افسردگی فعلی مادر رابطه ابراز احساسات مهم در مادران و نشانه های افسردگی در دختران نوجوان آنها را تعدیل می کند

Abstract Prior studies have examined critical expressed emotion (EE-Crit) in mothers in the intergenerational transmission of depression. However, the potential moderating effect of maternal depression diagnostic status in relation to EE-Crit and youth depressive symptoms has yet to be determined. A total of N=121 biological mother/daughter dyads that differed in maternal depression diagnostic status were recruited for the present study: (1) currently depressed mothers (current depression, n=29); (2) formerly depressed mothers (past depression, n=39); and (3) mothers free from any psychiatric history (healthy controls, n=53). Mothers were administered structured clinical interviews and completed self-report measures of EE-Crit and psychopathology, and daughters self-reported depressive symptoms. Results indicated no significant group differences in EE-Crit; however, current maternal depression status moderated EE-Crit such that the magnitude of the relation between EE-Crit and adolescent depressive symptoms was significantly greater in daughters of currently depressed mothers. These findings highlight the importance of considering current maternal depression, rather than a history of maternal depression, in relation to EE-Crit and adolescent depressive symptoms, providing impetus for future investigations.

1. Introduction Major Depressive Disorder (MDD) is a debilitating clinical syndrome that typically emerges during adolescence (Hankin et al., 1998). MDD is highly recurrent (Boland and Keller, 2009) and even subclinical depressive symptoms during adolescence are predictive of MDD in adulthood (Pine et al., 1999). Importantly, adolescence marks the time in which depression comes to disproportionately affect females, resulting in the 2:1 female to male ratio seen in adult depression (Rudolph, 2009). For these reasons among others, it is imperative to examine the factors that influence depressive symptoms among high-risk adolescents, particularly adolescent girls. While a variety of psychosocial factors increase risk for childhood depression (i.e., stressful life events, childhood abuse or neglect; Bhatia and Bhatia, 2007), maternal depression has been consistently implicated as a predominant risk factor for children of all ages—from infancy through adolescence (Goodman and Gotlib, 1999). Children of depressed mothers are at greater risk for developing psychopathology, with a specific risk for depression (Downey and Coyne, 1990). Maternal depression has adverse effects on the child׳s attachment style (Jameson et al., 1997) and contributes to negative, disengaged parent–child interactions (Lovejoy et al., 2000). Depressed mothers make more negative, critical, and hostile comments about their children (Brennan et al., 2002), further increasing risk for youth depression. Not surprisingly then, several studies have considered the effects of expressed emotion (EE) in mothers in relation to childhood (child and adolescent) depression. EE is a construct that refers to the presence of negative components such as criticism, hostility, and/or emotional over-involvement (EOI), and positive components such as warmth and positive comments (Hooley, 2007). This growing body of literature points to EE, particularly high maternal critical EE (EE-Crit), as a significant factor in the development of childhood depression (Burkhouse et al., 2012). Some of these studies have employed high-risk designs to examine youth at high familial risk for depression, before they experience a major depressive episode. These study designs are especially informative as they allow for the identification of factors that may increase vulnerability to depression (Gotlib et al., 2010). We are aware of three high-risk design studies that investigated the relations between maternal depression, EE-Crit, and youth depression (Tompson et al., 2010, Burkhouse et al., 2012 and Silk et al., 2009). We present these findings to highlight the established relations between these variables; however, it is not our intent to investigate causal relations in the current study. Conducting a high-risk cross-sectional study with children (ages 8–12) of mothers with a history of depression, Tompson et al. (2010) found that maternal depression was positively associated with EE-Crit. Both maternal depression history and EE-Crit were significantly associated with MDD in children; however, maternal depression history was the stronger predictor of child depression. Additionally, maternal depression history did not moderate the relation between EE-Crit and child depressive symptoms, but it is important to note that the authors made no distinction between current and past maternal depression. In examining EE-Crit׳s contribution to childhood initial MDD onset, Burkhouse et al. (2012) conducted a multi-wave high-risk longitudinal study of children (ages 8–12) of mothers with or without a history of depression. EE-Crit predicted children׳s depression onset over a 14-month period, even after controlling for maternal and child depressive symptoms at a 6-month follow-up. Maternal depression status, defined as a history of MDD during the child׳s lifetime, was examined as a moderator of EE-Crit and results were non-significant; however, this was attributed to low statistical power due to an insufficient sample size. Finally, a high-risk longitudinal study conducted by Silk et al. (2009) examined EE-Crit among mothers of children (ages 8–19) with differential depression statuses. EE-Crit was significantly higher among mothers of currently depressed, remitted, and high-risk children as compared to healthy controls. No significant differences were found between the first three groups on EE-Crit. Additionally, EE-Crit was associated with an increased likelihood of depressive onset in children at-risk or with a history of depression, suggesting a potential causal role of EE-Crit. Thus, Silk et al.׳s (2009) study emphasizes the transactional nature of EE-Crit. Collectively, the literature suggests that maternal depression is positively associated with EE-Crit, and EE-Crit appears to be a significant, independent predictor of youth depression. While this research is highly informative, the potential moderating effect of current, as compared to past, maternal depression on the relation between EE-Crit and adolescent depression has yet to be determined. Prior high-risk studies have generally operationalized maternal depression as mothers with a lifetime history of depression ( Tompson et al., 2010) or mothers who had a major depressive episode during the child׳s lifetime ( Burkhouse et al., 2012). This failure to distinguish between currently and formerly depressed mothers in the study of EE-Crit likely has implications, as distinct behavioral and emotional changes occur in mothers during the course of major depressive episodes. There are many ways in which current maternal depression is associated with compromised parenting (see Joormann et al., 2009). For instance, currently depressed mothers are more critical and disengaged from their children ( Cox et al., 1987). Moreover, current depression, rather than a history of psychopathology, is predictive of this negative parent–child interaction style ( Adrian, 1989). Taken together, these findings suggest that current, rather than past, maternal depression may distinctly influence the relation between EE-Crit and child depressive symptoms. Against this background, the present cross-sectional, high-risk study examined whether the relation between EE-Crit and depressive symptoms in adolescent daughters was moderated by maternal depression diagnostic status (current depression, past depression, and healthy controls). We were specifically concerned with depressive symptoms in adolescent girls due to the fact that a disproportionate number of girls become affected by depression during adolescence (Rudolph, 2009). Moreover, there are important female-specific factors, particularly in regards to social cognition (McClure, 2000) and interpersonal style (Purdie and Downey, 2000) that makes girls particularly susceptible to depressive reactions to EE-Crit and maternal depression (Davies and Windle, 1997). We had two hypotheses: 1) mothers with current depression would report significantly higher EE-Crit than mothers with past depression and healthy controls; and 2) maternal depression status would moderate the relation between EE-Crit and depressive symptoms in daughters, such that the magnitude of the relation between EE-Crit and depressive symptoms would be greater among mothers with current depression. While positive findings would not establish a causal relationship between EE-Crit and adolescent depression, they would suggest that emphasizing the treatment of EE-Crit, particularly in currently depressed mothers, is warranted. By examining EE-Crit specifically, rather than the general parenting impairments associated with maternal depression (i.e., less warmth and nurture), we aim to provide therapists with a well-operationalized treatment target.

3. Results 3.1. Participant characteristics For participant characteristics and group differences, see Table 1. Of note, daughters in the current depression group reported significantly higher MFQ scores than healthy controls (p=0.001), and there were no significant differences between groups on EE-Crit (p=0.302). 3.2. Moderation analyses Hierarchical multiple regression analyses were conducted to examine whether maternal depression status (current depression, past depression, or healthy control) moderated the relation between EE-Crit and adolescent depressive symptoms, controlling for general severity of maternal psychopathology (GHQ total scores). In step 1, GHQ scores were entered, with MFQ scores as the dependent variable. The model was significant, R2=0.10, F (1, 119)=12.71, p=0.001, accounting for 10% of the variance in MFQ scores. In step 2, dummy-coded group variables and FAS scores were entered as predictors. The model remained significant, R2=0.16, F (4, 116)=5.42, p<0.001, accounting for 16% of the variance; however, individual predictors were non-significant (p׳s ≤0.079). In step 3, group×FAS interaction terms were added, and the model account for an additional 7% of MFQ score variance, R2=0.23, F (5, 115)=5.65, p<0.001. There was a significant interaction between current depression status and FAS scores, β=0.89, SE=0.31, t=2.93, p=0.004, such that the magnitude of the relation between EE-Crit and depressive symptoms in daughters was significantly greater among currently depressed mothers (see Fig. 1). No other interactions were significant, but past depression group status had a main effect, β=0.58, SE=0.28, t=2.08, p=0.040, such that daughters of mothers with remitted depression reported significantly higher depressive symptoms compared to healthy controls. Of note, FAS scores were non-significant in the final model, β=−0.03, SE=0.21, t=−0.12, p=0.905. Current maternal depression moderates the relation between EE-Crit and ... Fig. 1. Current maternal depression moderates the relation between EE-Crit and adolescent depressive symptoms. High vs. low EE-Crit was determined by a median split of FAS scores within diagnostic groups. MFQ scores are standardized, M=0, S.D.=1. Current=depressive symptoms among daughters of mothers with current depression; Past=depressive symptoms among daughters of mothers with past depression; Healthy control=depressive symptoms of daughters of mothers with no psychiatric history.