*The full article is available through this link. This article may be available free of charge to those with university credentials.
This study examined the levels of depression reported by women who had used a domestic violence shelter. Depressive symptoms were assessed three times: immediately after shelter exit, 10 weeks thereafter, and 6 months later. Whereas 83% of the women reported at least mild depression on the Center for Epidemiological Studies Depression (CES-D) scale upon shelter exit, only 58% were depressed 10 weeks later. This did not change at the 6-month follow-up. An ecological, longitudinal model was evaluated to predict battered women's depression 8 1/2 months postshelter exit. Results of hierarchical regression analyses suggested that, after controlling for previous levels of depression, the women's feelings of powerlessness, experience of abuse, and decreased social support contributed to their depression symptoms. The women's scores on these three variables (feelings of powerlessness, abuse, and social support) at 10 weeks postshelter exit and at 6-month follow-up predicted depression at 6 months. Thus, there were both predictive and concurrent effects for these constructs. Implications for clinical and community interventions are discussed.