- Email: [email protected]
Depression symptoms and reasons for gambling sequentially mediate the associations between insecure attachment styles and problem gambling
Addictive Behaviors 78 (2018) 166–172
Contents lists available at ScienceDirect
Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh
Depression symptoms and reasons for gambling sequentially mediate the associations between insecure attachment styles and problem gambling
T
⁎
Matthew T. Keougha, , Trinda L. Pennistona, Natalie Vilhena-Churchillb, R. Michael Bagbyc, Lena C. Quiltyd,e a
University of Manitoba, Department of Psychology, 190 Dysart Road, P314 Duff Roblin Building, Winnipeg, MB R3T 2N2, Canada Altum Health, University Health Network, Krembil Discovery Tower, 399, Bathurst St., Toronto, Ontario M5T 2S8, Canada c University of Toronto, Departments of Psychology and Psychiatry, 1265 Military Trail, Toronto, Ontario M1C 1A4, Canada d Centre for Addiction and Mental Health, Campbell Family Mental Health Research Institute, 100 Stokes Street, Bell Gateway Building, Toronto, Ontario M6J 1H4, Canada e University of Toronto, Department of Psychiatry, 250 College St., Toronto, Ontario M5T 1R8, Canada b
H I G H L I G H T S test the associations between insecure attachment styles and gambling problems. • We clinical sample with mood disorders was used. • ADepression and coping motives mediated attachment-gambling associations. • Insecure attachment increases problem gambling risk in adults with mood disorders. •
A R T I C L E I N F O
A B S T R A C T
Keywords: Insecure attachment Depressive symptoms Gambling motives Mood disorders
One of the central pathways to problem gambling (PG) is gambling to cope with negative moods, which is a cardinal feature of depression. Insecure attachment styles are also etiologically related to depression; and, therefore, by extension, those who are insecurely attached may engage in excessive gambling behaviors to cope with depression. In this study, we aimed to evaluate this and to this end predicted that depression severity and coping motives for gambling would conjointly mediate the relations between insecure attachment styles and PG. Data came from a larger investigation of PG within mood disorders. Participants exhibited a lifetime depressive or bipolar disorder and endorsed a mood episode within the past ten years. Participants (N = 275) completed self-report measures during a two-day assessment. Path analysis supported two main indirect effects. First, anxious attachment predicted elevated depression, which in turn predicted increased coping motives for gambling, which subsequently predicted greater PG severity. Second, this double mediational pathway was also observed for avoidant attachment. Results suggest that insecure attachment relates to PG via depressive symptoms and coping-related gambling motives. Mood symptoms and associated gambling motives are malleable and are promising targets of gambling interventions for insecurely attached individuals.
1. Introduction
Dowling et al., 2014; Maniaci et al., 2015; Nehlin et al., 2013; Odlaug, Schreiber, & Grant, 2013), and high suicidality (Petry, Stinson, & Grant, 2005; Suissa, 2011). Clinical interventions would benefit from a better understanding of the risk factors that contribute to PG. Blaszczynski and Nower (2002) identified three distinct pathways to PG, including (1) behaviorally conditioned problem gamblers, (2) emotionally vulnerable problem gamblers, and (3) impulsivist/antisocial problem gamblers. Each pathway is believed to unfold over the course of development and each is thought to carry diverse but unique risk
About 2–3% of North Americans gamble at harmful levels (Kessler et al., 2008; Williams, Volberg, & Stevens, 2012). This rate more than triples in psychiatric populations (e.g., people with mood disorders) (Getty, Watson, & Frisch, 2000; Nehlin, Grönbladh, Fredriksson, & Jansson, 2013). Problem gambling (PG) is associated with many negative outcomes, like financial distress, high rates of addiction, poor mental and physical health (Afifi, Cox, Martens, Sareen, & Enns, 2010;
⁎
Corresponding author. E-mail address: [email protected] (M.T. Keough).
https://doi.org/10.1016/j.addbeh.2017.11.018 Received 19 July 2017; Received in revised form 9 November 2017; Accepted 9 November 2017 Available online 14 November 2017 0306-4603/ © 2017 Elsevier Ltd. All rights reserved.
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
Emotionally vulnerable gamblers are people who struggle with elevated levels of negative affect and emotional lability. Consequently, they have ample opportunities to learn that gambling has negatively reinforcing effects on mood (Blaszczynski & Nower, 2002; Keough, Wardell, Hendershot, Bagby, & Quilty, 2017). Through repeated experience, the association between dysregulated mood and gambling (as a coping strategy) becomes strengthened, resulting in the development of PG. In the current study, we used a complementary theoretical framework – attachment theory (Ainsworth & Bowlby, 1991) – to provide new insight on the individual differences and mechanisms of the emotionally vulnerable pathway to PG among people with mood disorders.
The insecurely attached person is thought to engage in maladaptive coping strategies, such as addictive behaviors, to reduce high levels of distress. Insecure attachment styles have been positively linked with alcohol (McNally, Palfai, Levine, & Moore, 2003) and substance use (Kassel, Wardle, & Roberts, 2007; Thorberg & Lyvers, 2010); surprisingly few studies have examined the risk impact of insecure attachment styles for dysfunctional gambling behavior. Recent work suggests that insecure attachment positively correlates with problem gambling (Di Trani, Renzi, Vari, Zavattini, & Solano, 2017; Testa et al., 2017a, 2017b). Moreover, some studies have examined the link between poor parental bonding (a correlate or sequelae of attachment) and PG, supporting the notion that poor parental attachment predicts increased adolescent engagement in gambling behaviors (Magoon & Ingersoll, 2006). Poor parental attachment is also related to the development of depressive symptoms (Demidenko, Manion, & Lee, 2015; Otowa, York, Gardner, Kendler, & Hettema, 2014); it is also possible then that depressed mood somehow plays a role in the association between insecure attachment and risk for problem gambling. Despite the potential causal relations between depressed mood, insecure attachment and PG, the interrelations of these factors has yet to be examined explicitly in pathway modeling.
1.2. Adult attachment, emotional dysregulation, and problem gambling
1.3. Role of gambling motives
Attachment is the tendency for us to form emotional bonds with other humans. It is believed to be biologically rooted and begins with infant-caregiver interactions during the initial years of life. Attachment theory (Ainsworth & Bowlby, 1991) states that early bonding with a caregiver is important for the formation of internal working models of communication, emotion regulation and coping, and interpersonal functioning that are carried through to adulthood. These internal models are believed to be stable across development and the same theories explaining parent-child bonds can also be applied to understand adult interpersonal relationships and associated developmental outcomes (Ainsworth & Bowlby, 1991; Hazan & Shaver, 1987). There are three general attachment styles categorized into secure and insecure (anxious and avoidant) styles; (1) secure attachment is characterized by positive relationship development, effective regulation of positive and negative emotions, and capability in seeking social support when needed, (2) anxious attachment is expressed by excessive need for closeness, worry about relationships, and fear of rejection, and (3) avoidant attachment is defined by emotional distance from others and compulsive self-reliance (Brennan & Shaver, 2005). Securely attached individuals typically have higher self-esteem and self-efficacy compared to insecure individuals (Ainsworth & Bowlby, 1991), and are also better at coping with challenging situations by seeking support from others (Caspers, Cadoret, Langbehn, Yucuis, & Troutman, 2005). Insecurely attached adults, in contrast, report having lower self-esteem, higher levels of emotional dysregulation, are less likely to seek support, and are more likely to engage in non-affiliative maladaptive coping strategies (Pietromonaco & Barrett, 2000). The quality of one's attachments has been shown to predict a myriad of outcomes during the lifespan. Insecure attachment style is, for example, associated with poor interpersonal and intellectual functioning (Marks, Horrocks, & Schutte, 2016; Thorberg & Lyvers, 2010), dysfunctional personality traits (Reiner & Spangler, 2013), and increased risk for mental disorders (Thorberg & Lyvers, 2010). Individuals with insecure attachment styles are particularly susceptible to developing mood disorders, and this association has been observed consistently in children (Abela et al., 2005, 2009), adolescents (Lee & Hankin, 2009) and adults (Hankin, Kassel, & Abela, 2005; Shaver, Schachner, & Mikulincer, 2005; Wei, Mallinckrodt, Larson, & Zakalik, 2005). Significant mood problems among insecurely attached people are thought to arise from poor working models that incorporate dysfunctional attitudes about oneself and others (Hankin et al., 2005; Shaver et al., 2005; Wei, Mallinckrodt, Larson & Zakalik, 2005).
Like any other addiction, PG is a learned, goal-directed behavior. Through repeated experience, individuals form certain reasons to gamble. In turn, these motives drive future gambling behavior and increase risk for associated harms. Similar to research examining substance use motivations (Cooper, 1994; Cooper, Frone, Russel, & Mudar, 1995), three central motivations have been identified for gambling: coping motives (e.g., gambling to reduce negative emotion), enhancement motives (i.e., gambling to increase positive affect), and social motives (i.e., gambling to increase social affiliation) (Stewart & Zack, 2008). Mirroring work on substance use motives, research demonstrates that each motive type relates to different aspects of PG behavior (Francis, Dowling, Jackson, Christensen, & Wardle, 2015; Wardell, Quilty, Hendershot, & Bagby, 2015). Notably, coping motives have been shown to predict gambling problems or harms, whereas enhancement motives predict gambling involvement and frequency, as well as related harms (Stewart & Zack, 2008; Wardell et al., 2015). Social motives are not typically associated with gambling harms or problems and only predict low gambling frequency (Canale, Santinello, & Griffiths, 2015; Stewart & Zack, 2008). We speculate that coping motives may be imperative to increasing risk for gambling harms among insecurely attached individuals.
factors for PG (Gupta et al., 2013), although these various risk factors have yet to be fully explored and it has been argued that more research is needed to investigate these complex pathways (see e.g., Milosevic & Ledgerwood, 2010). In this study, our specific goal was to explore insecure attachment as one of the possible risk factors in the emotionally vulnerable pathway to PG. 1.1. The emotionally vulnerable gambler
1.4. The current study Our main goal was to provide novel evidence for the role of insecure attachment in PG among individuals with mood disorders (Quilty, Mackew, & Bagby, 2014). Informed by attachment theory (Ainsworth & Bowlby, 1991), we expected that individuals with insecure attachment styles (anxious and avoidant) would have increased PG severity because of their elevated depressive symptoms and subsequently elevated coping motives for gambling. 2. Materials and methods 2.1. Participants and procedure Participants were 275 adults (63.6% female; Mage = 43.02, SDage = 11.58) with a lifetime diagnosis of a depressive disorder (n = 138, including major depressive disorder [n = 119], dysthymic disorder [n = 18], and depressive disorder not otherwise specified [n = 1]) or a bipolar disorder (n = 137, including bipolar I disorder [n = 110], bipolar II disorder [n = 21], bipolar disorder not otherwise 167
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
2.2. Measures
specified [n = 6]). The data in this study came from a large-scale study on PG within mood disorders (Quilty et al., 2014). Participants were recruited from the community using online and print advertisements. To be eligible for participation, individuals must have met criteria for a mood disorder diagnosis as indicated above within the past ten years (confirmed by the Structured Clinical Interview for DSM-IV, Axis I Disorders, Patient version; First, Spitzer, Gibbon, & Williams, 2002), as well as had the capacity to provide informed consent and to complete study measures in English. Participants were excluded if they were experiencing severe mania or rapid cycling mood episodes at the time of testing, due to safety and compliance concerns. Other psychiatric disorders were permitted, and exhibited by approximately 50% (n = 137) of participants, including substance use disorders (n = 39), anxiety disorders (n = 108), somatoform disorders (n = 10), eating disorders (n = 18), adjustment disorder (n = 1) and impulse control disorder (n = 1; n = 56 exhibited more than one comorbid diagnosis). Participants completed a clinical interview and self-report measures across two days, in a confidential space at the Centre for Addiction and Mental Health (CAMH). Clinical interviews were completed or supervised by registered Clinical Psychologists in Ontario. Trained research assistants oversaw the completion of all study assessment measures, also under the supervision of registered Psychologists. All procedures were approved by the REB at the CAMH before data collection. Participants received an honorarium for their participation. See Table 1 for complete participant sociodemographic information.
2.2.1. Experiences in close relationships scale-revised (Brennan, Clark, & Shaver, 1998; Fraley, Waller, & Brennan, 2000). This is a 36-item self-report capturing the two dimensions of insecure adult attachment styles: anxious attachment (18-items) and avoidant attachment (18-items). Insecure attachment styles reflect high scores on either or both of these subscales and secure attachment reflects low scores on both dimensions. Responses range on a 7-point scale (1 [strongly disagree] to 7 [strongly disagree]) and mean scores are used. Previous work supports very good to excellent reliability for the subscales (Sibley, Fischer, & Liu, 2005; Sibley & Liu, 2004). The Cronbach alphas in our sample were also excellent (anxious attachment = 0.92, and avoidant attachment = 0.94). 2.2.2. Hamilton Rating Scale for Depression (Hamilton, 1960). The Hamilton Rating Scale for Depression (HAMD) is a 17-item clinician-rated measure of major depression. The HAMD has been used extensively in clinical outcome trials and basic research studies. The severity of each symptom is assessed on a response scale of either 0–2 or 0–4, with higher scores indicating greater depressive symptom severity. A sum score is used to reflect the severity of depressive symptoms. Previous research supports the validity and reliability of the HAMD and associated measures (Bagby, Ryder, Schuller, & Marshall, 2004; Zimmerman, Martinez, Young, Chelminski, & Dalrymple, 2013). The Cronbach alpha of the HAMD in our sample was good (0.82).
Table 1 Summary of participant sociodemographic information.
2.2.3. Reasons for Gambling Questionnaire This self-report measure has 11 items (Quilty, Keough, Toneatto, Watson, & Bagby (under review)) and was first used by Wardell et al. (2015). This measure captures three motives for gambling: social (3items), enhancement (4-items), and coping (4-items). Responses were on a 7-point scale (−3 [strongly disagree] to 0 [neither agree nor disagree] to + 3 [strongly agree]) and mean scores were calculated. In their initial validation study, Quilty and colleagues found support for the three factor structure of this scale and results also indicated good concurrent validity. The Cronbach alphas in our sample were in the acceptable to good range (coping motives = 0.88; enhancement motives = 0.82, and social motives = 0.69).
Sociodemographic variable
Gender Male Female Ethnicity White Black South Asian Chinese Japanese Filipino Arab/West Indian Latin American Bi-racial Education < High school High school Part post-secondary Post-secondary Post-graduate Marital status Married/common law Widowed Divorced/separated Single Employment status Employed Unemployed Student Retired On Physical/mental disability Caregiver/homemaker Did not respond Primary gambling activity Lottery Bet on horse racing Bingo Casino Bet on recreation Internet gambling Invest in stock
n
%
102 173
37.1 62.9
233 9 4 2 2 4 5 4 12
84.7 3.3 1.5 0.7 0.7 1.5 1.8 1.5 4.4
12 22 78 127 36
4.4 8.0 28.4 46.2 13.0
73 3 66 133
26.5 1.1 24.0 48.4
104 29 16 5 98 5 18
37.8 10.5 5.8 1.8 35.6 1.8 6.5
149 2 11 18 13 6 75
54.3 0.7 4.0 6.6 4.7 2.3 27.4
2.2.4. South Oaks Gambling Screen We used the South Oaks Gambling Screen (SOGS; Lesieur & Blume, 1987) to reflect gambling severity. The SOGS has 20-items and is a selfreport measure that is widely used in gambling studies (e.g., Abbott & Volberg, 2006; Goodie et al., 2013). Responses are summed and scores of 3–4 (problem gambling) and 5 + (probable Gambling Disorder) are thought to reflect elevated gambling problems (Abbott & Volberg, 2006). Previous research has shown that the SOGS sum score aligns with DSM-IV criteria for pathological gambling (Goodie et al., 2013). Based on existing SOGS cutoff scores, 16% of our participants had significant problem gambling symptoms or probable Gambling Disorder. The SOGS scores in our sample ranged from 0 to 16, suggesting that our sample displayed good variability in gambling problem severity. Accordingly, consistent with previous work (e.g., Keough et al., 2017), we treated the SOGS sum score as a continuous outcome of gambling problem severity. The Cronbach alpha of the SOGS in our sample was good (0.83). 2.3. Overview of data analysis The analyses for the current study proceeded in two steps. First, we screened the data and inspected descriptive statistics and correlations (Kline, 2009; Wilkinson, 1999). There were no outliers on model variables. Attachment styles, depressive symptoms, and gambling motive variables were normally distributed (Zskew and Zkurtosis < ± 3.29; Kim, 168
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
significant covariances among insecure attachment styles and among motives for gambling. Given that our sample was comprised of individuals with a life history of depressive versus bipolar disorders, we examined potential group differences in the hypothesized model. Specifically, we compared fit across constrained and unconstrained models (i.e., only paths were restrained across groups, but not error terms or covariances). Results showed that both constrained (χ2(26) = 23.39, p = 0.61, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.06]) and unconstrained (χ2(18) = 16.05, p = 0.59, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.07]) models provided excellent fit to the data. According to Cheung and Rensvold (2002), the lack of difference in CFI fit values across models suggests no appreciable group differences in fit. Based on literature showing gender differences in mood, gambling motivations, and gambling harms, it is possible that components of the model differed between men and women in our study (e.g., Tavares, Zilberman, Beites, & Gentil, 2001; Walker, Hinch, & Weighill, 2005). For example, women tend to report gambling more frequently to cope, whereas, men tend to endorse enhancement motives for gambling (Walker et al., 2005). Therefore, we tested whether the model was invariant across gender. Results showed that unconstrained (χ2(18) = 16.23, p = 0.57, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.069] and constrained (χ2(26) = 21.77, p = 0.70, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.05]) models provided excellent fit to the data. Again, the lack of difference in CFI fit values indicated no differences in model fit between men and women. Accordingly, based on the above, we present the model collapsed across diagnostic and gender groups. Based on the bias-corrected bootstrapped 95% CIs, we found support for two hypothesized indirect effects. Consistent with predictions, the indirect effect from anxious-attachment to PG, via both depressive symptoms and coping motives, was supported (B = 0.079, 95% CI [0.025, 0.162]). This suggests that depressive symptoms and coping motives sequentially mediate the association between anxious-attachment and gambling severity. Moreover, this double mediational effect was also present for the avoidant-attachment pathway to PG severity (B = 0.037, 95% CI [0.003, 0.116]) (Fig. 1).
2013); however, the gambling problem severity outcome was positively skewed (Zskew = 17.71, Zkurtosis = 19.96). Second, the hypothesized model was tested using path analysis in Mplus version 7.3 (Muthén & Muthén, 2012). Given the non-normality of the gambling problem severity outcome, model fit indices and path parameters were estimated using robust maximum likelihood (MLR). In the hypothesized model, avoidant and anxious attachment styles were independent variables; gambling problem severity was the outcome; and depressive symptoms and gambling motives were sequential mediators. Covariances among attachment styles and among gambling motives were also included in the model. We were interested specifically in examining the unique mediational pathways from insecure attachment styles to PG via depressive symptoms and coping motives. As argued elsewhere (e.g., Cooper, 1994; Keough et al., 2017), when examining the role of certain substance use motives, it is important to include the other motives to control for overlapping variance. Model fit was considered excellent if the following criteria were met: Comparative Fit Index (CFI; Bentler, 1990) and the Tucker-Lewis Index (TLI; Tucker & Lewis, 1973) > 0.95; Root Mean Square Error of Approximation (RMSEA) < 0.05 (Hu & Bentler, 1999); and a nonsignificant model χ2 (Kline, 2011). Bias corrected bootstrapping was used to estimate indirect effects based on 95% confidence intervals [CI]). Mediation is said to be present if the 95% CI for an indirect effect does not contain zero (Fritz & MacKinnon, 2007). 3. Results 3.1. Descriptive statistics and correlations Presented in Table 2 are the descriptive statistics and bivariate correlations. At the bivariate level, avoidant attachment (but not anxious attachment) positively correlated with gambling problem severity. Both insecure attachment styles positively correlated with depressive symptom severity. Finally, depressive symptoms and all motives for gambling positively correlated with PG severity. 3.2. Testing hypothesized indirect effects
4. Discussion
The fit of the hypothesized model was excellent (χ2(9) = 9.94, p = 0.36, CFI = 0.998, TLI = 0.996 RMSEA = 0.019 [90%CI = 0.00, 0.07]). After controlling for intercorrelations among variables in this model, we observed that both anxious and avoidant attachment styles were positive predictors of depressive symptoms. Depressive severity was positive predictor of coping (but not enhancement or social) motives for gambling. Coping and enhancement motives were positive predictors of gambling problem severity; whereas, social motives was a negative predictor of gambling behavior. Finally, we observed
The purpose of this study was to provide insight on the emotionally vulnerable pathway to PG among individuals with mood disorders. The use of a clinical sample with mood disorders is a main strength of our study. This allowed us to provide a strong test of how insecure attachment and associated depression impact gambling severity among emotionally vulnerable individuals. We speculated that insecure attachment styles – characterized by poor working models of emotion regulation, interpersonal functioning, and coping – would be positive predictors of PG severity in this population (Ainsworth & Bowlby, 1991; Blaszczynski & Nower, 2002; Hazan & Shaver, 1987). Further, we expected high levels of depression and coping reasons for gambling to explain these associations. Although insecure attachment styles have been shown to relate to increased risk for substance use and mood disorders in adults (Kassel et al., 2007; McNally et al., 2003; Thorberg & Lyvers, 2010), there are comparatively few tests of the association between attachment styles and PG outcomes (Testa et al., 2017a, b). Moreover, to our knowledge, no studies to date have examined mechanisms (or mediators) of these associations. Our results supported hypotheses, showing that insecurely attached adults with mood disorders have more severe gambling pathology due to high levels of depression and coping reasons for gambling. These effects were applicable to both men and women, given the lack of model differences between gender groups. Our study is the first to provide a comprehensive model of the role of attachment in PG among adults with mood disorders and suggests that insecure attachment styles may be important developmental
Table 2 Descriptive statistics and bivariate correlations.
1. Anxiousattachment 2. Avoidantattachment 3. Depressive symptoms 4. Coping motives 5. Enhancement motives 6. Social motives 7. Problem gambling M SD ⁎ ⁎⁎
1
2
3
4
5
6
7
1.00
0.39⁎⁎
0.41⁎⁎
0.12⁎
0.09
0.08
0.11
1.00
0.28⁎⁎
0.14⁎
0.00
0.01
0.13⁎
1.00
0.16⁎⁎
0.09
0.04
0.13⁎
1.00
0.69⁎⁎ 1.00
0.62⁎⁎ 0.51⁎⁎
0.54⁎⁎ 0.43⁎⁎
1.00
0.14⁎ 1.00 1.52 3.38
3.95 1.44
3.36 1.30
10.39 6.90
− 5.01 6.73
− 0.99 7.18
− 3.95 4.59
p < 0.05. p < 0.01.
169
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
Coping Motives 2
R = .04 Anxious Attachment
0.15 (0.05); 0.15, p=.01
33.14 (4.15); 0.70, p<.001
0.31 (0.05); 0.63, p<.001
1.69 (0.31); 0.35, p<.001
Depressive Symptoms
0.74 (0.13); 0.39, p<.001
0.08 (0.06); 0.08, p=.17
Enhancement Motives
18.94 (2.33); 0.62, p<.001
0.02 (0.04); 0.04, p=.55
Problem Gambling 2
R = .01
2
R = .18 0.78 (0.38); 0.15, p=.04
0.07 (0.03); 0.16, p=.04
2
16.54 (4.53); 0.51, p<.001
R = .33
-0.24 (0.06); -0.33, p<.001
Avoidant Attachment Social Motives 2
R = .01 Fig. 1. This figure is the final path model for attachment styles predicting problem gambling as mediated by both depressive symptoms and reasons for gambling. The path estimates are presented in the following order: unstandardized coefficient (standard error); standardized coefficient. The bolded lines are specified paths that were statistically significant (p < 0.05) and the grey lines are specified paths that were non-statistically significant.
disorders and evaluate an analogous model incorporating anxiety in addition to depression. Although a substantial proportion of participants in the current study exhibited an anxiety disorder diagnosis, unfortunately an analogous clinician-rated measure of anxiety was not available to control for these effects or evaluate their specificity to depression. Phase of illness and treatment status will be important to incorporate in this research, such as including participants taking medications that have been associated gambling disorder symptoms, such as those used to treatment Parkinson's. Finally, though not central to our hypotheses, we observed a negative association between social motives and gambling severity. While this is at odds with correlational work (e.g., Stewart & Zack, 2008), results of larger path modelling studies show that social motives can be unrelated to gambling problems (Wardell et al., 2015) and even alcohol use (Merrill & Read, 2010). It may be that social motives for gambling – when pitted against the other “riskier” motives – may be adaptive and protective. Future research should replicate this negative association in other samples using path modelling. There are several notable clinical implications of the current study. First, consistent with psychotherapy moderator research (McBride, Atkinson, Quilty, & Bagby, 2006), our findings suggest that attachment style is an important individual difference to consider when treating PG. In addition to notable problems with mood, insecure attachment styles may exacerbate interpersonal difficulties among emotionally vulnerable gamblers. These interpersonal problems may further reduce coping abilities, increase feelings of social isolation, and may ultimately worsen gambling behavior. Accordingly, attachment-focused treatments (i.e., Interpersonal Psychotherapy) may have particular clinical for enhancing mood (Cuijpers, van Straten, Warmerdam, & Andersson, 2008), for improving social functioning, and ultimately for reducing PG. However, interpersonal approaches to psychotherapy have yet to be fully examined for PG. Second, given that mood symptoms and coping motives for gambling are both malleable factors, clinicians should work with emotionally vulnerable problem gamblers to foster more adaptive cognitions and coping strategies. Overall, our study is among the first to test a comprehensive model of insecure attachment among individuals at risk for emotionally vulnerable gambling. Results provide the foundation for future empirical work in this area.
precursors to gambling. The literature demonstrates that early dysfunctional attachment styles (i.e., in childhood) set the stage for significant problems with mood and coping that carry forward into adolescence and adulthood. For example, these studies illustrate that anxious and avoidant attachment styles in childhood prospectively predict depression and anxiety symptoms in adolescence (Allen, Hauser, & Borman-Spurrell, 1996; Lee & Hankin, 2009; Sund & Wichstrom, 2002). Accordingly, the adverse influence of insecure attachment on mood early in life may enhance the negatively reinforcing effects of addictive behaviors. In turn, young people with insecure attachment styles may show early initiation of addictive behavior, in an attempt to manage unwanted emotions. This is especially problematic because early initiation predicts more severe progression and presentation of PG in adulthood (Burge, Pietrzak, Molina, & Petry, 2004; Burge, Pietrzak, & Petry, 2006). The continuity of attachment styles throughout the lifespan suggests that insecure attachment may possibly be an early precursor to PG in adulthood. However, we cannot fully be certain of this in our study, given the cross-sectional nature of our methods. Ultimately, longitudinal work needs to be done in this area to fully understand how the associations between insecure attachment, mood symptoms, and PG unfold over time from early life to adulthood. The main strength of our study is that we tested a strong theoretically informed model in a sample with mood disorders (i.e., individuals who would be at risk for being emotionally vulnerable gamblers). However, some limitations should be noted. Our data is cross-sectional and therefore, we cannot make causal or definite temporal inferences about the direction of effects in our model. The specified direction was informed by theory, however, it will be important for future work to replicate (and extend) on our findings using well-designed prospective studies. Additionally, while use of a sample with a history of mood disorders provides a strong test for the role of insecure attachment in the emotionally vulnerable pathway, our results may not generalize to other populations. For example, individuals with anxiety disorders are at risk for harms associated with coping-motivated PG (el-Guebaly et al., 2006; Petry et al., 2005). Particularly relevant to the current study, people with anxiety disorders also have marked problems with attachment and associated emotion dysregulation (e.g., Eng, Heimberg, Hart, Schneier, & Liebowitz, 2001; Lee & Hankin, 2009; Warren, Huston, Egeland, & Sroufe, 1997). The pathways model would suggest that these individuals would also be at risk for becoming emotionally vulnerable gamblers. Accordingly, future work should test the present model in a sample of adults with a more fulsome range of anxiety
Role of funding sources This work was funded by Grant #2662 from the Ontario Problem 170
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
Gambling Research Centre, awarded to Dr. R. Michael Bagby. The financial support had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the manuscript for publication.
initiation and severity of gambling and health problems among older adult problem gamblers. Psychiatric Services (Washington, D.C.), 55(12), 1437–1439. Burge, A. N., Pietrzak, R. H., & Petry, N. M. (2006). Pre/early adolescent onset of gambling and psychosocial problems in treatment-seeking pathological gamblers. Journal of Gambling Studies (Online), 22(3), 263–274. Canale, N., Santinello, M., & Griffiths, M. D. (2015). Validation of the reasons for gambling questionnaire (RGQ) in a British population survey. Addictive Behaviors, 45, 276–280. Caspers, K. M., Cadoret, R. J., Langbehn, D., Yucuis, R., & Troutman, B. (2005). Contributions of attachment style and perceived social support to lifetime use of illicit substances. Addictive Behaviors, 30, 1007–1011. Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9, 233–255. Cooper, M. L. (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6(2), 117–128. Cooper, M. L., Frone, M. R., Russel, M., & Mudar, P. (1995). Drinking to regulate positive and negative emotions: A motivational model of alcohol use. Journal of Personality and Social Psychology, 69(5), 990–1005. Cuijpers, P., van Straten, A., Warmerdam, L., & Andersson, G. (2008). Psychological treatment of depression: A meta-analytic database of randomized studies. BMC Psychiatry, 8(1), 36. Demidenko, N., Manion, I., & Lee, C. (2015). Father–daughter attachment and communication in depressed and nondepressed adolescent girls. Journal of Child and Family Studies, 24(6), 1727–1734. Di Trani, M., Renzi, A., Vari, C., Zavattini, G. C., & Solano, L. (2017). Gambling disorder and affect regulation: The role of alexithymia and attachment style. Journal of Gambling Studies, 33(2), 649–659. Dowling, N. A., Cowlishaw, S., Jackson, A. C., Merkouris, S. S., Francis, K. L., & Christensen, D. R. (2014). The prevalence of comorbid personality disorders in treatment-seeking problem gamblers: A systematic review and meta-analysis. Journal of Personality Disorders, 29(6), 735–754. el-Guebaly, N., Patten, S. B., Currie, S. R., Williams, R., Beck, C., & Maxwell, C. (2006). Epidemiological associations between gambling behavior, substance use and mood and anxiety disorders. Journal of Gambling Studies, 22, 275–287. Eng, W., Heimberg, R. G., Hart, T. A., Schneier, F. R., & Liebowitz, M. R. (2001). Attachment in individuals with social anxiety disorder: The relationship among adult attachment styles, social anxiety, and depression. Emotion, 1(4), 365–380. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002). Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P) New York: Biometrics Research. New York State Psychiatric Institute. Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. Francis, K. L., Dowling, N. A., Jackson, A. C., Christensen, D. R., & Wardle, H. (2015). Gambling motives: Application of the reasons for gambling questionnaire in an Australian population survey. Journal of Gambling Studies, 31(3), 807–823. Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect. Psychological Science, 18(3), 233–239. Getty, H. A., Watson, J., & Frisch, G. R. (2000). A comparison of depression and styles of coping in male and female GA members and controls. Journal of Gambling Studies, 16(4), 377–391. Goodie, A. S., MacKillop, J., Miller, J. D., Fortune, E. E., Maples, J., Lance, C., & Campbell, W. K. (2013). Evaluating the south oaks gambling screen with DSM-IV and DSM-5 criteria. Assessment, 20(5), 523–531. Gupta, S., Mana-Capelli, S., Mclean, J. R., Chen, C. T., Ray, S., Gould, K. L., & Mccollum, D. (2013). Identification of SIN pathway targets reveals mechanisms of crosstalk between NDR kinase pathways. Current Biology, 23(4), 333–338. Hamilton, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery & Psychiatry, 23(1), 56–62. Hankin, B. L., Kassel, J. D., & Abela, J. R. Z. (2005). Adult attachment dimensions and specificity of emotional distress symptoms: Prospective investigations of cognitive risk and interpersonal stress generation as mediating mechanisms. Personality and Social Psychology Bulletin, 31(1), 136–151. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. Kassel, J., Wardle, M., & Roberts, J. (2007). Adult attachment security and college student substance use. Addictive Behaviors, 32(6), 1164–1176. Keough, M. T., Wardell, J. D., Hendershot, C. S., Bagby, R. M., & Quilty, L. C. (2017). Fun seeking and reward responsiveness moderate the effect of the behavioral inhibition system on coping-motivated problem gambling. Journal of Gambling Studies. Kessler, R. C., Hwang, I., Labrie, R., Petukhova, M., Sampson, N. A., Winters, K. C., & Shaffer, H. J. (2008). DSM-IV pathological gambling in the National Comorbidity Survey Replication. Psychological Medicine, 38, 1351–1360. Kim, H.-K. (2013). Statistical notes for clinical researchers: Assessing normal distribution using skewness and kurtosis. Restorative Dentistry and Endodontics, 38(1), 52–54. Kline, R. B. (2009). Becoming a behavioral science researcher: A guide to producing research that matters. New York: Guilford Press. Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford. Lee, A., & Hankin, B. L. (2009). Insecure attachment, dysfunctional attitudes, and low self-esteem predicting prospective symptoms of depression and anxiety during adolescence. Journal of Clinical Child and Adolescent Psychology, 38(2), 219–231. Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. American Journal of
Contributors All authors contributed meaningfully to the conceptual model presented in the manuscript. Dr. Keough conceptualized and analyzed the research model; conducted literature searches and wrote the first draft of the manuscript. Dr. Vilhena-Churchill wrote sections of the methods, and provided conceptual feedback on the full manuscript. Dr. Bagby secured funding for the project, oversaw data collection and operating procedures, and provided extensive feedback on the manuscript. Dr. Quilty contributed substantially to data collection and to the conceptual model tested in the manuscript. She also provided ongoing feedback on several manuscript drafts. The final manuscript reflects the combined substantial effort of all co-authors and together we declare that we approve of this submission. Conflict of interest All authors declare that they have no conflicts of interest. Acknowledgements N/A. Disclosures The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection and preparation of this manuscript was supported by Grant #2662 from the Ontario Problem Gambling Research Centre. Dr. Bagby was the PI on this grant. References Abbott, M. W., & Volberg, R. A. (2006). The measurement of adult problem and pathological gambling. International Gambling Studies, 6(2), 175–200. Abela, J. R. Z., Hankin, B. L., Haigh, E. A. P., Adams, P., Vinokuroff, T., & Trayhern, L. (2005). Interpersonal vulnerability to depression in high-risk children: The role of insecure attachment and reassurance seeking. Journal of Clinical Child and Adolescent Psychology, 34(1), 182–192. Abela, J. R. Z., Zinck, S., Kryger, S., Zilber, I., & Hankin, B. L. (2009). Contagious depression: Negative attachment cognitions as a moderator of the temporal association between parental depression and child depression. Journal of Clinical Child and Adolescent Psychology, 38(1), 16–26. Afifi, T. O., Cox, B. J., Martens, P. J., Sareen, J., & Enns, M. W. (2010). The relationship between problem gambling and mental and physical health correlates among a nationally representative sample of Canadian women. Canadian Journal of Public Health, 101(2), 171–175. Ainsworth, M. S., & Bowlby, J. (1991). An ethological approach to personality development. American Psychologist, 46(4), 333–341. Allen, J. P., Hauser, S. T., & Borman-Spurrell, E. (1996). Attachment theory as a framework for understanding sequelae of severe adolescent psychopathology: An 11-year follow-up study. Journal of Consulting and Clinical Psychology, 64(2), 254–263. Bagby, R. M., Ryder, A. G., Schuller, D. R., & Marshall, M. B. (2004). The Hamilton Depression Rating Scale: Has the gold standard become a lead weight? The American Journal of Psychiatry, 161(12), 2163–2177. Bentler, P. M. (1990). Comparatives fit indexes in structural models. Psychological Bulletin, 107, 238–246. Blaszczynski, A., & Nower, L. (2002). A pathways model of problem and pathological gambling. Addiction, 97, 487–499. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. Attachment theory and close relationships. New York: Guilford Press. Brennan, K. A., & Shaver, P. R. (2005). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21(3), 267–283. Burge, A. N., Pietrzak, R. H., Molina, C. A., & Petry, N. M. (2004). Age of gambling
171
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
experiences in close relationships (ECR-R) self-report measure of adult romantic attachment. Personality and Social Psychology Bulletin, 31(11), 1524–1536. Sibley, C. G., & Liu, J. H. (2004). Short-term temporal stability and factor structure of the revised experiences in close relationships (ECR-R) measure of adult attachment. Personality and Individual Differences, 36(4), 969–975. Stewart, S. H., & Zack, M. (2008). Development and psychometric evaluation of a threedimensional gambling motives questionnaire. Addiction, 103(7), 1110–1117. Suissa, A. J. (2011). Vulnerability and gambling addiction: Psychological benchmarks and avenues for intervention. International Journal of Mental Health and Addiction, 9(1), 12–23. Sund, A. M., & Wichstrom, L. (2002). Insecure attachment as a risk factor for future depressive symptoms in early adolescence. Journal of the American Academy of Child and Adolescent Psychiatry, 41(12), 1478–1485. Tavares, H., Zilberman, M., Beites, F., & Gentil, V. (2001). Gender differences in gambling progression. Journal of Gambling Studies, 17, 151–159. Testa, R., Grandinetti, P., Pascucci, M., Bruschi, A., Parente, P., Pozzi, G., & Janiri, L. (2017a). Attachment styles and severity of pathological gambling: Preliminary evaluations. European Psychiatry, 41, S396. Testa, R., Grandinetti, P., Pascucci, M., Bruschi, A., Parente, P., Pozzi, G., & Janiri, L. (2017b). Attachment disorders in alcohol and gambling addicted patients: Preliminary evaluations. European Psychiatry, 41, S396. Thorberg, F. A., & Lyvers, M. (2010). Attachment in relation to affect regulation and interpersonal functioning among substance use disorder in patients. Addiction Research and Theory, 18(4), 464–478. Tucker, L. R., & Lewis, C. (1973). The reliability coefficient for maximum likelihood factor analysis. Psychometrika, 38, 1–10. Walker, G. J., Hinch, T. D., & Weighill, A. (2005). Inter-and intra-gender similarities and differences in motivations for casino gambling. Leisure Sciences, 27(2), 111–130. Wardell, J. D., Quilty, L. C., Hendershot, C. S., & Bagby, R. M. (2015). Motivational pathways from reward sensitivity and punishment sensitivity to gambling frequency and gambling-related problems. Psychology of Addictive Behaviors, 29(4), 1022–1030. Warren, S. L., Huston, L., Egeland, B., & Sroufe, L. A. (1997). Child and adolescent anxiety disorders and early attachment. Journal of the American Academy of Child & Adolescent Psychiatry, 36(5), 637–644. Wei, M., Mallinckrodt, B., Larson, L. M., & Zakalik, R. A. (2005). Adult attachment, depressive symptoms, and validation from self versus others. Journal of Counselling Psychology, 52(3), 368–677. Wilkinson, L., & Task Force on Statistical Inference (1999). Statistical methods in psychology journals: Guidelines and explanations. American Psychologist, 54, 594–604. Williams, R. J., Volberg, R. A., & Stevens, R. M. G. (2012). The population prevalence of problem gambling: Methodological influences, standardized rates, jurisdictional differences, and worldwide trends. Report prepared for the Ontario Problem Gambling Research Centre and the Ontario Ministry of Health and Long Term Care. (May 8, 2012) http://hdl.handle.net/10133/3068. Zimmerman, M., Martinez, J. H., Young, D., Chelminski, I., & Dalrymple, K. (2013). Severity classification on the Hamilton depression rating scale. Journal of Affective Disorders, 150(2), 384–388.
Psychiatry, 144, 1184–1188. Magoon, M., & Ingersoll, E. (2006). Parental modeling, attachment, and supervision as moderators of adolescent gambling. Journal of Gambling Studies, 22(1), 1–22. Maniaci, G., Picone, F., Dimarco, T., Lipari, A., Brancato, A., & Cannizzaro, C. (2015). Psychodiagnostic assessment of pathological gamblers: A focus on personality disorders, clinical syndromes and alexithymia. International Journal of Mental Health and Addiction, 13(6), 728–739. Marks, A. D. G., Horrocks, K. A., & Schutte, N. S. (2016). Emotional intelligence mediates the relationship between insecure attachment and subjective health outcomes. Personality and Individual Differences, 98, 188–192. McBride, C., Atkinson, L., Quilty, L. C., & Bagby, R. M. (2006). Attachment as moderator of treatment outcome in major depression: A randomized control trial of interpersonal psychotherapy versus cognitive behavior therapy. Journal of Consulting and Clinical Psychology, 74(6), 1041–1054. McNally, A. M., Palfai, T. P., Levine, R. V., & Moore, B. M. (2003). Attachment dimensions and drinking-related problems among young adults: The mediational role of coping motives. Addictive Behaviors, 28(6), 1115–1127. Merrill, J. E., & Read, J. P. (2010). Motivational pathways to unique types of alcohol consequences. Psychology of Addictive Behaviors, 24(4), 705–711. Milosevic, A., & Ledgerwood, D. M. (2010). The subtyping of pathological gambling: A comprehensive review. Clinical Psychology Review, 30(8), 988–998. Muthén, L. K., & Muthén, B. O. (2012). Mplus user's guide: Version 7. Los Angeles: Muthén and Muthén. Nehlin, C., Grönbladh, L., Fredriksson, A., & Jansson, L. (2013). Alcohol and drug use, smoking, and gambling among psychiatric outpatients: A 1-year prevalence study. Substance Abuse, 34(2), 162–168. Odlaug, B. L., Schreiber, L. R., & Grant, J. E. (2013). Personality dimensions and disorders in pathological gambling. Current Opinion in Psychiatry, 26, 107–112. Otowa, T., York, T. P., Gardner, C. O., Kendler, K. S., & Hettema, J. M. (2014). The impact of childhood parental loss on risk for mood, anxiety and substance use disorders in a population-based sample of male twins. Psychiatry Research, 220(1–2), 404–409. Petry, N. M., Stinson, F. S., & Grant, B. F. (2005). Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: Results from the national epidemiologic survey on alcohol and related conditions. The Journal of Clinical Psychiatry, 66(5), 564–574. Pietromonaco, P. R., & Barrett, L. F. (2000). The internal working models concept: What do we really know about the self in relation to others? Review of General Psychology, 4(2), 155–175. Quilty, L. C., Mackew, L., & Bagby, R. M. (2014). Distinct profiles of behavioural inhibition and activation system sensitivity in unipolar vs. bipolar mood disorders. Psychiatry Research, 219(1), 228–231. Reiner, I., & Spangler, G. (2013). Representations of early attachment experiences and personality in adulthood. Journal of Adult Development, 20(1), 38–45. Shaver, P. R., Schachner, D. A., & Mikulincer, M. (2005). Attachment style, excessive reassurance seeking, relationship processes, and depression. Personality and Social Psychology Bulletin, 31(3), 343–359. Sibley, C. G., Fischer, R., & Liu, J. H. (2005). Reliability and validity of the revised
172
Contents lists available at ScienceDirect
Addictive Behaviors journal homepage: www.elsevier.com/locate/addictbeh
Depression symptoms and reasons for gambling sequentially mediate the associations between insecure attachment styles and problem gambling
T
⁎
Matthew T. Keougha, , Trinda L. Pennistona, Natalie Vilhena-Churchillb, R. Michael Bagbyc, Lena C. Quiltyd,e a
University of Manitoba, Department of Psychology, 190 Dysart Road, P314 Duff Roblin Building, Winnipeg, MB R3T 2N2, Canada Altum Health, University Health Network, Krembil Discovery Tower, 399, Bathurst St., Toronto, Ontario M5T 2S8, Canada c University of Toronto, Departments of Psychology and Psychiatry, 1265 Military Trail, Toronto, Ontario M1C 1A4, Canada d Centre for Addiction and Mental Health, Campbell Family Mental Health Research Institute, 100 Stokes Street, Bell Gateway Building, Toronto, Ontario M6J 1H4, Canada e University of Toronto, Department of Psychiatry, 250 College St., Toronto, Ontario M5T 1R8, Canada b
H I G H L I G H T S test the associations between insecure attachment styles and gambling problems. • We clinical sample with mood disorders was used. • ADepression and coping motives mediated attachment-gambling associations. • Insecure attachment increases problem gambling risk in adults with mood disorders. •
A R T I C L E I N F O
A B S T R A C T
Keywords: Insecure attachment Depressive symptoms Gambling motives Mood disorders
One of the central pathways to problem gambling (PG) is gambling to cope with negative moods, which is a cardinal feature of depression. Insecure attachment styles are also etiologically related to depression; and, therefore, by extension, those who are insecurely attached may engage in excessive gambling behaviors to cope with depression. In this study, we aimed to evaluate this and to this end predicted that depression severity and coping motives for gambling would conjointly mediate the relations between insecure attachment styles and PG. Data came from a larger investigation of PG within mood disorders. Participants exhibited a lifetime depressive or bipolar disorder and endorsed a mood episode within the past ten years. Participants (N = 275) completed self-report measures during a two-day assessment. Path analysis supported two main indirect effects. First, anxious attachment predicted elevated depression, which in turn predicted increased coping motives for gambling, which subsequently predicted greater PG severity. Second, this double mediational pathway was also observed for avoidant attachment. Results suggest that insecure attachment relates to PG via depressive symptoms and coping-related gambling motives. Mood symptoms and associated gambling motives are malleable and are promising targets of gambling interventions for insecurely attached individuals.
1. Introduction
Dowling et al., 2014; Maniaci et al., 2015; Nehlin et al., 2013; Odlaug, Schreiber, & Grant, 2013), and high suicidality (Petry, Stinson, & Grant, 2005; Suissa, 2011). Clinical interventions would benefit from a better understanding of the risk factors that contribute to PG. Blaszczynski and Nower (2002) identified three distinct pathways to PG, including (1) behaviorally conditioned problem gamblers, (2) emotionally vulnerable problem gamblers, and (3) impulsivist/antisocial problem gamblers. Each pathway is believed to unfold over the course of development and each is thought to carry diverse but unique risk
About 2–3% of North Americans gamble at harmful levels (Kessler et al., 2008; Williams, Volberg, & Stevens, 2012). This rate more than triples in psychiatric populations (e.g., people with mood disorders) (Getty, Watson, & Frisch, 2000; Nehlin, Grönbladh, Fredriksson, & Jansson, 2013). Problem gambling (PG) is associated with many negative outcomes, like financial distress, high rates of addiction, poor mental and physical health (Afifi, Cox, Martens, Sareen, & Enns, 2010;
⁎
Corresponding author. E-mail address: [email protected] (M.T. Keough).
https://doi.org/10.1016/j.addbeh.2017.11.018 Received 19 July 2017; Received in revised form 9 November 2017; Accepted 9 November 2017 Available online 14 November 2017 0306-4603/ © 2017 Elsevier Ltd. All rights reserved.
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
Emotionally vulnerable gamblers are people who struggle with elevated levels of negative affect and emotional lability. Consequently, they have ample opportunities to learn that gambling has negatively reinforcing effects on mood (Blaszczynski & Nower, 2002; Keough, Wardell, Hendershot, Bagby, & Quilty, 2017). Through repeated experience, the association between dysregulated mood and gambling (as a coping strategy) becomes strengthened, resulting in the development of PG. In the current study, we used a complementary theoretical framework – attachment theory (Ainsworth & Bowlby, 1991) – to provide new insight on the individual differences and mechanisms of the emotionally vulnerable pathway to PG among people with mood disorders.
The insecurely attached person is thought to engage in maladaptive coping strategies, such as addictive behaviors, to reduce high levels of distress. Insecure attachment styles have been positively linked with alcohol (McNally, Palfai, Levine, & Moore, 2003) and substance use (Kassel, Wardle, & Roberts, 2007; Thorberg & Lyvers, 2010); surprisingly few studies have examined the risk impact of insecure attachment styles for dysfunctional gambling behavior. Recent work suggests that insecure attachment positively correlates with problem gambling (Di Trani, Renzi, Vari, Zavattini, & Solano, 2017; Testa et al., 2017a, 2017b). Moreover, some studies have examined the link between poor parental bonding (a correlate or sequelae of attachment) and PG, supporting the notion that poor parental attachment predicts increased adolescent engagement in gambling behaviors (Magoon & Ingersoll, 2006). Poor parental attachment is also related to the development of depressive symptoms (Demidenko, Manion, & Lee, 2015; Otowa, York, Gardner, Kendler, & Hettema, 2014); it is also possible then that depressed mood somehow plays a role in the association between insecure attachment and risk for problem gambling. Despite the potential causal relations between depressed mood, insecure attachment and PG, the interrelations of these factors has yet to be examined explicitly in pathway modeling.
1.2. Adult attachment, emotional dysregulation, and problem gambling
1.3. Role of gambling motives
Attachment is the tendency for us to form emotional bonds with other humans. It is believed to be biologically rooted and begins with infant-caregiver interactions during the initial years of life. Attachment theory (Ainsworth & Bowlby, 1991) states that early bonding with a caregiver is important for the formation of internal working models of communication, emotion regulation and coping, and interpersonal functioning that are carried through to adulthood. These internal models are believed to be stable across development and the same theories explaining parent-child bonds can also be applied to understand adult interpersonal relationships and associated developmental outcomes (Ainsworth & Bowlby, 1991; Hazan & Shaver, 1987). There are three general attachment styles categorized into secure and insecure (anxious and avoidant) styles; (1) secure attachment is characterized by positive relationship development, effective regulation of positive and negative emotions, and capability in seeking social support when needed, (2) anxious attachment is expressed by excessive need for closeness, worry about relationships, and fear of rejection, and (3) avoidant attachment is defined by emotional distance from others and compulsive self-reliance (Brennan & Shaver, 2005). Securely attached individuals typically have higher self-esteem and self-efficacy compared to insecure individuals (Ainsworth & Bowlby, 1991), and are also better at coping with challenging situations by seeking support from others (Caspers, Cadoret, Langbehn, Yucuis, & Troutman, 2005). Insecurely attached adults, in contrast, report having lower self-esteem, higher levels of emotional dysregulation, are less likely to seek support, and are more likely to engage in non-affiliative maladaptive coping strategies (Pietromonaco & Barrett, 2000). The quality of one's attachments has been shown to predict a myriad of outcomes during the lifespan. Insecure attachment style is, for example, associated with poor interpersonal and intellectual functioning (Marks, Horrocks, & Schutte, 2016; Thorberg & Lyvers, 2010), dysfunctional personality traits (Reiner & Spangler, 2013), and increased risk for mental disorders (Thorberg & Lyvers, 2010). Individuals with insecure attachment styles are particularly susceptible to developing mood disorders, and this association has been observed consistently in children (Abela et al., 2005, 2009), adolescents (Lee & Hankin, 2009) and adults (Hankin, Kassel, & Abela, 2005; Shaver, Schachner, & Mikulincer, 2005; Wei, Mallinckrodt, Larson, & Zakalik, 2005). Significant mood problems among insecurely attached people are thought to arise from poor working models that incorporate dysfunctional attitudes about oneself and others (Hankin et al., 2005; Shaver et al., 2005; Wei, Mallinckrodt, Larson & Zakalik, 2005).
Like any other addiction, PG is a learned, goal-directed behavior. Through repeated experience, individuals form certain reasons to gamble. In turn, these motives drive future gambling behavior and increase risk for associated harms. Similar to research examining substance use motivations (Cooper, 1994; Cooper, Frone, Russel, & Mudar, 1995), three central motivations have been identified for gambling: coping motives (e.g., gambling to reduce negative emotion), enhancement motives (i.e., gambling to increase positive affect), and social motives (i.e., gambling to increase social affiliation) (Stewart & Zack, 2008). Mirroring work on substance use motives, research demonstrates that each motive type relates to different aspects of PG behavior (Francis, Dowling, Jackson, Christensen, & Wardle, 2015; Wardell, Quilty, Hendershot, & Bagby, 2015). Notably, coping motives have been shown to predict gambling problems or harms, whereas enhancement motives predict gambling involvement and frequency, as well as related harms (Stewart & Zack, 2008; Wardell et al., 2015). Social motives are not typically associated with gambling harms or problems and only predict low gambling frequency (Canale, Santinello, & Griffiths, 2015; Stewart & Zack, 2008). We speculate that coping motives may be imperative to increasing risk for gambling harms among insecurely attached individuals.
factors for PG (Gupta et al., 2013), although these various risk factors have yet to be fully explored and it has been argued that more research is needed to investigate these complex pathways (see e.g., Milosevic & Ledgerwood, 2010). In this study, our specific goal was to explore insecure attachment as one of the possible risk factors in the emotionally vulnerable pathway to PG. 1.1. The emotionally vulnerable gambler
1.4. The current study Our main goal was to provide novel evidence for the role of insecure attachment in PG among individuals with mood disorders (Quilty, Mackew, & Bagby, 2014). Informed by attachment theory (Ainsworth & Bowlby, 1991), we expected that individuals with insecure attachment styles (anxious and avoidant) would have increased PG severity because of their elevated depressive symptoms and subsequently elevated coping motives for gambling. 2. Materials and methods 2.1. Participants and procedure Participants were 275 adults (63.6% female; Mage = 43.02, SDage = 11.58) with a lifetime diagnosis of a depressive disorder (n = 138, including major depressive disorder [n = 119], dysthymic disorder [n = 18], and depressive disorder not otherwise specified [n = 1]) or a bipolar disorder (n = 137, including bipolar I disorder [n = 110], bipolar II disorder [n = 21], bipolar disorder not otherwise 167
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
2.2. Measures
specified [n = 6]). The data in this study came from a large-scale study on PG within mood disorders (Quilty et al., 2014). Participants were recruited from the community using online and print advertisements. To be eligible for participation, individuals must have met criteria for a mood disorder diagnosis as indicated above within the past ten years (confirmed by the Structured Clinical Interview for DSM-IV, Axis I Disorders, Patient version; First, Spitzer, Gibbon, & Williams, 2002), as well as had the capacity to provide informed consent and to complete study measures in English. Participants were excluded if they were experiencing severe mania or rapid cycling mood episodes at the time of testing, due to safety and compliance concerns. Other psychiatric disorders were permitted, and exhibited by approximately 50% (n = 137) of participants, including substance use disorders (n = 39), anxiety disorders (n = 108), somatoform disorders (n = 10), eating disorders (n = 18), adjustment disorder (n = 1) and impulse control disorder (n = 1; n = 56 exhibited more than one comorbid diagnosis). Participants completed a clinical interview and self-report measures across two days, in a confidential space at the Centre for Addiction and Mental Health (CAMH). Clinical interviews were completed or supervised by registered Clinical Psychologists in Ontario. Trained research assistants oversaw the completion of all study assessment measures, also under the supervision of registered Psychologists. All procedures were approved by the REB at the CAMH before data collection. Participants received an honorarium for their participation. See Table 1 for complete participant sociodemographic information.
2.2.1. Experiences in close relationships scale-revised (Brennan, Clark, & Shaver, 1998; Fraley, Waller, & Brennan, 2000). This is a 36-item self-report capturing the two dimensions of insecure adult attachment styles: anxious attachment (18-items) and avoidant attachment (18-items). Insecure attachment styles reflect high scores on either or both of these subscales and secure attachment reflects low scores on both dimensions. Responses range on a 7-point scale (1 [strongly disagree] to 7 [strongly disagree]) and mean scores are used. Previous work supports very good to excellent reliability for the subscales (Sibley, Fischer, & Liu, 2005; Sibley & Liu, 2004). The Cronbach alphas in our sample were also excellent (anxious attachment = 0.92, and avoidant attachment = 0.94). 2.2.2. Hamilton Rating Scale for Depression (Hamilton, 1960). The Hamilton Rating Scale for Depression (HAMD) is a 17-item clinician-rated measure of major depression. The HAMD has been used extensively in clinical outcome trials and basic research studies. The severity of each symptom is assessed on a response scale of either 0–2 or 0–4, with higher scores indicating greater depressive symptom severity. A sum score is used to reflect the severity of depressive symptoms. Previous research supports the validity and reliability of the HAMD and associated measures (Bagby, Ryder, Schuller, & Marshall, 2004; Zimmerman, Martinez, Young, Chelminski, & Dalrymple, 2013). The Cronbach alpha of the HAMD in our sample was good (0.82).
Table 1 Summary of participant sociodemographic information.
2.2.3. Reasons for Gambling Questionnaire This self-report measure has 11 items (Quilty, Keough, Toneatto, Watson, & Bagby (under review)) and was first used by Wardell et al. (2015). This measure captures three motives for gambling: social (3items), enhancement (4-items), and coping (4-items). Responses were on a 7-point scale (−3 [strongly disagree] to 0 [neither agree nor disagree] to + 3 [strongly agree]) and mean scores were calculated. In their initial validation study, Quilty and colleagues found support for the three factor structure of this scale and results also indicated good concurrent validity. The Cronbach alphas in our sample were in the acceptable to good range (coping motives = 0.88; enhancement motives = 0.82, and social motives = 0.69).
Sociodemographic variable
Gender Male Female Ethnicity White Black South Asian Chinese Japanese Filipino Arab/West Indian Latin American Bi-racial Education < High school High school Part post-secondary Post-secondary Post-graduate Marital status Married/common law Widowed Divorced/separated Single Employment status Employed Unemployed Student Retired On Physical/mental disability Caregiver/homemaker Did not respond Primary gambling activity Lottery Bet on horse racing Bingo Casino Bet on recreation Internet gambling Invest in stock
n
%
102 173
37.1 62.9
233 9 4 2 2 4 5 4 12
84.7 3.3 1.5 0.7 0.7 1.5 1.8 1.5 4.4
12 22 78 127 36
4.4 8.0 28.4 46.2 13.0
73 3 66 133
26.5 1.1 24.0 48.4
104 29 16 5 98 5 18
37.8 10.5 5.8 1.8 35.6 1.8 6.5
149 2 11 18 13 6 75
54.3 0.7 4.0 6.6 4.7 2.3 27.4
2.2.4. South Oaks Gambling Screen We used the South Oaks Gambling Screen (SOGS; Lesieur & Blume, 1987) to reflect gambling severity. The SOGS has 20-items and is a selfreport measure that is widely used in gambling studies (e.g., Abbott & Volberg, 2006; Goodie et al., 2013). Responses are summed and scores of 3–4 (problem gambling) and 5 + (probable Gambling Disorder) are thought to reflect elevated gambling problems (Abbott & Volberg, 2006). Previous research has shown that the SOGS sum score aligns with DSM-IV criteria for pathological gambling (Goodie et al., 2013). Based on existing SOGS cutoff scores, 16% of our participants had significant problem gambling symptoms or probable Gambling Disorder. The SOGS scores in our sample ranged from 0 to 16, suggesting that our sample displayed good variability in gambling problem severity. Accordingly, consistent with previous work (e.g., Keough et al., 2017), we treated the SOGS sum score as a continuous outcome of gambling problem severity. The Cronbach alpha of the SOGS in our sample was good (0.83). 2.3. Overview of data analysis The analyses for the current study proceeded in two steps. First, we screened the data and inspected descriptive statistics and correlations (Kline, 2009; Wilkinson, 1999). There were no outliers on model variables. Attachment styles, depressive symptoms, and gambling motive variables were normally distributed (Zskew and Zkurtosis < ± 3.29; Kim, 168
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
significant covariances among insecure attachment styles and among motives for gambling. Given that our sample was comprised of individuals with a life history of depressive versus bipolar disorders, we examined potential group differences in the hypothesized model. Specifically, we compared fit across constrained and unconstrained models (i.e., only paths were restrained across groups, but not error terms or covariances). Results showed that both constrained (χ2(26) = 23.39, p = 0.61, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.06]) and unconstrained (χ2(18) = 16.05, p = 0.59, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.07]) models provided excellent fit to the data. According to Cheung and Rensvold (2002), the lack of difference in CFI fit values across models suggests no appreciable group differences in fit. Based on literature showing gender differences in mood, gambling motivations, and gambling harms, it is possible that components of the model differed between men and women in our study (e.g., Tavares, Zilberman, Beites, & Gentil, 2001; Walker, Hinch, & Weighill, 2005). For example, women tend to report gambling more frequently to cope, whereas, men tend to endorse enhancement motives for gambling (Walker et al., 2005). Therefore, we tested whether the model was invariant across gender. Results showed that unconstrained (χ2(18) = 16.23, p = 0.57, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.069] and constrained (χ2(26) = 21.77, p = 0.70, CFI = 1.00, TLI = 1.00, RMSEA = 0.00 [90%CI = 0.00, 0.05]) models provided excellent fit to the data. Again, the lack of difference in CFI fit values indicated no differences in model fit between men and women. Accordingly, based on the above, we present the model collapsed across diagnostic and gender groups. Based on the bias-corrected bootstrapped 95% CIs, we found support for two hypothesized indirect effects. Consistent with predictions, the indirect effect from anxious-attachment to PG, via both depressive symptoms and coping motives, was supported (B = 0.079, 95% CI [0.025, 0.162]). This suggests that depressive symptoms and coping motives sequentially mediate the association between anxious-attachment and gambling severity. Moreover, this double mediational effect was also present for the avoidant-attachment pathway to PG severity (B = 0.037, 95% CI [0.003, 0.116]) (Fig. 1).
2013); however, the gambling problem severity outcome was positively skewed (Zskew = 17.71, Zkurtosis = 19.96). Second, the hypothesized model was tested using path analysis in Mplus version 7.3 (Muthén & Muthén, 2012). Given the non-normality of the gambling problem severity outcome, model fit indices and path parameters were estimated using robust maximum likelihood (MLR). In the hypothesized model, avoidant and anxious attachment styles were independent variables; gambling problem severity was the outcome; and depressive symptoms and gambling motives were sequential mediators. Covariances among attachment styles and among gambling motives were also included in the model. We were interested specifically in examining the unique mediational pathways from insecure attachment styles to PG via depressive symptoms and coping motives. As argued elsewhere (e.g., Cooper, 1994; Keough et al., 2017), when examining the role of certain substance use motives, it is important to include the other motives to control for overlapping variance. Model fit was considered excellent if the following criteria were met: Comparative Fit Index (CFI; Bentler, 1990) and the Tucker-Lewis Index (TLI; Tucker & Lewis, 1973) > 0.95; Root Mean Square Error of Approximation (RMSEA) < 0.05 (Hu & Bentler, 1999); and a nonsignificant model χ2 (Kline, 2011). Bias corrected bootstrapping was used to estimate indirect effects based on 95% confidence intervals [CI]). Mediation is said to be present if the 95% CI for an indirect effect does not contain zero (Fritz & MacKinnon, 2007). 3. Results 3.1. Descriptive statistics and correlations Presented in Table 2 are the descriptive statistics and bivariate correlations. At the bivariate level, avoidant attachment (but not anxious attachment) positively correlated with gambling problem severity. Both insecure attachment styles positively correlated with depressive symptom severity. Finally, depressive symptoms and all motives for gambling positively correlated with PG severity. 3.2. Testing hypothesized indirect effects
4. Discussion
The fit of the hypothesized model was excellent (χ2(9) = 9.94, p = 0.36, CFI = 0.998, TLI = 0.996 RMSEA = 0.019 [90%CI = 0.00, 0.07]). After controlling for intercorrelations among variables in this model, we observed that both anxious and avoidant attachment styles were positive predictors of depressive symptoms. Depressive severity was positive predictor of coping (but not enhancement or social) motives for gambling. Coping and enhancement motives were positive predictors of gambling problem severity; whereas, social motives was a negative predictor of gambling behavior. Finally, we observed
The purpose of this study was to provide insight on the emotionally vulnerable pathway to PG among individuals with mood disorders. The use of a clinical sample with mood disorders is a main strength of our study. This allowed us to provide a strong test of how insecure attachment and associated depression impact gambling severity among emotionally vulnerable individuals. We speculated that insecure attachment styles – characterized by poor working models of emotion regulation, interpersonal functioning, and coping – would be positive predictors of PG severity in this population (Ainsworth & Bowlby, 1991; Blaszczynski & Nower, 2002; Hazan & Shaver, 1987). Further, we expected high levels of depression and coping reasons for gambling to explain these associations. Although insecure attachment styles have been shown to relate to increased risk for substance use and mood disorders in adults (Kassel et al., 2007; McNally et al., 2003; Thorberg & Lyvers, 2010), there are comparatively few tests of the association between attachment styles and PG outcomes (Testa et al., 2017a, b). Moreover, to our knowledge, no studies to date have examined mechanisms (or mediators) of these associations. Our results supported hypotheses, showing that insecurely attached adults with mood disorders have more severe gambling pathology due to high levels of depression and coping reasons for gambling. These effects were applicable to both men and women, given the lack of model differences between gender groups. Our study is the first to provide a comprehensive model of the role of attachment in PG among adults with mood disorders and suggests that insecure attachment styles may be important developmental
Table 2 Descriptive statistics and bivariate correlations.
1. Anxiousattachment 2. Avoidantattachment 3. Depressive symptoms 4. Coping motives 5. Enhancement motives 6. Social motives 7. Problem gambling M SD ⁎ ⁎⁎
1
2
3
4
5
6
7
1.00
0.39⁎⁎
0.41⁎⁎
0.12⁎
0.09
0.08
0.11
1.00
0.28⁎⁎
0.14⁎
0.00
0.01
0.13⁎
1.00
0.16⁎⁎
0.09
0.04
0.13⁎
1.00
0.69⁎⁎ 1.00
0.62⁎⁎ 0.51⁎⁎
0.54⁎⁎ 0.43⁎⁎
1.00
0.14⁎ 1.00 1.52 3.38
3.95 1.44
3.36 1.30
10.39 6.90
− 5.01 6.73
− 0.99 7.18
− 3.95 4.59
p < 0.05. p < 0.01.
169
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
Coping Motives 2
R = .04 Anxious Attachment
0.15 (0.05); 0.15, p=.01
33.14 (4.15); 0.70, p<.001
0.31 (0.05); 0.63, p<.001
1.69 (0.31); 0.35, p<.001
Depressive Symptoms
0.74 (0.13); 0.39, p<.001
0.08 (0.06); 0.08, p=.17
Enhancement Motives
18.94 (2.33); 0.62, p<.001
0.02 (0.04); 0.04, p=.55
Problem Gambling 2
R = .01
2
R = .18 0.78 (0.38); 0.15, p=.04
0.07 (0.03); 0.16, p=.04
2
16.54 (4.53); 0.51, p<.001
R = .33
-0.24 (0.06); -0.33, p<.001
Avoidant Attachment Social Motives 2
R = .01 Fig. 1. This figure is the final path model for attachment styles predicting problem gambling as mediated by both depressive symptoms and reasons for gambling. The path estimates are presented in the following order: unstandardized coefficient (standard error); standardized coefficient. The bolded lines are specified paths that were statistically significant (p < 0.05) and the grey lines are specified paths that were non-statistically significant.
disorders and evaluate an analogous model incorporating anxiety in addition to depression. Although a substantial proportion of participants in the current study exhibited an anxiety disorder diagnosis, unfortunately an analogous clinician-rated measure of anxiety was not available to control for these effects or evaluate their specificity to depression. Phase of illness and treatment status will be important to incorporate in this research, such as including participants taking medications that have been associated gambling disorder symptoms, such as those used to treatment Parkinson's. Finally, though not central to our hypotheses, we observed a negative association between social motives and gambling severity. While this is at odds with correlational work (e.g., Stewart & Zack, 2008), results of larger path modelling studies show that social motives can be unrelated to gambling problems (Wardell et al., 2015) and even alcohol use (Merrill & Read, 2010). It may be that social motives for gambling – when pitted against the other “riskier” motives – may be adaptive and protective. Future research should replicate this negative association in other samples using path modelling. There are several notable clinical implications of the current study. First, consistent with psychotherapy moderator research (McBride, Atkinson, Quilty, & Bagby, 2006), our findings suggest that attachment style is an important individual difference to consider when treating PG. In addition to notable problems with mood, insecure attachment styles may exacerbate interpersonal difficulties among emotionally vulnerable gamblers. These interpersonal problems may further reduce coping abilities, increase feelings of social isolation, and may ultimately worsen gambling behavior. Accordingly, attachment-focused treatments (i.e., Interpersonal Psychotherapy) may have particular clinical for enhancing mood (Cuijpers, van Straten, Warmerdam, & Andersson, 2008), for improving social functioning, and ultimately for reducing PG. However, interpersonal approaches to psychotherapy have yet to be fully examined for PG. Second, given that mood symptoms and coping motives for gambling are both malleable factors, clinicians should work with emotionally vulnerable problem gamblers to foster more adaptive cognitions and coping strategies. Overall, our study is among the first to test a comprehensive model of insecure attachment among individuals at risk for emotionally vulnerable gambling. Results provide the foundation for future empirical work in this area.
precursors to gambling. The literature demonstrates that early dysfunctional attachment styles (i.e., in childhood) set the stage for significant problems with mood and coping that carry forward into adolescence and adulthood. For example, these studies illustrate that anxious and avoidant attachment styles in childhood prospectively predict depression and anxiety symptoms in adolescence (Allen, Hauser, & Borman-Spurrell, 1996; Lee & Hankin, 2009; Sund & Wichstrom, 2002). Accordingly, the adverse influence of insecure attachment on mood early in life may enhance the negatively reinforcing effects of addictive behaviors. In turn, young people with insecure attachment styles may show early initiation of addictive behavior, in an attempt to manage unwanted emotions. This is especially problematic because early initiation predicts more severe progression and presentation of PG in adulthood (Burge, Pietrzak, Molina, & Petry, 2004; Burge, Pietrzak, & Petry, 2006). The continuity of attachment styles throughout the lifespan suggests that insecure attachment may possibly be an early precursor to PG in adulthood. However, we cannot fully be certain of this in our study, given the cross-sectional nature of our methods. Ultimately, longitudinal work needs to be done in this area to fully understand how the associations between insecure attachment, mood symptoms, and PG unfold over time from early life to adulthood. The main strength of our study is that we tested a strong theoretically informed model in a sample with mood disorders (i.e., individuals who would be at risk for being emotionally vulnerable gamblers). However, some limitations should be noted. Our data is cross-sectional and therefore, we cannot make causal or definite temporal inferences about the direction of effects in our model. The specified direction was informed by theory, however, it will be important for future work to replicate (and extend) on our findings using well-designed prospective studies. Additionally, while use of a sample with a history of mood disorders provides a strong test for the role of insecure attachment in the emotionally vulnerable pathway, our results may not generalize to other populations. For example, individuals with anxiety disorders are at risk for harms associated with coping-motivated PG (el-Guebaly et al., 2006; Petry et al., 2005). Particularly relevant to the current study, people with anxiety disorders also have marked problems with attachment and associated emotion dysregulation (e.g., Eng, Heimberg, Hart, Schneier, & Liebowitz, 2001; Lee & Hankin, 2009; Warren, Huston, Egeland, & Sroufe, 1997). The pathways model would suggest that these individuals would also be at risk for becoming emotionally vulnerable gamblers. Accordingly, future work should test the present model in a sample of adults with a more fulsome range of anxiety
Role of funding sources This work was funded by Grant #2662 from the Ontario Problem 170
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
Gambling Research Centre, awarded to Dr. R. Michael Bagby. The financial support had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the manuscript for publication.
initiation and severity of gambling and health problems among older adult problem gamblers. Psychiatric Services (Washington, D.C.), 55(12), 1437–1439. Burge, A. N., Pietrzak, R. H., & Petry, N. M. (2006). Pre/early adolescent onset of gambling and psychosocial problems in treatment-seeking pathological gamblers. Journal of Gambling Studies (Online), 22(3), 263–274. Canale, N., Santinello, M., & Griffiths, M. D. (2015). Validation of the reasons for gambling questionnaire (RGQ) in a British population survey. Addictive Behaviors, 45, 276–280. Caspers, K. M., Cadoret, R. J., Langbehn, D., Yucuis, R., & Troutman, B. (2005). Contributions of attachment style and perceived social support to lifetime use of illicit substances. Addictive Behaviors, 30, 1007–1011. Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9, 233–255. Cooper, M. L. (1994). Motivations for alcohol use among adolescents: Development and validation of a four-factor model. Psychological Assessment, 6(2), 117–128. Cooper, M. L., Frone, M. R., Russel, M., & Mudar, P. (1995). Drinking to regulate positive and negative emotions: A motivational model of alcohol use. Journal of Personality and Social Psychology, 69(5), 990–1005. Cuijpers, P., van Straten, A., Warmerdam, L., & Andersson, G. (2008). Psychological treatment of depression: A meta-analytic database of randomized studies. BMC Psychiatry, 8(1), 36. Demidenko, N., Manion, I., & Lee, C. (2015). Father–daughter attachment and communication in depressed and nondepressed adolescent girls. Journal of Child and Family Studies, 24(6), 1727–1734. Di Trani, M., Renzi, A., Vari, C., Zavattini, G. C., & Solano, L. (2017). Gambling disorder and affect regulation: The role of alexithymia and attachment style. Journal of Gambling Studies, 33(2), 649–659. Dowling, N. A., Cowlishaw, S., Jackson, A. C., Merkouris, S. S., Francis, K. L., & Christensen, D. R. (2014). The prevalence of comorbid personality disorders in treatment-seeking problem gamblers: A systematic review and meta-analysis. Journal of Personality Disorders, 29(6), 735–754. el-Guebaly, N., Patten, S. B., Currie, S. R., Williams, R., Beck, C., & Maxwell, C. (2006). Epidemiological associations between gambling behavior, substance use and mood and anxiety disorders. Journal of Gambling Studies, 22, 275–287. Eng, W., Heimberg, R. G., Hart, T. A., Schneier, F. R., & Liebowitz, M. R. (2001). Attachment in individuals with social anxiety disorder: The relationship among adult attachment styles, social anxiety, and depression. Emotion, 1(4), 365–380. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002). Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition. (SCID-I/P) New York: Biometrics Research. New York State Psychiatric Institute. Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. Francis, K. L., Dowling, N. A., Jackson, A. C., Christensen, D. R., & Wardle, H. (2015). Gambling motives: Application of the reasons for gambling questionnaire in an Australian population survey. Journal of Gambling Studies, 31(3), 807–823. Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediated effect. Psychological Science, 18(3), 233–239. Getty, H. A., Watson, J., & Frisch, G. R. (2000). A comparison of depression and styles of coping in male and female GA members and controls. Journal of Gambling Studies, 16(4), 377–391. Goodie, A. S., MacKillop, J., Miller, J. D., Fortune, E. E., Maples, J., Lance, C., & Campbell, W. K. (2013). Evaluating the south oaks gambling screen with DSM-IV and DSM-5 criteria. Assessment, 20(5), 523–531. Gupta, S., Mana-Capelli, S., Mclean, J. R., Chen, C. T., Ray, S., Gould, K. L., & Mccollum, D. (2013). Identification of SIN pathway targets reveals mechanisms of crosstalk between NDR kinase pathways. Current Biology, 23(4), 333–338. Hamilton, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery & Psychiatry, 23(1), 56–62. Hankin, B. L., Kassel, J. D., & Abela, J. R. Z. (2005). Adult attachment dimensions and specificity of emotional distress symptoms: Prospective investigations of cognitive risk and interpersonal stress generation as mediating mechanisms. Personality and Social Psychology Bulletin, 31(1), 136–151. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. Kassel, J., Wardle, M., & Roberts, J. (2007). Adult attachment security and college student substance use. Addictive Behaviors, 32(6), 1164–1176. Keough, M. T., Wardell, J. D., Hendershot, C. S., Bagby, R. M., & Quilty, L. C. (2017). Fun seeking and reward responsiveness moderate the effect of the behavioral inhibition system on coping-motivated problem gambling. Journal of Gambling Studies. Kessler, R. C., Hwang, I., Labrie, R., Petukhova, M., Sampson, N. A., Winters, K. C., & Shaffer, H. J. (2008). DSM-IV pathological gambling in the National Comorbidity Survey Replication. Psychological Medicine, 38, 1351–1360. Kim, H.-K. (2013). Statistical notes for clinical researchers: Assessing normal distribution using skewness and kurtosis. Restorative Dentistry and Endodontics, 38(1), 52–54. Kline, R. B. (2009). Becoming a behavioral science researcher: A guide to producing research that matters. New York: Guilford Press. Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford. Lee, A., & Hankin, B. L. (2009). Insecure attachment, dysfunctional attitudes, and low self-esteem predicting prospective symptoms of depression and anxiety during adolescence. Journal of Clinical Child and Adolescent Psychology, 38(2), 219–231. Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. American Journal of
Contributors All authors contributed meaningfully to the conceptual model presented in the manuscript. Dr. Keough conceptualized and analyzed the research model; conducted literature searches and wrote the first draft of the manuscript. Dr. Vilhena-Churchill wrote sections of the methods, and provided conceptual feedback on the full manuscript. Dr. Bagby secured funding for the project, oversaw data collection and operating procedures, and provided extensive feedback on the manuscript. Dr. Quilty contributed substantially to data collection and to the conceptual model tested in the manuscript. She also provided ongoing feedback on several manuscript drafts. The final manuscript reflects the combined substantial effort of all co-authors and together we declare that we approve of this submission. Conflict of interest All authors declare that they have no conflicts of interest. Acknowledgements N/A. Disclosures The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection and preparation of this manuscript was supported by Grant #2662 from the Ontario Problem Gambling Research Centre. Dr. Bagby was the PI on this grant. References Abbott, M. W., & Volberg, R. A. (2006). The measurement of adult problem and pathological gambling. International Gambling Studies, 6(2), 175–200. Abela, J. R. Z., Hankin, B. L., Haigh, E. A. P., Adams, P., Vinokuroff, T., & Trayhern, L. (2005). Interpersonal vulnerability to depression in high-risk children: The role of insecure attachment and reassurance seeking. Journal of Clinical Child and Adolescent Psychology, 34(1), 182–192. Abela, J. R. Z., Zinck, S., Kryger, S., Zilber, I., & Hankin, B. L. (2009). Contagious depression: Negative attachment cognitions as a moderator of the temporal association between parental depression and child depression. Journal of Clinical Child and Adolescent Psychology, 38(1), 16–26. Afifi, T. O., Cox, B. J., Martens, P. J., Sareen, J., & Enns, M. W. (2010). The relationship between problem gambling and mental and physical health correlates among a nationally representative sample of Canadian women. Canadian Journal of Public Health, 101(2), 171–175. Ainsworth, M. S., & Bowlby, J. (1991). An ethological approach to personality development. American Psychologist, 46(4), 333–341. Allen, J. P., Hauser, S. T., & Borman-Spurrell, E. (1996). Attachment theory as a framework for understanding sequelae of severe adolescent psychopathology: An 11-year follow-up study. Journal of Consulting and Clinical Psychology, 64(2), 254–263. Bagby, R. M., Ryder, A. G., Schuller, D. R., & Marshall, M. B. (2004). The Hamilton Depression Rating Scale: Has the gold standard become a lead weight? The American Journal of Psychiatry, 161(12), 2163–2177. Bentler, P. M. (1990). Comparatives fit indexes in structural models. Psychological Bulletin, 107, 238–246. Blaszczynski, A., & Nower, L. (2002). A pathways model of problem and pathological gambling. Addiction, 97, 487–499. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. Attachment theory and close relationships. New York: Guilford Press. Brennan, K. A., & Shaver, P. R. (2005). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21(3), 267–283. Burge, A. N., Pietrzak, R. H., Molina, C. A., & Petry, N. M. (2004). Age of gambling
171
Addictive Behaviors 78 (2018) 166–172
M.T. Keough et al.
experiences in close relationships (ECR-R) self-report measure of adult romantic attachment. Personality and Social Psychology Bulletin, 31(11), 1524–1536. Sibley, C. G., & Liu, J. H. (2004). Short-term temporal stability and factor structure of the revised experiences in close relationships (ECR-R) measure of adult attachment. Personality and Individual Differences, 36(4), 969–975. Stewart, S. H., & Zack, M. (2008). Development and psychometric evaluation of a threedimensional gambling motives questionnaire. Addiction, 103(7), 1110–1117. Suissa, A. J. (2011). Vulnerability and gambling addiction: Psychological benchmarks and avenues for intervention. International Journal of Mental Health and Addiction, 9(1), 12–23. Sund, A. M., & Wichstrom, L. (2002). Insecure attachment as a risk factor for future depressive symptoms in early adolescence. Journal of the American Academy of Child and Adolescent Psychiatry, 41(12), 1478–1485. Tavares, H., Zilberman, M., Beites, F., & Gentil, V. (2001). Gender differences in gambling progression. Journal of Gambling Studies, 17, 151–159. Testa, R., Grandinetti, P., Pascucci, M., Bruschi, A., Parente, P., Pozzi, G., & Janiri, L. (2017a). Attachment styles and severity of pathological gambling: Preliminary evaluations. European Psychiatry, 41, S396. Testa, R., Grandinetti, P., Pascucci, M., Bruschi, A., Parente, P., Pozzi, G., & Janiri, L. (2017b). Attachment disorders in alcohol and gambling addicted patients: Preliminary evaluations. European Psychiatry, 41, S396. Thorberg, F. A., & Lyvers, M. (2010). Attachment in relation to affect regulation and interpersonal functioning among substance use disorder in patients. Addiction Research and Theory, 18(4), 464–478. Tucker, L. R., & Lewis, C. (1973). The reliability coefficient for maximum likelihood factor analysis. Psychometrika, 38, 1–10. Walker, G. J., Hinch, T. D., & Weighill, A. (2005). Inter-and intra-gender similarities and differences in motivations for casino gambling. Leisure Sciences, 27(2), 111–130. Wardell, J. D., Quilty, L. C., Hendershot, C. S., & Bagby, R. M. (2015). Motivational pathways from reward sensitivity and punishment sensitivity to gambling frequency and gambling-related problems. Psychology of Addictive Behaviors, 29(4), 1022–1030. Warren, S. L., Huston, L., Egeland, B., & Sroufe, L. A. (1997). Child and adolescent anxiety disorders and early attachment. Journal of the American Academy of Child & Adolescent Psychiatry, 36(5), 637–644. Wei, M., Mallinckrodt, B., Larson, L. M., & Zakalik, R. A. (2005). Adult attachment, depressive symptoms, and validation from self versus others. Journal of Counselling Psychology, 52(3), 368–677. Wilkinson, L., & Task Force on Statistical Inference (1999). Statistical methods in psychology journals: Guidelines and explanations. American Psychologist, 54, 594–604. Williams, R. J., Volberg, R. A., & Stevens, R. M. G. (2012). The population prevalence of problem gambling: Methodological influences, standardized rates, jurisdictional differences, and worldwide trends. Report prepared for the Ontario Problem Gambling Research Centre and the Ontario Ministry of Health and Long Term Care. (May 8, 2012) http://hdl.handle.net/10133/3068. Zimmerman, M., Martinez, J. H., Young, D., Chelminski, I., & Dalrymple, K. (2013). Severity classification on the Hamilton depression rating scale. Journal of Affective Disorders, 150(2), 384–388.
Psychiatry, 144, 1184–1188. Magoon, M., & Ingersoll, E. (2006). Parental modeling, attachment, and supervision as moderators of adolescent gambling. Journal of Gambling Studies, 22(1), 1–22. Maniaci, G., Picone, F., Dimarco, T., Lipari, A., Brancato, A., & Cannizzaro, C. (2015). Psychodiagnostic assessment of pathological gamblers: A focus on personality disorders, clinical syndromes and alexithymia. International Journal of Mental Health and Addiction, 13(6), 728–739. Marks, A. D. G., Horrocks, K. A., & Schutte, N. S. (2016). Emotional intelligence mediates the relationship between insecure attachment and subjective health outcomes. Personality and Individual Differences, 98, 188–192. McBride, C., Atkinson, L., Quilty, L. C., & Bagby, R. M. (2006). Attachment as moderator of treatment outcome in major depression: A randomized control trial of interpersonal psychotherapy versus cognitive behavior therapy. Journal of Consulting and Clinical Psychology, 74(6), 1041–1054. McNally, A. M., Palfai, T. P., Levine, R. V., & Moore, B. M. (2003). Attachment dimensions and drinking-related problems among young adults: The mediational role of coping motives. Addictive Behaviors, 28(6), 1115–1127. Merrill, J. E., & Read, J. P. (2010). Motivational pathways to unique types of alcohol consequences. Psychology of Addictive Behaviors, 24(4), 705–711. Milosevic, A., & Ledgerwood, D. M. (2010). The subtyping of pathological gambling: A comprehensive review. Clinical Psychology Review, 30(8), 988–998. Muthén, L. K., & Muthén, B. O. (2012). Mplus user's guide: Version 7. Los Angeles: Muthén and Muthén. Nehlin, C., Grönbladh, L., Fredriksson, A., & Jansson, L. (2013). Alcohol and drug use, smoking, and gambling among psychiatric outpatients: A 1-year prevalence study. Substance Abuse, 34(2), 162–168. Odlaug, B. L., Schreiber, L. R., & Grant, J. E. (2013). Personality dimensions and disorders in pathological gambling. Current Opinion in Psychiatry, 26, 107–112. Otowa, T., York, T. P., Gardner, C. O., Kendler, K. S., & Hettema, J. M. (2014). The impact of childhood parental loss on risk for mood, anxiety and substance use disorders in a population-based sample of male twins. Psychiatry Research, 220(1–2), 404–409. Petry, N. M., Stinson, F. S., & Grant, B. F. (2005). Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: Results from the national epidemiologic survey on alcohol and related conditions. The Journal of Clinical Psychiatry, 66(5), 564–574. Pietromonaco, P. R., & Barrett, L. F. (2000). The internal working models concept: What do we really know about the self in relation to others? Review of General Psychology, 4(2), 155–175. Quilty, L. C., Mackew, L., & Bagby, R. M. (2014). Distinct profiles of behavioural inhibition and activation system sensitivity in unipolar vs. bipolar mood disorders. Psychiatry Research, 219(1), 228–231. Reiner, I., & Spangler, G. (2013). Representations of early attachment experiences and personality in adulthood. Journal of Adult Development, 20(1), 38–45. Shaver, P. R., Schachner, D. A., & Mikulincer, M. (2005). Attachment style, excessive reassurance seeking, relationship processes, and depression. Personality and Social Psychology Bulletin, 31(3), 343–359. Sibley, C. G., Fischer, R., & Liu, J. H. (2005). Reliability and validity of the revised
172