Addictive Behaviors 39 (2014) 1404–1407
Contents lists available at ScienceDirect
Addictive Behaviors
Short Communication
Pathways of family influence: Alcohol use and disordered eating in daughters Annette S. Kluck ⁎, Lucille Carriere, Starla Dallesasse 1, Batsirai Bvunzawabaya 2, Erin English, Megan Cobb 3, Therese Borges, Kseniya Zhuzha, Daniel Fry Special Education, Rehabilitation, Counseling (SERC), Auburn University, United States
H I G H L I G H T S • • • • •
Evaluated distal and proximal risk factors for addiction-related coping behaviors Family dynamics directly related to parental addiction-related coping behaviors Family dynamics indirectly related to addiction-related coping behaviors Perceived parental alcohol problems best predicted problematic alcohol use Perceived parental emphasis on appearance best predicted disordered eating
a r t i c l e
i n f o
Available online 28 May 2014 Keywords: Disordered eating Alcohol use Family Behavioral model
a b s t r a c t Models of addiction etiology and treatment emphasize the influence of family-of-origin experiences. Using two addiction-related coping behaviors (ARCBs) common among college women (i.e., problematic alcohol use, disordered eating), we examined whether ARCBs in parents related to matching ARCBs in college women offspring. We expected that matching parental ARCBs would relate more strongly to the ARCBs in offspring than more distal/general family factors. A total of 197 college women completed measures of family dynamics, parental difficulties with alcohol, family focus on appearance and weight, personal difficulties with alcohol use, and disordered eating. A significant indirect effect for family dysfunction on disordered eating and alcohol-related problems was found. That is, family relationship difficulties predicted parents' ARCBs, which predicted matching ARCBs in participants (e.g., parental alcohol problems predicted participant alcohol problems). Matched parental ARCBs were better predictors of participants' ARCBs than more general/distal family factors and non-matched ARCBs. Specifically, path analysis and testing of beta weights supported specificity of parental ARCBs for predicting matching offspring ARCBs. Implications of study findings for tailoring prevention efforts are discussed. © 2014 Published by Elsevier Ltd.
1. Introduction Empirical studies have linked family dysfunction with maladaptive emotion-focused coping behaviors such as those associated with addictive processes (also called addiction-related coping behaviors; ARCBs) related to alcohol use/abuse (Downs & Robertson, 1987; West, Hosie, & Zarski, 1987) and eating disorders (Bruch, 1971; Humphrey, 1988). However, not all individuals originating from families with dysfunctional dynamics develop psychological difficulties. The authors of the present study suggest that parents' problematic coping behavior may
⁎ Corresponding author at: SERC, Auburn, AL 36849, United States. E-mail address: [email protected] (A.S. Kluck). 1 Now at Central Alabama Veterans Health Care System. 2 Now at University of Pennsylvania Counseling and Psychological Services. 3 Now at University of Wisconsin - La Crosse.
http://dx.doi.org/10.1016/j.addbeh.2014.05.015 0306-4603/© 2014 Published by Elsevier Ltd.
provide more specific and proximal sources of risk for a particular ARCB among offspring. Specifically, behavioral theories (e.g., Bandura, Ross, & Ross, 1963) may help explain the tendency for some coping mechanisms to appear to run in families. Addictive behaviors like substance use and disordered eating are often viewed as maladaptive coping mechanisms (e.g., Dube, Anda, Felitti, Edwards, & Croft, 2002; Wagener & Much, 2010). They may reflect somewhat unique psychological difficulties in that they are believed to be a means through which individuals use avoidance to cope with stress, including stress from dysfunctional family environments. In line with behavioral theory (e.g., Bandura et al., 1963), parents may model problematic attitudes and behaviors related to alcohol, body weight, and food as ways of dealing with (through avoidance) difficulties such as family intimacy dysfunction. In addition, addiction research has documented an increased risk in developing an alcohol-related disorder in offspring of alcoholic parents (Sher, Gershuny, Peterson, &
A.S. Kluck et al. / Addictive Behaviors 39 (2014) 1404–1407
Raskin, 1997; Tildesley & Andrews, 2008; Vungkhanching, Sher, Jackson, & Parra, 2004), and parental alcoholism is predictive of the development of problematic alcohol use in offspring after controlling for the potential role of other parental psychological disorders (Chassin, Pitts, DeLucia, & Todd, 1999) and parental relational factors, including parental warmth and hostility (White, Johnson, & Buyske, 2000). Similarly, research indicates that parental modeling of problematic eating may predict problematic eating in their daughters (e.g., Francis & Birch, 2005; Kluck, 2008; Stice, Agras, & Hammer, 1999), and family modeling and use of reinforcement contingencies mediated the relationship between dysfunctional family dynamics and disordered eating (Kluck, 2008). Thus, in families with dysfunctional dynamics, specific ARCBs found among parent(s) may influence the type of ARCBs that offspring develop. If family intimacy dysfunction is a general risk factor and parental modeling behaviors are more specific and proximal risk factors in the development of a particular type of ARCB, modeling of problematic alcohol use in parents should be associated with problematic alcohol use in daughters, and parental modeling of disordered eating attitudes that emphasize thinness and appearance should be associated with disordered eating in daughters. Thus, the purpose of the present study was to test a model where family intimacy dysfunction has an indirect effect on ARCBs in daughters through parents' ARCBs. We hypothesized that (1) family intimacy dysfunction would be associated with ARCBs in parents (i.e., parental alcohol problems, family appearance focus), (2) ARCBs among parents would be associated with increases in similar ARCBs among participants, (3) parents' ARCB (e.g., family appearance focus) that matched the daughters' ARCB (e.g., disordered eating) would be more strongly associated with that specific ARCB among daughters than would general family climate (i.e., intimacy dysfunction) or another type of parental ARCB (e.g., parental problematic alcohol use), and (4) family intimacy dysfunction would have an indirect effect on collegiate women's ARCBs through parents' ARCBs. 2. Method Women (N = 203) recruited from undergraduate psychology courses at a large Southeastern University in the United States served as participants. Only participants under age 25 who were never married were included in study analyses. The resulting sample of 197 participants had a mean age of 20.26 (SD = 1.24). The majority of participants self-identified as Caucasian (81.7%) and heterosexual (99.0%). The balanced cohesion subscale from the Family Adaptability and Cohesion Scales — IV (FACES IV; Olson, 2011) was used to assess participants' perceptions of family intimacy in their family environment with higher scores indicating healthier relationships. ARCBs of parents were assessed using the Children of Alcoholics Screening Test (CAST; Jones, 1983), a measure of perceptions of parental problematic alcohol use during participants' childhoods, and the Family Influence Scale (FIS; Young, Clopton, & Bleckley, 2004), a measure of family emphasis on appearance and thinness. ARCBs of participants were assessed using the Michigan Alcoholism Screening Test (MAST; Selzer, 1971) to measure problematic alcohol use and the Eating Disorder Examination Questionnaire (EDE-Q; Fairburn & Beglin, 1994) to measure disordered eating. Following approval from the university Institutional Review Board, participants were recruited from an undergraduate psychology participant pool. Participants received information about the study purpose, benefits, risks, and their right to discontinue participation without penalty. After providing consent, participants completed a questionnaire packet containing a demographic sheet and the five questionnaires, which were partially counterbalanced to control for order effects. When finished, participants returned the packets to a researcher and received course credit in exchange for their participation.
1405
3. Results All measures were scored such that higher scores were associated with higher levels of the construct. Reliabilities for the study measures were adequate to excellent ranging from .70 (for the MAST) to .97 (for the CAST). Tests of order effects across the four orders were not significant. Consistent with what would be expected if ARCBs in parents were associated with more problematic family functioning, lack of intimacy (as measured by low cohesion, M = 29.74; SD = 5.13) was associated with daughters' perceptions of problematic behaviors in their parents (for the CAST, M = 2.58, SD = 5.59, r = − .22, p b .01; for the FIS, M = 41.23, SD = 10.49, r = − .20, p b .01). Family climate variables that represented the specific perceived parental ARCBs were significantly associated with ARCBs in daughters in the respective domains. Increased perceived parental alcohol use was associated with increased alcohol-related problems among participants (M = 1.52, SD = 1.62), r = .34, p b .01, and increased perceived focus upon appearance and thinness in the family was positively associated with disordered eating (M = 1.88, SD = 1.29), r = .44, p b .01. However, the more diffuse family climate variable (i.e., difficulties with family intimacy) was generally unrelated to problematic behaviors in daughters (for EDE-Q, r = −.04; for MAST, r = −.06). Although participants' scores on the measure of alcohol problems were unrelated to their scores on the measure of disordered eating (r = .07), their perceptions of their parents' alcohol problems were predictive of greater difficulties with disordered eating among participants (r = .17, p b .05). Perceived emphasis on thinness and appearance was unrelated to problematic alcohol use (r = −.05). Next, we examined the predictive power of matched parental– participant behaviors (i.e., alcoholism–alcoholism, appearance focus– disordered eating) compared to family dysfunction and unmatched parental–participant behaviors (e.g., alcoholism–disordered eating) by statistically testing the differences in the regression weights using a modified (with Bonferroni correction) alpha cutoff of .013. Perceived parental alcohol problems were a significantly stronger predictor of participant alcohol problems than were perceived family focus on appearance and thinness (z = 3.76, p b .001) and family cohesion (z = 3.71; p b .001), and there was no significant difference in the predictive power of the latter two variables (z = 0.09, p = .928). Similarly, perceived family focus on appearance and thinness was a significantly stronger predictor of disordered eating than perceived parental alcohol problems (z = 2.74, p = .006) or family cohesion (z = 3.87; p b .001), and there was no significant difference in the predictive power of the latter two variables (z = 1.89; p = .059). As such, only the perceived parental ARCB that matched the ARCB in participants was significantly stronger as a predictor than other family climate variables. Following the recommendations by Preacher and Hayes (2004), we used a bootstrapping technique for the Sobel test to test for the presence of a significant indirect effect. Sobel (z = − 2.65, p = .008; z = − 2.58, p = .010) and bootstrapping (95% CI = − .046 to − .007; 95% CI = −.038 to − .008) approaches revealed a significant indirect effect for perceived parental problems in the relationships between indicators of perceived family functioning (i.e., cohesion) and problematic alcohol use and disordered eating, respectively, in participants. We used path analysis to further test for specificity of the parental ARCB model. Fig. 1 includes three sets of path coefficients for three models (i.e., the full model containing both direct and indirect effects, the model containing only indirect effects, the model with paths between unmatched ARCBs). The direct paths between cohesion to the participants' ARCBs in the full effects models were near zero. The full effects model had adequate fit (Table 1). The indirect effects only model also had adequate fit (Table 1). The χ2 difference between the two models was not significant (p = .748), suggesting that dropping the direct effects paths did not result in over-trimming of the model. We ran a third model, referred to as the crossover model, in which we allowed
1406
A.S. Kluck et al. / Addictive Behaviors 39 (2014) 1404–1407
CASTTOT -0.24 (-0.24) -0.24
0.33 (0.34) 0.34
MASTTOT
(0.02) -0.01
0.02 COHESION (0.05)
0.12
0.07
-0.20 (-0.20) -0.20
0.44 (0.45) 0.44
FISTOT
EDETOT
Fig. 1. Indirect and full effects models of family climate predicting ARCBs in study participants. Perceived family dynamics (measured by FACES IV cohesion balance subscale) are indirectly (and directly in the case of the full model, values for this model appear in parentheses) related to problematic behavior in participants through perceived parental behaviors. CASTTOT represents perceived problems with alcohol in parents, MASTTOT represents self-reported problems with alcohol among participants, FISTOT represents perceived family emphasis on appearance and weight, and EDETOT represents self-reported disordered eating. Plain text path coefficients in parentheses reflect the full effects model (with a large dashed line), those outside of parentheses reflect the indirect effects only model (with a solid line), and those in italics reflect the model with unmatched ARCB paths included (with a small dashed line).
perceived ARCBs to predict unmatched and matched participant ARCBs (see Fig. 1). Consulting the Wald and LaGrange tests revealed that dropping the paths between perceived family focus on appearance and thinness and participant difficulties with alcohol, and between family cohesion and participant difficulties with alcohol, was estimated to improve the fit of the model. Despite having adequate fit (Table 1), the χ2 difference (p = .435) between the indirect effects only model and the crossover model revealed that the crossover model did not fit better than the indirect effects only model. In addition, the Akaike Information Criterion (AIC) revealed slightly less good fit for the third model, supporting the hypothesis that perceived parent ARCBs have specificity for ARCBs in participants.
4. Discussion Children can be influenced by their parents' modeling of certain behaviors and through covert messages that communicate parents' beliefs and expectations (White et al., 2000). The authors expected that perceived parental ARCBs would be more proximal sources of influence in the development of ARCBs among offspring than perceived problems in family dynamics. Our results support the notion that family dynamics might relate to the presence of other problems within the family (like parent ARCBs) and that the type of ARCBs exhibited by daughters may be influenced by the ARCBs modeled by their parents. The relationships between family dynamics and ARCBs among participants were indirect and more distal than were the relationships between perceived parental ARCBs and self-reported ARCBs in participants. The findings are consistent with the idea that perceived family dysfunction may create an environment where problematic emotion-based coping behaviors occur, but it is the specific perceived problematic behaviors in parents that most relate to the specific difficulties in offspring. The present study contains several limitations to consider. These include use of a correlational design that prohibits conclusions regarding
causality, reliance on self-report of a single informant, and use of a homogenous sample of young adults. The findings from the present study have potential implications for mental health professionals. First, knowing about parental ARCBs may be useful in identifying current or potential ARCBs in young women. Second, in the treatment of an adult woman with an addiction, consideration of prevention efforts for offspring may be warranted (e.g., referring other family members for therapy, support or 12-step groups, or psychoeducation). Haller and Chassin (2010) found that offspring of alcoholics who perceived alcohol consumption to be more risky used alcohol less than offspring who did not perceive alcohol consumption as risky, pointing to the possibility of modifying the pattern of intergenerational transmission. Third, it may be helpful to determine the types of ARCBs found in the parents and family members of young adult women seeking treatment because these may relate to current or potential ARCBs in those women. In some cases, therapy, psychoeducational interventions (should parents or family members hold misconceptions about alcohol use and eating/weight or lack knowledge about healthy coping mechanisms), or other supports may be needed for family members of individuals who struggle with problems with alcohol use (such as Al-ANON) and disordered eating. Role of funding sources There were no funding sources for conducting the research or preparing the manuscript. Contributors The authors and their contributions are described below. All authors approve the final manuscript. Annette S. Kluck: designed study, wrote protocol, coordinated data collection, conducted literature searches, conducted statistical analyses, wrote multiple draft of manuscript. Lucille J. Carriere: designed study, wrote protocol, collected data, conducted literature searches, co-wrote first draft of manuscript. Starla Dallesasse: wrote protocol, collected data, entered data, co-wrote multiple drafts of manuscript.
Table 1 Fit indices for path analyses. (Fig. 1). Model
χ2
p
χ2/df
CFI
NNFI
SRMR
RMSEA (90% CI)
AIC
Cohesion indirect & direct effects Cohesion indirect effects only Cohesion crossover model
4.52 5.10 1.31
0.34 0.53 0.52
1.13 0.85 0.66
.99 1.00 1.00
.98 1.02 1.05
.05 .05 .02
.03 (.00–.11) .00 (.00–.09) .00 (.00–.13)
−3.48 −6.90 −2.69
Note, none of the χ2 were significant at p b .05. CFI = Comparative fit index; NNFI = Bentler–Bonnet non-normed fit index; SRMR = standardized root mean-square residual; RMSEA = root mean-square error of approximation; CI = confidence interval; AIC = Akaike Information Criterion.
A.S. Kluck et al. / Addictive Behaviors 39 (2014) 1404–1407 Batsirai Bvunzawabaya: wrote protocol, collected data, entered data, co-wrote multiple drafts of manuscript. Erin English: conducted literature searches, assisted in writing final draft of manuscript. Megan Cobb: collected data. Therese Borges: conducted literature searches, assisted in writing final draft of manuscript. Kseniya Zhuzha: conducted literature searches, assisted in writing final draft of manuscript. Daniel Fry: conducted literature searches, assisted in writing final draft of manuscript. Conflict of interest The authors have no conflicts of interest to disclose. Acknowledgments The authors thank Sophie Ahmad for her assistance with data collection and Natalie Reiner and Sally Kirklewski for their assistance with manuscript preparation.
References Bandura, A., Ross, D., & Ross, S. A. (1963). Vicarious reinforcement and imitative learning. The Journal of Abnormal and Social Psychology, 67, 601–607. http://dx.doi.org/10.1037/ h0045550. Bruch, H. (1971). Family transactions in eating disorders. Comprehensive Psychiatry, 12, 238–248. http://dx.doi.org/10.1016/0010-440X(71)90021-6. Chassin, L., Pitts, S.C., DeLucia, C., & Todd, M. (1999). A longitudinal study of children of alcoholics: Predicting young adult substance use disorders, anxiety, and depression. Journal of Abnormal Psychology, 108, 106–119. http://dx.doi.org/10.1037/0021-843X. 108.1.106. Downs, W. R., & Robertson, J. F. (1987). Family dynamics and adolescent alcohol use: A path analysis with a treatment sample. Journal of Applied Social Sciences, 12, 24–45. Dube, S. R., Anda, R. F., Felitti, V. J., Edwards, V. J., & Croft, J. B. (2002). Adverse childhood experiences and personal alcohol abuse as an adult. Addictive Behaviors, 27, 713–725. http://dx.doi.org/10.1016/S0306-4603(01)00204-0. Fairburn, C. G., & Beglin, S. J. (1994). Assessment of eating disorders: Interview or selfreport questionnaire? International Journal for Eating Disorders, 16, 363–370. http:// dx.doi.org/10.1002/(ISSN)1098-108X. Francis, L. A., & Birch, L. L. (2005). Maternal influences on daughters' restrained eating behavior. Health Psychology, 24, 548–554. http://dx.doi.org/10.1037/0278-6133.24.6. 548. Haller, M. M., & Chassin, L. (2010). The reciprocal influences of perceived risk for alcoholism and alcohol use over time: Evidence of aversive transmission of parental alcoholism. Journal of Studies on Alcohol & Drugs, 71, 588–596.
1407
Humphrey, L. L. (1988). Relationships within subtypes of anorexic, bulimic, and normal families. Journal of American Academy of Child & Adolescent Psychiatry, 27, 544–551. http://dx.doi.org/10.1097/00004583-198809000-00005. Jones, J. W. (1983). The children of alcoholics screening test: A validity study. Bulletin of the Society of Psychologists in Addictive Behaviors, 2, 155–163. Kluck, A. S. (2008). Family factors in the development of disordered eating: Integrating dynamic and behavioral explanations. Eating Behaviors, 9, 471–483. http://dx.doi. org/10.1016/j.eatbeh.2008.07.006. Olson, D. (2011). FACES IV and the Circumplex model: Validation study. Journal of Marital and Family Therapy, 37, 64–80. http://dx.doi.org/10.1111/j.1752-0606.2009.00175.x. Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717–731. http://dx.doi.org/10.3758/BF03206553. Selzer, M. L. (1971). The Michigan alcoholism screening test: The quest for a new diagnostic instrument. American Journal of Psychiatry, 127, 1653–1658. Sher, K. J., Gershuny, B.S., Peterson, L., & Raskin, G. (1997). The role of childhood stressors in the intergenerational transmission of alcohol use disorders. Journal of Studies on Alcohol and Drugs, 58, 414–427. Stice, E., Agras, W. S., & Hammer, L. D. (1999). Risk factors for the emergence of childhood eating disturbances: A five-year prospective study. International Journal of Eating Disorders, 25, 375–387. http://dx.doi.org/10.1002/(SICI)1098-108X(199905)25: 4b375::AID-EAT2N3.3.CO;2-B. Tildesley, E. A., & Andrews, J. A. (2008). The development of children's intentions to use alcohol: Direct and indirect effects of parent alcohol use and parenting behavior. Psychology of Addictive Behaviors, 22, 326–339. http://dx.doi.org/10.1037/0893164X.22.3.326. Vungkhanching, M., Sher, K. J., Jackson, K. M., & Parra, G. R. (2004). Relation of attachment style to family history of alcoholism and alcohol use disorders in early adulthood. Drug & Alcohol Dependence, 75, 47–53. http://dx.doi.org/10.1016/j.drugalcdep.2004. 01.013. Wagener, A.M., & Much, K. (2010). Eating disorders as coping mechanisms. Journal of College Student Development, 24, 203–212. http://dx.doi.org/10.1080/87568225. 486291. West, J.D., Hosie, T. W., & Zarski, J. J. (1987). Family dynamics and substance abuse: A preliminary study. Journal of Counseling & Development, 65, 487–490. White, H. R., Johnson, V., & Buyske, S. (2000). Parental modeling and parenting behavior effects on offspring alcohol and cigarette use: A growth curve analysis. Journal of Substance Abuse, 12, 287–310. http://dx.doi.org/10.1016/S0899-3289(00) 00056-0. Young, E. A., Clopton, J. R., & Bleckley, M. K. (2004). Perfectionism, low self-esteem, and family factors as predictors of bulimic behavior. Eating Behaviors, 5, 273–283. http://dx.doi.org/10.1016/j.eatbeh.2003.12.001.
Contents lists available at ScienceDirect
Addictive Behaviors
Short Communication
Pathways of family influence: Alcohol use and disordered eating in daughters Annette S. Kluck ⁎, Lucille Carriere, Starla Dallesasse 1, Batsirai Bvunzawabaya 2, Erin English, Megan Cobb 3, Therese Borges, Kseniya Zhuzha, Daniel Fry Special Education, Rehabilitation, Counseling (SERC), Auburn University, United States
H I G H L I G H T S • • • • •
Evaluated distal and proximal risk factors for addiction-related coping behaviors Family dynamics directly related to parental addiction-related coping behaviors Family dynamics indirectly related to addiction-related coping behaviors Perceived parental alcohol problems best predicted problematic alcohol use Perceived parental emphasis on appearance best predicted disordered eating
a r t i c l e
i n f o
Available online 28 May 2014 Keywords: Disordered eating Alcohol use Family Behavioral model
a b s t r a c t Models of addiction etiology and treatment emphasize the influence of family-of-origin experiences. Using two addiction-related coping behaviors (ARCBs) common among college women (i.e., problematic alcohol use, disordered eating), we examined whether ARCBs in parents related to matching ARCBs in college women offspring. We expected that matching parental ARCBs would relate more strongly to the ARCBs in offspring than more distal/general family factors. A total of 197 college women completed measures of family dynamics, parental difficulties with alcohol, family focus on appearance and weight, personal difficulties with alcohol use, and disordered eating. A significant indirect effect for family dysfunction on disordered eating and alcohol-related problems was found. That is, family relationship difficulties predicted parents' ARCBs, which predicted matching ARCBs in participants (e.g., parental alcohol problems predicted participant alcohol problems). Matched parental ARCBs were better predictors of participants' ARCBs than more general/distal family factors and non-matched ARCBs. Specifically, path analysis and testing of beta weights supported specificity of parental ARCBs for predicting matching offspring ARCBs. Implications of study findings for tailoring prevention efforts are discussed. © 2014 Published by Elsevier Ltd.
1. Introduction Empirical studies have linked family dysfunction with maladaptive emotion-focused coping behaviors such as those associated with addictive processes (also called addiction-related coping behaviors; ARCBs) related to alcohol use/abuse (Downs & Robertson, 1987; West, Hosie, & Zarski, 1987) and eating disorders (Bruch, 1971; Humphrey, 1988). However, not all individuals originating from families with dysfunctional dynamics develop psychological difficulties. The authors of the present study suggest that parents' problematic coping behavior may
⁎ Corresponding author at: SERC, Auburn, AL 36849, United States. E-mail address: [email protected] (A.S. Kluck). 1 Now at Central Alabama Veterans Health Care System. 2 Now at University of Pennsylvania Counseling and Psychological Services. 3 Now at University of Wisconsin - La Crosse.
http://dx.doi.org/10.1016/j.addbeh.2014.05.015 0306-4603/© 2014 Published by Elsevier Ltd.
provide more specific and proximal sources of risk for a particular ARCB among offspring. Specifically, behavioral theories (e.g., Bandura, Ross, & Ross, 1963) may help explain the tendency for some coping mechanisms to appear to run in families. Addictive behaviors like substance use and disordered eating are often viewed as maladaptive coping mechanisms (e.g., Dube, Anda, Felitti, Edwards, & Croft, 2002; Wagener & Much, 2010). They may reflect somewhat unique psychological difficulties in that they are believed to be a means through which individuals use avoidance to cope with stress, including stress from dysfunctional family environments. In line with behavioral theory (e.g., Bandura et al., 1963), parents may model problematic attitudes and behaviors related to alcohol, body weight, and food as ways of dealing with (through avoidance) difficulties such as family intimacy dysfunction. In addition, addiction research has documented an increased risk in developing an alcohol-related disorder in offspring of alcoholic parents (Sher, Gershuny, Peterson, &
A.S. Kluck et al. / Addictive Behaviors 39 (2014) 1404–1407
Raskin, 1997; Tildesley & Andrews, 2008; Vungkhanching, Sher, Jackson, & Parra, 2004), and parental alcoholism is predictive of the development of problematic alcohol use in offspring after controlling for the potential role of other parental psychological disorders (Chassin, Pitts, DeLucia, & Todd, 1999) and parental relational factors, including parental warmth and hostility (White, Johnson, & Buyske, 2000). Similarly, research indicates that parental modeling of problematic eating may predict problematic eating in their daughters (e.g., Francis & Birch, 2005; Kluck, 2008; Stice, Agras, & Hammer, 1999), and family modeling and use of reinforcement contingencies mediated the relationship between dysfunctional family dynamics and disordered eating (Kluck, 2008). Thus, in families with dysfunctional dynamics, specific ARCBs found among parent(s) may influence the type of ARCBs that offspring develop. If family intimacy dysfunction is a general risk factor and parental modeling behaviors are more specific and proximal risk factors in the development of a particular type of ARCB, modeling of problematic alcohol use in parents should be associated with problematic alcohol use in daughters, and parental modeling of disordered eating attitudes that emphasize thinness and appearance should be associated with disordered eating in daughters. Thus, the purpose of the present study was to test a model where family intimacy dysfunction has an indirect effect on ARCBs in daughters through parents' ARCBs. We hypothesized that (1) family intimacy dysfunction would be associated with ARCBs in parents (i.e., parental alcohol problems, family appearance focus), (2) ARCBs among parents would be associated with increases in similar ARCBs among participants, (3) parents' ARCB (e.g., family appearance focus) that matched the daughters' ARCB (e.g., disordered eating) would be more strongly associated with that specific ARCB among daughters than would general family climate (i.e., intimacy dysfunction) or another type of parental ARCB (e.g., parental problematic alcohol use), and (4) family intimacy dysfunction would have an indirect effect on collegiate women's ARCBs through parents' ARCBs. 2. Method Women (N = 203) recruited from undergraduate psychology courses at a large Southeastern University in the United States served as participants. Only participants under age 25 who were never married were included in study analyses. The resulting sample of 197 participants had a mean age of 20.26 (SD = 1.24). The majority of participants self-identified as Caucasian (81.7%) and heterosexual (99.0%). The balanced cohesion subscale from the Family Adaptability and Cohesion Scales — IV (FACES IV; Olson, 2011) was used to assess participants' perceptions of family intimacy in their family environment with higher scores indicating healthier relationships. ARCBs of parents were assessed using the Children of Alcoholics Screening Test (CAST; Jones, 1983), a measure of perceptions of parental problematic alcohol use during participants' childhoods, and the Family Influence Scale (FIS; Young, Clopton, & Bleckley, 2004), a measure of family emphasis on appearance and thinness. ARCBs of participants were assessed using the Michigan Alcoholism Screening Test (MAST; Selzer, 1971) to measure problematic alcohol use and the Eating Disorder Examination Questionnaire (EDE-Q; Fairburn & Beglin, 1994) to measure disordered eating. Following approval from the university Institutional Review Board, participants were recruited from an undergraduate psychology participant pool. Participants received information about the study purpose, benefits, risks, and their right to discontinue participation without penalty. After providing consent, participants completed a questionnaire packet containing a demographic sheet and the five questionnaires, which were partially counterbalanced to control for order effects. When finished, participants returned the packets to a researcher and received course credit in exchange for their participation.
1405
3. Results All measures were scored such that higher scores were associated with higher levels of the construct. Reliabilities for the study measures were adequate to excellent ranging from .70 (for the MAST) to .97 (for the CAST). Tests of order effects across the four orders were not significant. Consistent with what would be expected if ARCBs in parents were associated with more problematic family functioning, lack of intimacy (as measured by low cohesion, M = 29.74; SD = 5.13) was associated with daughters' perceptions of problematic behaviors in their parents (for the CAST, M = 2.58, SD = 5.59, r = − .22, p b .01; for the FIS, M = 41.23, SD = 10.49, r = − .20, p b .01). Family climate variables that represented the specific perceived parental ARCBs were significantly associated with ARCBs in daughters in the respective domains. Increased perceived parental alcohol use was associated with increased alcohol-related problems among participants (M = 1.52, SD = 1.62), r = .34, p b .01, and increased perceived focus upon appearance and thinness in the family was positively associated with disordered eating (M = 1.88, SD = 1.29), r = .44, p b .01. However, the more diffuse family climate variable (i.e., difficulties with family intimacy) was generally unrelated to problematic behaviors in daughters (for EDE-Q, r = −.04; for MAST, r = −.06). Although participants' scores on the measure of alcohol problems were unrelated to their scores on the measure of disordered eating (r = .07), their perceptions of their parents' alcohol problems were predictive of greater difficulties with disordered eating among participants (r = .17, p b .05). Perceived emphasis on thinness and appearance was unrelated to problematic alcohol use (r = −.05). Next, we examined the predictive power of matched parental– participant behaviors (i.e., alcoholism–alcoholism, appearance focus– disordered eating) compared to family dysfunction and unmatched parental–participant behaviors (e.g., alcoholism–disordered eating) by statistically testing the differences in the regression weights using a modified (with Bonferroni correction) alpha cutoff of .013. Perceived parental alcohol problems were a significantly stronger predictor of participant alcohol problems than were perceived family focus on appearance and thinness (z = 3.76, p b .001) and family cohesion (z = 3.71; p b .001), and there was no significant difference in the predictive power of the latter two variables (z = 0.09, p = .928). Similarly, perceived family focus on appearance and thinness was a significantly stronger predictor of disordered eating than perceived parental alcohol problems (z = 2.74, p = .006) or family cohesion (z = 3.87; p b .001), and there was no significant difference in the predictive power of the latter two variables (z = 1.89; p = .059). As such, only the perceived parental ARCB that matched the ARCB in participants was significantly stronger as a predictor than other family climate variables. Following the recommendations by Preacher and Hayes (2004), we used a bootstrapping technique for the Sobel test to test for the presence of a significant indirect effect. Sobel (z = − 2.65, p = .008; z = − 2.58, p = .010) and bootstrapping (95% CI = − .046 to − .007; 95% CI = −.038 to − .008) approaches revealed a significant indirect effect for perceived parental problems in the relationships between indicators of perceived family functioning (i.e., cohesion) and problematic alcohol use and disordered eating, respectively, in participants. We used path analysis to further test for specificity of the parental ARCB model. Fig. 1 includes three sets of path coefficients for three models (i.e., the full model containing both direct and indirect effects, the model containing only indirect effects, the model with paths between unmatched ARCBs). The direct paths between cohesion to the participants' ARCBs in the full effects models were near zero. The full effects model had adequate fit (Table 1). The indirect effects only model also had adequate fit (Table 1). The χ2 difference between the two models was not significant (p = .748), suggesting that dropping the direct effects paths did not result in over-trimming of the model. We ran a third model, referred to as the crossover model, in which we allowed
1406
A.S. Kluck et al. / Addictive Behaviors 39 (2014) 1404–1407
CASTTOT -0.24 (-0.24) -0.24
0.33 (0.34) 0.34
MASTTOT
(0.02) -0.01
0.02 COHESION (0.05)
0.12
0.07
-0.20 (-0.20) -0.20
0.44 (0.45) 0.44
FISTOT
EDETOT
Fig. 1. Indirect and full effects models of family climate predicting ARCBs in study participants. Perceived family dynamics (measured by FACES IV cohesion balance subscale) are indirectly (and directly in the case of the full model, values for this model appear in parentheses) related to problematic behavior in participants through perceived parental behaviors. CASTTOT represents perceived problems with alcohol in parents, MASTTOT represents self-reported problems with alcohol among participants, FISTOT represents perceived family emphasis on appearance and weight, and EDETOT represents self-reported disordered eating. Plain text path coefficients in parentheses reflect the full effects model (with a large dashed line), those outside of parentheses reflect the indirect effects only model (with a solid line), and those in italics reflect the model with unmatched ARCB paths included (with a small dashed line).
perceived ARCBs to predict unmatched and matched participant ARCBs (see Fig. 1). Consulting the Wald and LaGrange tests revealed that dropping the paths between perceived family focus on appearance and thinness and participant difficulties with alcohol, and between family cohesion and participant difficulties with alcohol, was estimated to improve the fit of the model. Despite having adequate fit (Table 1), the χ2 difference (p = .435) between the indirect effects only model and the crossover model revealed that the crossover model did not fit better than the indirect effects only model. In addition, the Akaike Information Criterion (AIC) revealed slightly less good fit for the third model, supporting the hypothesis that perceived parent ARCBs have specificity for ARCBs in participants.
4. Discussion Children can be influenced by their parents' modeling of certain behaviors and through covert messages that communicate parents' beliefs and expectations (White et al., 2000). The authors expected that perceived parental ARCBs would be more proximal sources of influence in the development of ARCBs among offspring than perceived problems in family dynamics. Our results support the notion that family dynamics might relate to the presence of other problems within the family (like parent ARCBs) and that the type of ARCBs exhibited by daughters may be influenced by the ARCBs modeled by their parents. The relationships between family dynamics and ARCBs among participants were indirect and more distal than were the relationships between perceived parental ARCBs and self-reported ARCBs in participants. The findings are consistent with the idea that perceived family dysfunction may create an environment where problematic emotion-based coping behaviors occur, but it is the specific perceived problematic behaviors in parents that most relate to the specific difficulties in offspring. The present study contains several limitations to consider. These include use of a correlational design that prohibits conclusions regarding
causality, reliance on self-report of a single informant, and use of a homogenous sample of young adults. The findings from the present study have potential implications for mental health professionals. First, knowing about parental ARCBs may be useful in identifying current or potential ARCBs in young women. Second, in the treatment of an adult woman with an addiction, consideration of prevention efforts for offspring may be warranted (e.g., referring other family members for therapy, support or 12-step groups, or psychoeducation). Haller and Chassin (2010) found that offspring of alcoholics who perceived alcohol consumption to be more risky used alcohol less than offspring who did not perceive alcohol consumption as risky, pointing to the possibility of modifying the pattern of intergenerational transmission. Third, it may be helpful to determine the types of ARCBs found in the parents and family members of young adult women seeking treatment because these may relate to current or potential ARCBs in those women. In some cases, therapy, psychoeducational interventions (should parents or family members hold misconceptions about alcohol use and eating/weight or lack knowledge about healthy coping mechanisms), or other supports may be needed for family members of individuals who struggle with problems with alcohol use (such as Al-ANON) and disordered eating. Role of funding sources There were no funding sources for conducting the research or preparing the manuscript. Contributors The authors and their contributions are described below. All authors approve the final manuscript. Annette S. Kluck: designed study, wrote protocol, coordinated data collection, conducted literature searches, conducted statistical analyses, wrote multiple draft of manuscript. Lucille J. Carriere: designed study, wrote protocol, collected data, conducted literature searches, co-wrote first draft of manuscript. Starla Dallesasse: wrote protocol, collected data, entered data, co-wrote multiple drafts of manuscript.
Table 1 Fit indices for path analyses. (Fig. 1). Model
χ2
p
χ2/df
CFI
NNFI
SRMR
RMSEA (90% CI)
AIC
Cohesion indirect & direct effects Cohesion indirect effects only Cohesion crossover model
4.52 5.10 1.31
0.34 0.53 0.52
1.13 0.85 0.66
.99 1.00 1.00
.98 1.02 1.05
.05 .05 .02
.03 (.00–.11) .00 (.00–.09) .00 (.00–.13)
−3.48 −6.90 −2.69
Note, none of the χ2 were significant at p b .05. CFI = Comparative fit index; NNFI = Bentler–Bonnet non-normed fit index; SRMR = standardized root mean-square residual; RMSEA = root mean-square error of approximation; CI = confidence interval; AIC = Akaike Information Criterion.
A.S. Kluck et al. / Addictive Behaviors 39 (2014) 1404–1407 Batsirai Bvunzawabaya: wrote protocol, collected data, entered data, co-wrote multiple drafts of manuscript. Erin English: conducted literature searches, assisted in writing final draft of manuscript. Megan Cobb: collected data. Therese Borges: conducted literature searches, assisted in writing final draft of manuscript. Kseniya Zhuzha: conducted literature searches, assisted in writing final draft of manuscript. Daniel Fry: conducted literature searches, assisted in writing final draft of manuscript. Conflict of interest The authors have no conflicts of interest to disclose. Acknowledgments The authors thank Sophie Ahmad for her assistance with data collection and Natalie Reiner and Sally Kirklewski for their assistance with manuscript preparation.
References Bandura, A., Ross, D., & Ross, S. A. (1963). Vicarious reinforcement and imitative learning. The Journal of Abnormal and Social Psychology, 67, 601–607. http://dx.doi.org/10.1037/ h0045550. Bruch, H. (1971). Family transactions in eating disorders. Comprehensive Psychiatry, 12, 238–248. http://dx.doi.org/10.1016/0010-440X(71)90021-6. Chassin, L., Pitts, S.C., DeLucia, C., & Todd, M. (1999). A longitudinal study of children of alcoholics: Predicting young adult substance use disorders, anxiety, and depression. Journal of Abnormal Psychology, 108, 106–119. http://dx.doi.org/10.1037/0021-843X. 108.1.106. Downs, W. R., & Robertson, J. F. (1987). Family dynamics and adolescent alcohol use: A path analysis with a treatment sample. Journal of Applied Social Sciences, 12, 24–45. Dube, S. R., Anda, R. F., Felitti, V. J., Edwards, V. J., & Croft, J. B. (2002). Adverse childhood experiences and personal alcohol abuse as an adult. Addictive Behaviors, 27, 713–725. http://dx.doi.org/10.1016/S0306-4603(01)00204-0. Fairburn, C. G., & Beglin, S. J. (1994). Assessment of eating disorders: Interview or selfreport questionnaire? International Journal for Eating Disorders, 16, 363–370. http:// dx.doi.org/10.1002/(ISSN)1098-108X. Francis, L. A., & Birch, L. L. (2005). Maternal influences on daughters' restrained eating behavior. Health Psychology, 24, 548–554. http://dx.doi.org/10.1037/0278-6133.24.6. 548. Haller, M. M., & Chassin, L. (2010). The reciprocal influences of perceived risk for alcoholism and alcohol use over time: Evidence of aversive transmission of parental alcoholism. Journal of Studies on Alcohol & Drugs, 71, 588–596.
1407
Humphrey, L. L. (1988). Relationships within subtypes of anorexic, bulimic, and normal families. Journal of American Academy of Child & Adolescent Psychiatry, 27, 544–551. http://dx.doi.org/10.1097/00004583-198809000-00005. Jones, J. W. (1983). The children of alcoholics screening test: A validity study. Bulletin of the Society of Psychologists in Addictive Behaviors, 2, 155–163. Kluck, A. S. (2008). Family factors in the development of disordered eating: Integrating dynamic and behavioral explanations. Eating Behaviors, 9, 471–483. http://dx.doi. org/10.1016/j.eatbeh.2008.07.006. Olson, D. (2011). FACES IV and the Circumplex model: Validation study. Journal of Marital and Family Therapy, 37, 64–80. http://dx.doi.org/10.1111/j.1752-0606.2009.00175.x. Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717–731. http://dx.doi.org/10.3758/BF03206553. Selzer, M. L. (1971). The Michigan alcoholism screening test: The quest for a new diagnostic instrument. American Journal of Psychiatry, 127, 1653–1658. Sher, K. J., Gershuny, B.S., Peterson, L., & Raskin, G. (1997). The role of childhood stressors in the intergenerational transmission of alcohol use disorders. Journal of Studies on Alcohol and Drugs, 58, 414–427. Stice, E., Agras, W. S., & Hammer, L. D. (1999). Risk factors for the emergence of childhood eating disturbances: A five-year prospective study. International Journal of Eating Disorders, 25, 375–387. http://dx.doi.org/10.1002/(SICI)1098-108X(199905)25: 4b375::AID-EAT2N3.3.CO;2-B. Tildesley, E. A., & Andrews, J. A. (2008). The development of children's intentions to use alcohol: Direct and indirect effects of parent alcohol use and parenting behavior. Psychology of Addictive Behaviors, 22, 326–339. http://dx.doi.org/10.1037/0893164X.22.3.326. Vungkhanching, M., Sher, K. J., Jackson, K. M., & Parra, G. R. (2004). Relation of attachment style to family history of alcoholism and alcohol use disorders in early adulthood. Drug & Alcohol Dependence, 75, 47–53. http://dx.doi.org/10.1016/j.drugalcdep.2004. 01.013. Wagener, A.M., & Much, K. (2010). Eating disorders as coping mechanisms. Journal of College Student Development, 24, 203–212. http://dx.doi.org/10.1080/87568225. 486291. West, J.D., Hosie, T. W., & Zarski, J. J. (1987). Family dynamics and substance abuse: A preliminary study. Journal of Counseling & Development, 65, 487–490. White, H. R., Johnson, V., & Buyske, S. (2000). Parental modeling and parenting behavior effects on offspring alcohol and cigarette use: A growth curve analysis. Journal of Substance Abuse, 12, 287–310. http://dx.doi.org/10.1016/S0899-3289(00) 00056-0. Young, E. A., Clopton, J. R., & Bleckley, M. K. (2004). Perfectionism, low self-esteem, and family factors as predictors of bulimic behavior. Eating Behaviors, 5, 273–283. http://dx.doi.org/10.1016/j.eatbeh.2003.12.001.
