Change in Maternal Depressive Mood: Unique Contributions to Adolescent Functioning over Time
Thomas, Amanda McCombs, Forehand, Rex, Neighbors, Bryan, Adolescence
There is substantial literature on the relationship between maternal depression and child adjustment (e.g., Downey & Coyne, 1990). Mothers diagnosed with major depressive disorder have been found to have children with higher rates of psychiatric disorders in general (Orvaschel, Walsh-Allis, & Ye, 1988), dysthymia (Klein, Clark, Dansky, & Margolis, 1988), cognitive deficits (Kaplan, Beardslee, & Keller, 1987), and a full range of adjustment problems (Downey & Coyne, 1990). Even less severe forms of maternal depression have been associated with difficulties in child functioning (for a review, see Forehand, McCombs, & Brody, 1987). Thus, parental depression, whether defined as a disorder or simply as mood, predicts child adjustment.
Most studies concerned with prediction of child problems resulting from parental depression have utilized a cross-sectional design. An arguably more effective method of studying the relationship between maternal depression/depressed mood and child functioning is a longitudinal design. With this type of design, one can begin to move from simple correlations to data which are temporally congruent with the causal ordering of variables. Longitudinal data also can address other questions which are not open to examination by cross-sectional data. The current project utilizes such a design to examine one such question: Do changes in level of maternal depression contribute to child functioning beyond the static level of depression? One study in the literature sheds some preliminary light on this question.
Billings and Moos (1985) examined the longitudinal changes in child and family functioning when parental depression either remits or continues. Their clinically depressed sample was divided into two groups at follow-up: one in a state of remission (no moderate or severe symptoms of depression in the last month); and one nonremitted (continuing to report moderate to severe symptoms of depression). At a one-year follow-up, children of remitted parents exhibited significantly less dysfunction than did children of nonremitted parents. This study suggests that the change in maternal clinical depression is associated with changes in child functioning.
The current project was designed to extend the Billings and Moos study by examining whether changes in less severe forms of depression (i.e., maternal depressive mood) similarly relate to child functioning. This is an important issue since variations in mood are more common than clinical depression, and potentially influence far more children. The current study also improved on the Billings and Moos study by statistically controlling for the initial level of maternal depressed mood, rather than changing the labeling of parents from depressed to nondepressed.
We hypothesized that maternal mood which worsens over time would contribute to the prediction of adolescent problems beyond the static level of mood ascertained in an initial assessment in a sample of early adolescents. Important caveats to take into account when examining the current question are gender of child, marital status of mother, raters of child functioning, and aspects of child functioning that are measured. Susceptibility to psychological problems (e.g., Graham, 1979) and resiliency to stressors in general in early adolescence (Simmons, Burgeson, Carlton-Ford, & Blyth, 1987) differ by gender; therefore, gender of adolescent was taken into account. Marital status has been shown repeatedly to be related to child functioning (e.g., Emery, 1988) and to maternal depressive mood, with depression being a mechanism through which divorce may affect child functioning (Forehand, Thomas, Wierson, Brody, & Fauber, 1990). Therefore, the potentially moderating influences of marital status on the relation between change in maternal depressive mood and child functioning were examined. Raters of adolescent adjustment should not be the depressed mothers as previous research indicates they may have a negative bias in reporting child behavior (Griest, Wells, & Forehand, 1979). …