DATA FROM THE World Health Organization shows that depressive disorders are much more frequent among women than men (Ustun & Kessler, 2002). This holds true for Latinos, the largest ethnic group after Anglos in the United States (U.S.; U.S. Census Bureau, 2003). Among the estimated 38 million Latinos, most (24 million) are of Mexican descent (U.S. Census Bureau, 2003) and among Latinos of Mexican descent, the lifetime rates for major depressive episode are two times higher for women compared with men (Vega et al., 1998). Also, women of Mexican descent experience an earlier onset of depressive symptoms than men ( Sorenson, Rutter, Aneshensel, 1991; Vega, Warheit, & Meinhardt, 1984). Poverty increases women’s vulnerability to depression ( Kessler et al 2001; Miranda & Green 1999 and Miranda et al 1997) and more Latinas live in poverty in the U.S. compared with whites or Asians (U.S. Census, 2001). Furthermore, Latinos living in the U.S. are more likely than any other racial group to be without health insurance (U. S. Census Bureau, 2001. Annual Demographic Survey March CPS Supplement [On-line], Available: http://www.bls.census.gov/cps/ads/2001/susenote.htm .U.S. Census, 2001).

Compounding this is the fact that symptomatic Latinos are less likely than non-Latino whites to report previous appropriate mental health care for the treatment of any mental illness (Alegria et al 2002; Hough et al 2002; Wells et al 2001 and Padgett et al 1994) including depression (Miranda et al., 2003). Research also shows that up to 50% of Latinos (Sue, 1977) who receive mental health care discontinue treatment after their first visit ( La Roche 2002 and Lopez 1997). Cultural insensitivity of mental health professionals is a barrier, affecting patients’ trust of their therapists and acceptance of their treatment plan La Roche 2002; Lopez 1997 and Sue 1998). Lack of self-recognition of the need for mental health treatment (Kessler et al., 2001) and social stigma ( Rubenstein et al 1999 and Rubenstein et al 1999) are also barriers to appropriate mental health treatment for depression among Latinos. All of these factors may influence perceptions about the cause of symptoms ( La Roche 2002 and Lopez 1997). This then can influence whether, and how, a person manages their symptoms of depression.

The Symptom Management Model (SMM; UCSF School of Nursing Symptom Management Faculty Group 1994 and Dodd et al 2001) offers a useful approach for the study of the perceptions of women of Mexican descent related to the causes of their depressive symptoms. The model points out that symptoms are subjective experiences that reflect changes in a person’s bio-psycho-social function, sensation, or cognition. Once an individual appraises their symptoms they are able to make meaningful judgments about the cause, treatability, and potential impact of their symptoms on their life. These perceptions and judgments form the basis of the individual’s symptom management strategies. To effectively manage the symptoms, however, the SMM holds that personal, environmental, and health/illness factors should be taken into consideration (Dodd et al., 2001). Because no published research has explored the perceptions of Latina women related to the source of their depressive symptoms, little is known about their sense of the cause of their distress. Such data could aid in the design of treatment programs, making them more attractive to low income, Latina women who are struggling with depression, but reluctant or unable to access care. Therefore, this article is focused on the perceived reasons behind episodes of sadness, hopelessness, or depression experienced by low income, Latina women of Mexican descent who are at risk for depression.

Guided by the SMM (Dodd et al., 2001) this secondary analysis was done with a subsample of 107 women from a larger cross-sectional study which included survey data from 315 women of Mexican descent (see Heilemann, Lee, & Kury, 2002). Women who were at risk for depression as identified by the Center for Epidemiologic Studies Depression (CES-D) scale and who also reported a qualitative, subjective reason for their most recent episode of sadness, hopelessness, or depression were included in this analysis. Our first aim was to describe the women in terms of person variables (including income, education, partner status), environment variables (including language preference and place of birth as parameter estimates of acculturation), and health/illness variables (including risk factors such as trauma history and alcohol, cigarette, or substance use). Our second aim was to analyze and describe reasons given by the women for their most recent episode of sadness, hopelessness, or depression.

Method

Sample

Survey methods were used to collect cross-sectional data from a convenience sample of women from three family community clinics and one closely associated dual immersion school (Spanish/English) in a Northern California community. Women of Mexican descent between 21 and 40 years of age who could read and write in Spanish or English were eligible to participate. Because it was likely that those who were not citizens would be reluctant to participate if a signature was required, written consent was waived by the Committee on Human Research and verbal consent was obtained. Each participant received $10.00 in cash and a small gift after completion of the questionnaire. Of approximately 380 women approached, 55 declined participation and 325 completed the study. Ten women were later excluded because they indicated on the questionnaire that they were not of Mexican descent, which resulted in a total sample of 315. For this secondary analysis, we include only the data from 107 women with CES-D scores of 16 or higher who also gave short answer qualitative answers to the question about the reason for their experience of sadness, hopelessness, or depression within the last month, including data from women who stated that they are “always” depressed. In a related question about worry, some but not all of the 107 women, extended their discussion of the reason for their depression. This data was also included in the analysis.