment in the sample. The instrument allows assessment of five types of maltreatment rather than a single form of childhood trauma (Bernstein et al., 2003, Goldstein et al., 2010, Jasinski et al., 2000),
using a continuous severity score which is a better predictor than dichotomous measures (presence or absence of trauma) (Huang et al., 2012, Schwandt et al., 2013). The total childhood maltreatment score also creates dimensional scales that maximize reliability and statistical power (Bernstein et al., 2003, Huang et al., 2012). The other instruments have also shown good reliability and validity. In addition, because we used a computer program for the assessment, all the data were recorded automatically, thus could ensure the patient’s confidentiality and minimize errors from manual input.
4.2. Limitations and Further Research
The study has several limitations. First, the cross-sectional analysis limits causal interpretation. Causality could not be inferred because all of the data were collected at the same time. Further research using prospective methods could solve the issue. Next to that, variables in this study were examined using self-report questionnaires on childhood maltreatment history and drinking habits. Although a standard method, the assessment is still vulnerable to recall bias. As the history of maltreatment was measured retrospectively, there is a potential problem with the data accuracy due to degradation of memories or pathological reasons that makes the trauma underreported (Bernstein et al., 2003). In addition, social desirability might also occur when reporting behaviours considered socially inappropriate such as alcohol use (Cross et al., 2015, Reinert et al., 2007). Accordingly, future research might consider combining the questionnaires with other sources of information such as alcohol biomarkers to screen alcohol use, and interviews with family members or informants to support data collection of childhood maltreatment.
The other limitation is regarding the sample characteristic. Because the participants in the sample were all university students, it limits the generalizability of the result to the non-university populations. Moreover, education has been shown to be associated with alcohol use (Grenard et al., 2008). It is possible with different levels of education in the sample, the study would yield a different result. Replication with a sample that includes different education levels will help to address the issue.
Next, we could not eliminate the possibility of other factors not examined in this study were related to the outcomes. For example, we did not have data on participant’s current life stressors or trauma, family history of alcoholism, genetic variation, and psychiatric comorbidity, which all have been found associated with alcohol use (Schwandt et al., 2013, De
Martini et al., 2012, Jasinski et al., 2000). Future research might need to account for these variables as potential confounders.
We did not observe working memory performance mediating the association between childhood trauma and alcohol misuse. Based on prior studies, it was likely that the levels of childhood trauma severity moderated the effects of childhood trauma to working memory capacity (Fuge et al., 2014, Philip et al., 2016). These studies, however, were still at infancy. Thus, future studies should further explore the adverse consequences of different levels of childhood trauma severity to cognitive impairments such as working memory capacity, and how it associates to vulnerability of various psychopathologies.
4.3. Implications and Conclusions
From the present study, we observed that higher rates of childhood trauma associated with alcohol misuse in a student population. The result indicates that co-occurrence is common. Thus, it is of importance for health practitioners to include screening tests for both trauma history and alcohol use problems in routine medical appointments, especially when one of the conditions exists. Besides, with the fact that alcoholics have low rates of treatment-seeking (De Martini et al., 2012, Schwandt et al., 2013), screening tests can increase earlier intervention and access to treatment. Tailored treatment which considers the long term effects of childhood maltreatment will also be beneficial for individuals with drinking problems to attain better treatment outcomes, as comorbidities likely result in poorer outcomes when it is not handled (Debell et al., 2014).
These study results add to a body of evidence that have shown adverse consequences following childhood maltreatment. Trauma creates a profound and lasting influence that affects individuals’ quality of life. Victims are reported more likely to suffer from mental and physical health problems (McEwen et al., 2008), which, as demonstrated in this study, includes alcohol problems. With the consideration that prevention is better than cure, early interventions targeting child’s environment to reduce maltreatment can be expected to minimize problems which may arise later in life. Prevention measures can be taken, such as home visits by social workers, increasing parenting skills, and developing social support to create a healthy environment for the children development.
To summarize, the current study found that childhood maltreatment predicted alcohol misuse in a healthy sample using student population. We propose several measures for
treatment and prevention with regard to the interplay of childhood trauma and drinking problems. On the other hand, we did not find working memory performance as the mediator in the underlying mechanism of the association between childhood trauma and alcohol misuse, which indicates other mechanisms not examined may account for that association. Given that the childhood maltreatment prevalence is found high and correlates with alcohol misuse and other adverse consequences, further studies addressing the possible underlying mechanism in the healthy population is still much needed to advance current treatment and prevention for alcohol misuse.
REFERENCES
Anda, R. F., Felitti, V. J., Bremner, J. D., Walker, J. D., Whitfield, C., Perry, B. D., … Giles, W. H. (2006). The enduring effects of abuse and related adverse experiences in childhood: A convergence of evidence from neurobiology and epidemiology. European Archives of Psychiatry and Clinical Neuroscience, 256(3), 174–186. https://doi.org/10.1007/s00406-005-0624-4
Andersen, S. L. (2019). Stress, sensitive periods, and substance abuse. Neurobiology of Stress, 10 (August). https://doi.org/10.1016/j.ynstr.2018.100140
Baddeley, A., Jarrold, C., & Vargha-Khadem, F. (2011). Working memory and the hippocampus. Journal of Cognitive Neuroscience, 23(12), 3855–3861. https://doi.org/10.1162/jocn_a_00066
Banerjee, N., Ironson, G., Fitch, C., Boroughs, M. S., Safren, S. A., Powell, A., & O’Cleirigh, C. (2018). The Indirect Effect of Posttraumatic Stress Disorder Symptoms on Current Alcohol Use Through Negative Cognitions in Sexual Minority Men. Journal of Traumatic Stress, 31(4), 602–612. https://doi.org/10.1002/jts.22304
Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., … Zule, W. (2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse and Neglect, 27(2), 169–190. https://doi.org/10.1016/S0145-2134(02)00541-0
Bradley, K. A., Bush, K. R., Epler, A. J., Dobie, D. J., Davis, T. M., Sporleder, J. L., … Kivlahan, D. R. (2003). Two brief alcohol-screening tests from the Alcohol Use Disorders Identification Test (AUDIT): Validation in a female Veterans Affairs patient population.
Archives of Internal Medicine, 163(7), 821–829.
Clark, D. B., Lesnick, L., & Hegedus, A. M. (1997). Traumas and other adverse life events in adolescents with alcohol abuse and dependence. Journal of the American Academy of Child and Adolescent Psychiatry, 36(12), 1744–1751. https://doi.org/10.1097/00004583-199712000-00023
Cowell, R. A., Cicchetti, D., Rogosch, F. A., & Toth, S. L. (2015). Childhood maltreatment and its effect on neurocognitive functioning: Timing and chronicity matter. Development and Psychopathology, 27(2), 521–533. https://doi.org/10.1017/S0954579415000139
Cross, D., Crow, T., Powers, A., & Bradley, B. (2015). Childhood trauma, PTSD, and problematic alcohol and substance use in low-income, African-American men and women. Child Abuse and Neglect, 44, 26–35. https://doi.org/10.1016/j.chiabu.2015.01.007 Day, A. M., Metrik, J., Spillane, N. S., & Kahler, C. W. (2013). Working memory and
impulsivity predict marijuana-related problems among frequent users. Drug and Alcohol Dependence, 131(1–3), 171–174. https://doi.org/10.1016/j.drugalcdep.2012.12.016 Debell, F., Fear, N. T., Head, M., Batt-Rawden, S., Greenberg, N., Wessely, S., & Goodwin,
L. (2014). A systematic review of the comorbidity between PTSD and alcohol misuse. Social Psychiatry and Psychiatric Epidemiology, 49(9), 1401–1425. https://doi.org/10.1007/s00127-014-0855-7
DeMartini, K. S., & Carey, K. B. (2012). Optimizing the use of the AUDIT for alcohol screening in college students. Psychological Assessment, 24(4), 954–963. https://doi.org/10.1037/a0028519
Dube, S. R., Anda, R. F., Felitti, V. J., Edwards, V. J., & Croft, J. B. (2002). Adverse childhood experiences and personal alcohol abuse as an adult. Addictive Behaviors, 27(5), 713–725. https://doi.org/10.1016/S0306-4603(01)00204-0
Ellingson, J. M., Fleming, K. A., Vergés, A., Bartholow, B. D., & Sher, K. J. (2014). Working memory as a moderator of impulsivity and alcohol involvement: Testing the cognitive-motivational theory of alcohol use with prospective and working memory updating data. Addictive Behaviors, 39(11), 1622–1631. https://doi.org/10.1016/j.addbeh.2014.01.004
Enoch, M. A. (2011). The role of early life stress as a predictor for alcohol and drug dependence. Psychopharmacology, 214(1), 17–31. https://doi.org/10.1007/s00213-010-1916-6
Evren, C., Kural, S., & Cakmak, D. (2006). Clinical correlates of childhood abuse and neglect in substance dependents. Addictive Behaviors, 31(3), 475–485. https://doi.org/10.1016/j.addbeh.2005.05.030
Fuge, P., Aust, S., Fan, Y., Weigand, A., Gärtner, M., Feeser, M., … Grimm, S. (2014). Interaction of early life stress and corticotropin-releasing hormone receptor gene: Effects on working memory. Biological Psychiatry, 76(11), 888–894. https://doi.org/10.1016/j.biopsych.2014.04.016
Goldstein, A. L., Flett, G. L., & Wekerle, C. (2010). Child maltreatment, alcohol use and drinking consequences among male and female college students: An examination of drinking motives as mediators. Addictive Behaviors, 35(6), 636–639. https://doi.org/10.1016/j.addbeh.2010.02.002
Goodman, J.B., Freeman, E.E., Chalmers, K.A. (2019). The relationship between early life stress and working memory in adulthood: A systematic review and meta-analysis, Memory, 27:6, 868-880, DOI: 10.1080/09658211.2018.1561897
Grenard, J. L., Ames, S. L., Wiers, R. W., Thush, C., Sussman, S., & Stacy, A. W. (2008). Working Memory Capacity Moderates the Predictive Effects of Drug-Related Associations on Substance Use. Psychology of Addictive Behaviors, 22(3), 426–432. https://doi.org/10.1037/0893-164X.22.3.426
Gunn, R. L., Gerst, K. R., Wiemers, E. A., Redick, T. S., & Finn, P. R. (2018). Predictors of Effective Working Memory Training in Individuals with Alcohol Use Disorders. Alcoholism: Clinical and Experimental Research, 42(12), 2432–2441. https://doi.org/10.1111/acer.13892
Gunn, R. L., Gerst, K. R., Lake, A. J., & Finn, P. R. (2018). The effects of working memory load and attention refocusing on delay discounting rates in alcohol use disorder with comorbid antisocial personality disorder. Alcohol, 66, 9–14. https://doi.org/10.1016/j.alcohol.2017.07.009
Hayes, A. F. (2013). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. London: The Guilford Press
Hendershot, C. S., Wardell, J. D., Vandervoort, J., McPhee, M. D., Keough, M. T., & Quilty, L. C. (2018). Randomized Trial of Working Memory Training as an Adjunct to Inpatient Substance Use Disorder Treatment. Psychology of Addictive Behaviors, 32(8), 861–872. https://doi.org/10.1037/adb0000415
Houben, K., Wiers, R. W., & Jansen, A. (2011). Getting a Grip on Drinking Behavior: Training Working Memory to Reduce Alcohol Abuse. Psychological Science, 22(7), 968–975. https://doi.org/10.1177/0956797611412392
Huang, M. C., Schwandt, M. L., Ramchandani, V. A., George, D. T., & Heilig, M. (2012). Impact of Multiple Types of Childhood Trauma Exposure on Risk of Psychiatric Comorbidity Among Alcoholic Inpatients. Alcoholism: Clinical and Experimental Research, 36(6), 598–606. https://doi.org/10.1111/j.1530-0277.2011.01695.x
Jaeggi, S. M., Studer-Luethi, B., Buschkuehl, M., Su, Y. F., Jonides, J., & Perrig, W. J. (2010). The relationship between n-back performance and matrix reasoning - implications for
training and transfer. Intelligence, 38(6), 625–635. https://doi.org/10.1016/j.intell.2010.09.001
Jasinski, J. L., Williams, L. M., & Siegel, J. (2000). Childhood physical and sexual abuse as risk factors for heavy drinking among African-American women: A prospective study. Child Abuse and Neglect, 24(8), 1061–1071. https://doi.org/10.1016/S0145-2134(00)00158-7
Khurana, A., Romer, D., Betancourt, L. M., Brodsky, N. L., Giannetta, J. M., & Hurt, H. (2012). Working memory ability predicts trajectories of early alcohol use in adolescents: The mediational role of impulsivity. Addiction, 108(3), 506–515. https://doi.org/10.1111/add.12001
Magnusson, Å., Lundholm, C., Göransson, M., Copeland, W., Heilig, M., & Pedersen, N. L. (2012). Familial influence and childhood trauma in female alcoholism. Psychological Medicine, 42(2), 381–389. https://doi.org/10.1017/S0033291711001310
McEwen, B. S. (2008). Understanding the potency of stressful early life experiences on brain and body function. Metabolism: Clinical and Experimental, 57(SUPL.2), 11–15. https://doi.org/10.1016/j.metabol.2008.07.006
Mirsal, H., Kalyoncu, A., Pektaş, Ö., Tan, D., & Beyazyürek, M. (2004). Childhood trauma in alcoholics. Alcohol and Alcoholism, 39(2), 126–129. https://doi.org/10.1093/alcalc/agh025
Morrison, A. B., & Chein, J. M. (2011). Does working memory training work? the promise and challenges of enhancing cognition by training working memory. Psychonomic Bulletin and Review, 18(1), 46–60. https://doi.org/10.3758/s13423-010-0034-0
Pechtel, P., & Pizzagalli, D. A. (2011). Effects of early life stress on cognitive and affective function: An integrated review of human literature. Psychopharmacology, 214(1), 55–70. https://doi.org/10.1007/s00213-010-2009-2
Peeters, M., Janssen, T., Monshouwer, K., Boendermaker, W., Pronk, T., Wiers, R., & Vollebergh, W. (2015). Weaknesses in executive functioning predict the initiating of adolescents’ alcohol use. Developmental Cognitive Neuroscience, 16, 139–146. https://doi.org/10.1016/j.dcn.2015.04.003
Philip, N. S., Sweet, L. H., Tyrka, A. R., Carpenter, S. L., Albright, S. E., Price, L. H., & Carpenter, L. L. (2016). Exposure to childhood trauma is associated with altered n-back activation and performance in healthy adults: implications for a commonly used working memory task. Brain Imaging and Behavior, 10(1), 124–135. https://doi.org/10.1007/s11682-015-9373-9
Reinert, D. F., & Allen, J. P. (2007). The alcohol use disorders identification test: An update of research findings. Alcoholism: Clinical and Experimental Research, 31(2), 185–199. https://doi.org/10.1111/j.1530-0277.2006.00295.x
Rothman, E. F., Edwards, E. M., Heeren, T., & Hingson, R. W. (2008). Adverse childhood experiences predict earlier age of drinking onset: Results from a representative US sample of current or former drinkers. Pediatrics, 122(2). https://doi.org/10.1542/peds.2007-3412
Saunders, J.B., Aasland, O. G., Babor, T. F., Fuente, J. R., Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful Alcohol Consumption-II. Addiction. 88, 791-804 Schwandt, M. L., Heilig, M., Hommer, D. W., George, D. T., & Ramchandani, V. A. (2013). Childhood Trauma Exposure and Alcohol Dependence Severity in Adulthood: Mediation by Emotional Abuse Severity and Neuroticism. Alcoholism: Clinical and Experimental Research, 37(6), 984–992. https://doi.org/10.1111/acer.12053
Sher, K. J., Talley, A. E., Littlefield, A. K., & Martinez, J. A. (2012). Alcohol Use and Alcohol Use Disorders. In The Oxford Handbook of Health Psychology. https://doi.org/10.1093/oxfordhb/9780195342819.013.0028
Shin, S. H., Edwards, E. M., & Heeren, T. (2009). Child abuse and neglect: Relations to adolescent binge drinking in the national longitudinal study of Adolescent Health (AddHealth) Study. Addictive Behaviors, 34(3), 277–280. https://doi.org/10.1016/j.addbeh.2008.10.023
Tahaney, K. D., & Palfai, T. P. (2018). Working memory moderates the association between perceived norms and heavy episodic drinking among college students. Addictive Behaviors, 81(September 2017), 46–49. https://doi.org/10.1016/j.addbeh.2018.01.011 Thombs, B. D., Bernstein, D. P., Lobbestael, J., & Arntz, A. (2009). A validation study of the
Dutch Childhood Trauma Questionnaire-Short Form: Factor structure, reliability, and known-groups validity. Child Abuse and Neglect, 33(8), 518–523. https://doi.org/10.1016/j.chiabu.2009.03.001
Trautmann, S., Kräplin, A., Dieterich, R., Richter, J., & Muehlhan, M. (2018). The role of childhood trauma and stress reactivity for increased alcohol craving after induced psychological trauma: an experimental analogue study. Psychopharmacology, 235(10), 2883–2895. https://doi.org/10.1007/s00213-018-4979-4
Widom, C. S., Hiller-Sturmhofel, S.H. (2001). Alcohol Abuse as a Risk Factor for and Consequence of Child Abuse. Alcohol Research and Health. Vol. 25 No. 1, 52-57
World Health Organization. (2018). Global status report on alcohol and health 2018. In Global status report on alcohol. Retrieved from:
https://www.who.int/substance_abuse/publications/global_alcohol_report/en/
World Health Organization. (2014). Global status report on violence prevention 2014. In Global status report on violence prevention. Retrieved from:
https://www.who.int/violence_injury_prevention/violence/status_report/2014/en/
World Health Organization. (2001). AUDIT The Alcohol Use Disorders Identification Test Guidelines for Use in Primary Care. Retrieved from:
https://www.who.int/publications-detail/audit-the-alcohol-use-disorders-identification-test-guidelines-for-use-in-primary-health-care
Young-Wolff, K. C., Kendler, K. S., Ericson, M. L., & Prescott, C. A. (2011). Accounting for the association between childhood maltreatment and alcohol-use disorders in males: A twin study. Psychological Medicine, 41(1), 59–70. https://doi.org/10.1017/S0033291710000425
