Parental Alcoholism and Adolescent Depression? Does Adolescent-Parent Communication Moderate the Relationship Between Parental Alcoholism and Adolescent Depression? Christine McCauley Ohannessian, University of Delaware The primary aim of this study was to explore whether adolescent-parent communication moderates the relationship between parental alcoholism and adolescent depression. Surveys were administered to 1,001 adolescents in the spring of 2007. Regression results indicated that maternal alcoholism predicted depression for girls, but not for boys. In addition, adolescent-mother communication problems and (low) adolescent-mother open communication predicted depression for girls. Adolescent-father communication problems predicted depression for both girls and boys. Importantly, open adolescent-parent communication (maternal and paternal) protected girls with problem drinking fathers from experiencing high levels of depression. Abstract Measures Adolescent Depression. The Center for Epidemiological Studies Depression Scale for Children (CES-DC; Weissman et al., 1980) was used to measure adolescent depressive symptomatology. Participants were asked to respond to the CES-DC items in regard to how they felt or acted during the past week. A sample item from this measure is “I felt sad.” The response scale ranges from 1 = not at all to 4 = a lot. The 20 CES-DC items were summed to create a total score. The CES-DC has been shown to be a reliable and valid measure of depressive symptomatology (Faulstich et al., 1986). The Cronbach alpha coefficient for the CES-DC total score based on this sample was .90. Procedures Sample During the spring of 2007, adolescents who provided assent, and who had parental consent, were administered a self-report survey in school by trained research personnel. The survey took approximately 40 minutes to complete. After completing the survey, the adolescents were given a movie pass for their participation. ◦ 1,001 10th and 11th grade high school students (53% girls) ◦ 57% Caucasian; 22% African-American; 12% Hispanic; 2% Asian ◦ Mean age = 16.09 (SD=.68); age range 15-17 ◦ All participants attended a public high school in Delaware, Pennsylvania, or Maryland ◦ Most of the adolescents (56%) lived with both biological parents; 89% lived with their biological mother and 61% lived with their biological father The interaction model (Model 2) only was significant for girls (F(4,375)=5.64, p<.001, ΔR2=.05). For girls, significant interactions were found between paternal problem drinking and adolescent-father communication problems (β = .17, p<.01) and between paternal problem drinking and adolescent-mother communication problems (β = -.12, p<.05). These interactions suggested that girls who had fathers with low levels of problem drinking were more depressed if they experienced high levels of communication problems with their parents. In addition, for girls, significant interactions were found between paternal problem drinking and adolescent-father open communication (β=-.18, p<.01) and between paternal problem drinking and adolescent-mother open communication (β=-.11, p<.05). As shown in Figures 1 and 2, adolescent girls who had fathers with high levels of problem drinking had relatively lower levels of depression if they had high levels of open communication with their parents. Maternal Problem Drinking The main effects maternal problem drinking model (Model 1) was significant for girls (F(5,384)=20.89, p<.001, R2=.21) and boys (F(5,314)=8.29, p<.001, R2=.12). In addition to the significant main effects found in the paternal drinking model, maternal problem drinking significantly predicted depression for girls, but not for boys (β=.12, p<.05). The interaction model (Model 2) was significant for boys (F(4,310)=5.69, p<.001, ΔR2=.06), but not for girls. A significant interaction was observed between maternal problem drinking and adolescent-mother open communication for boys (β=.29, p<.05), indicating that adolescent-mother open communication was more closely linked to depression for boys who had mothers with relatively low levels of problem drinking. Analyses Results Pearson product-moment correlations were calculated to examine the bivariate associations between the study variables. Next, hierarchical regression analyses were conducted to examine whether adolescent-parent communication moderates the relationship between parental problem drinking and adolescent depression. These analyses were conducted separately by the gender of the adolescent. Step 1 (Model 1) included parental problem drinking (paternal or maternal) and the adolescent-parent communication variables. Step 2 (Model 2) included these variables and the two-way interaction terms between parental problem drinking and the adolescent-parent communication variables. Measures Parental Problem Drinking. A revised 9-item version of the Short Michigan Alcoholism Screening Test (M-SMAST/F-SMAST; Crews & Sher, 1992) was used to assess alcohol problems in mothers and fathers. A sample item is “Has your mother/father ever attended a meeting of Alcoholics Anonymous?” The items were summed to create separate total scores for mothers and fathers. Because this measure yields continuous scores, parental alcoholism is referred to as problem drinking. In this study, adolescents who responded in regard to non-biological parents were not included. Cronbach alpha coefficients for the M-SMAST and the F-SMAST based on this sample were .80 and .86, respectively. Conclusions In the present study, maternal problem drinking predicted depression for girls, but not for boys. In addition, adolescent-mother communication problems and (low) adolescent-mother open communication predicted depression for girls, but not for boys. Adolescent-father communication problems predicted depression for both boys and girls. Taken together, these results are generally consistent with social learning theory (Bussey & Bandura, 1984), which suggests that children are more likely to emulate same-sex role models than opposite-sex role models. In addition, the findings are in line with research on parental psychopathology suggesting that psychopathology in the same-sex parent may have a more profound effect on adolescent adjustment than psychopathology in the opposite-sex parent (e.g., Crawford et al., 2001). Gender differences also were observed for the moderation analyses. As noted, open communication was found to “protect” girls with problem drinking fathers from experiencing high levels of depressive symptomatology. However, open communication was not found to have the same “protective” effect for boys . Results from this study underscore the importance of considering both the gender of the parent and the gender of the adolescent when examining the adolescent-parent relationship. Results Bivariate Relationships As shown in Table 1, paternal problem drinking and maternal problem drinking were positively associated with one another. Both paternal problem drinking and maternal problem drinking were related to depression for girls, but not for boys. Open communication was negatively associated with depression and communication problems were positively associated with depression for both girls and boys. Adolescent-Parent Communication. The 20-item Parent-Adolescent Communication Scale (Barnes & Olson, 2003) was used to measure communication between adolescents and their parents. This measure includes two subscales – open family communication and problems in family communication. Respective sample items are “I find it easy to discuss problems with my mother/father” and “There are topics I avoid discussing with my mother/father.” The response scale ranges from 1 = strongly disagree to 5 = strongly agree. Separate scale scores were calculated for communication with mothers and fathers. Cronbach alpha coefficients were .92 for adolescent-mother open communication, .78 for adolescent-mother communication problems, .94 for adolescent-father open communication, and .82 for adolescent-father communication problems. Regression Models - Paternal Problem Drinking The main effects paternal problem drinking model (Model 1) was significant for both girls (F(5,379)=19.64, p<.001, R2=.21) and boys (F(5,316)=8.49, p<.001, R2=.12). Adolescent-mother communication problems and (low) adolescent-mother open communication significantly predicted depression for girls (β = .17, p<.01; β = -.23, p<.001, respectively). In addition, adolescent-father communication problems significantly predicted depression for girls (β = .12, p<.05) and boys (β = .17, p<.01). To obtain a copy of this poster, please visit www.aap.udel.edu. *Supported by NIAAA K01AA015059