Bridging Disciplinary Boundaries (January 11 - 14, 2007)

PURPOSE: Depression is a disorder seen frequently after childbirth affecting as many as 19% of women (Gavin et al., 2005). Maternal depression has deleterious effects on mothers Misri et al., 2000). Moreover, untreated maternal depression is associated with poor parenting outcomes and can have a significant impact on the offspring's future cognitive, emotional and social development (Dennis, 2004). The literature suggests an association between depression and marital status and social support (Beck, 2001). Yet, little is known about specific aspects of social support, particularly those associated with the relationship quality between mother and biological father. Our objective is to describe the prevalence and predictors of maternal depression with an emphasis on relationship status and quality.

METHODS: Data are from the Fragile Families and Child Wellbeing Study (FFCW), a national survey designed to analyze new unwed parents and their children. Following approximately 5,000 children born in 1998-2000 and their parents, the study over-sampled births to unmarried couples; data are representative of non-marital births in large U.S. cities. This is the first longitudinal study to focus on unwed urban parents and the first study to include data from non-cohabiting unwed fathers. In addition to detailed socioeconomic information, the study provides data about relationship status and quality. Relationship status of participants is married/cohabiting; not cohabiting yet romantically involved/not romantically involved but visiting; and no contact. DSM-IV depression was measured by the Composite International Diagnostic Interview (CIDI), developed by the World Health Organization (WHO) to assess 12-month prevalence. We included mothers who have completed the first year follow-up and those who completed the CIDI questions (N = 4,348). Chi-Square statistics were used to identify bivariate association of depression with relationship status, quality, and other key variables identified in the literature (e.g., other social support, infant characteristics and socioeconomic indicators). We used logistic regressions for multivariate analyses modeling depression.

RESULTS: Prevalence of depression was 12%. Mothers cohabiting with the father at birth were nearly half as likely to be depressed in bivariate analysis (16 vs. 9%, p<0.01). Yet, once relationship quality within the first year was controlled for, effect of relationship status was not significant. Those with fair/poor relationships were approximately three times more likely to be depressed compared to those that rated relationship quality as excellent (odds ratio = 2.91, p<0.01). Some characteristics of the infant (very low birth weight, fussy, fair/poor health) were significantly associated with higher rates of depression. Women with three or more children were also more likely to be depressed (odds ratio = 1.65, p<0.01). These factors remained significant after controlling for relationship status and quality.

IMPLICATIONS: The identification of risk factors is important because preventive measures--early detection and treatment-- can alleviate suffering and potentially harmful effects on maternal health, infant outcomes and family well-being. Our results suggest that social work assessment, intervention and research in health and mental health settings ought to go beyond relationship status and focus on the quality of the relationship.