Generalized Anxiety Disorder or Poverty:? A Study of Poor Mothers and Their Children

The nosology of Generalized Anxiety Disorder (GAD) is highly controversial in part because diagnoses are often confounded by the criteria of other mood disorders. Even more notable is the fact that GAD is commonly diagnosed in the socially disadvantaged, especially those who live in poverty (Blazer, Hughes, & George, 1991; Brawman-Mintzer, & Lydiard, 1996; Kessler & Wittchen, 2002; Wittchen, Zhao, Kessler et al., 1999). The relationship between poverty and the etiology of GAD is not well understood. However, stress is implicated. Stress research proposes a stress response which affects parental functioning and especially levels of involvement with offspring (Belsky, 1984). Anxiety may also be involved. Thus, we investigated anxiety, parental stress and levels of maternal involvement in a sample of poor mothers. Our hypotheses were that poverty would predict anxiety in mothers and their offspring and would also affect maternal levels of involvement. Racial, ethnic and gender differences were also examined.

Method:

This was a secondary analysis of the Fragile Families data and our sample included 4,116 (Age: M = 28.23, SD = 6.09, range = 16 to 53) out of 4,231 mothers who were interviewed at the 3-year follow-up mother survey. The focal child's average age at the time of mother interview was 35.74 months (SD = 2.56, range = 30 to 50). Analysis consisted of Structural Equation Modeling with Mplus 5.0. The probability of mother's generalized anxiety disorder (GAD) was measured by the Composite International Diagnostic Interview - Short Form (CIDI-SF: Kessler, Andrews, Mroczek, Ustun, & Wittchen, 1998) a standardized instrument for use in epidemiological, cross-cultural, and other research studies. A follow-up question captured symptom duration. Maternal involvement was measured by activities representing the mother's engagement with the child. Poverty was measured by ability to pay utilities, rent, buy food, etc. The child's anxiety was measured via maternal reports of anxious symptomatology in the child. Parental stress was measured by asking about difficulties in parenting such as feeling worn out, feeling overwhelmed with parenting, etc.

Results

The model had good fit, Chi-Square = .000, CFI = .99, TLI = .99, RMSEA .05. Findings showed that the odds of mothers reporting parental stress was significant for the poorest mothers (OR 33, p<.000, 95% CI) and for mothers with anxious children (OR 1.05, p<.02, 95% CI). Furthermore, the poorest mothers reported fewer interactions with their children (OR 2.5, p<.000, 95% CI). The GAD mothers were less likely to report anxiety symptoms in their children (OR .96, p<.05, 95% CI), and as in previously published studies, we found that all mothers were more likely to report anxiety symptoms in their female than male offspring (OR 2.2, p<.000, 95% CI). There were no significant racial/ethnic differences or differences by marital status.

Discussion:

The findings suggest focal points for social work intervention as well as underscore the importance of both social and financial support for poor families. While we were not able to capture the direct effects of poverty on adult or childhood anxiety, such an explication maybe more promising in longitudinal analysis.