Low-income and ethnic minority women have been described as at risk for experiencing depression during and around the time of pregnancy, a finding complicated by low levels of mental health service use within this population. This study retrospectively examined data from a community-based perinatal depression project targeting low-income women in which many barriers to care were removed and a range of services could be elected from social work, specialty mental health, primary care, and peer support. The study focused on 206 women who self-referred to the project after community-based screening. In this sample, the mean age was 25.49 years, with 53% of participants identified as women of color, and 76% had income at or below 185% of the poverty level. The characteristics within this sample most strongly associated with service use varied among sectors of care. Women of color and women with elevated psychosocial risk were significantly more likely to use social work home visiting, whereas current depressive symptom level predicted specialty mental health sector treatment but not other sectors of care. Findings from this study compel future research to consider the complex factors influencing women's use of mental health services among multiple sectors of care.
KEY WORDS: community-based research; depression; mental health services; service utilization; women
An estimated 15% or more of all women who become pregnant experience depression (Edwards et al., 2006; Evans, Heron, Francomb, Oke, & Golding, 2001). Reviews of both medical and social-behavioral research in this area have placed a needed emphasis on emerging clinical trends in identification and treatment (that is, American College of Obstetricians and Gynecologists, 2008; Bledsoe & Grote, 2006). However, elevated risk for mental health challenges for pregnant women may also collide with barriers to full use of mental health services and, thus, jeopardize the receipt of accessible, affordable, and culturally congruent care (Alegria et al., 2002; Anderson et al., 2006; Song, Sands, & Wong, 2004).
Currently, there is a gap in knowledge about where and why low-income pregnant and newly parenting women with depression choose to seek mental health services when services are made available to them across sectors of care. The present study begins to address this gap by examining demographics, current symptoms, and psychosocial risk factors and their influence on service use patterns within one specific community program serving low-income, rural women during and around the time of pregnancy. This study supports Abrams and Curran's (2007) assertion that social work's research agenda for perinatal and maternal depression must be positioned to highlight the experiences of diverse groups of women within a biopsychosocial context.
DOMAINS OF RISK FOR DEPRESSION
A substantive body of literature illustrates demographic risk factors for perinatal and maternal depression. Elevated rates of depressive symptoms, for example, have been detected early in the postpartum period for African American and Latina women (Howell, Mora, Horowitz, & Leventhal, 2005), as have both-prevalence and continuity of elevated symptoms for African American mothers through 18 months postpartum (Beeghly et al., 2003). In addition, low socioeconomic status reportedly increases the risk for ethnic minority women (Beeghly et al., 2003; Miranda et al., 2003; Rosen, Tolman, & Warner, 2004).
Life stressors and psychosocial challenges also contribute to increased risk for depression. In a recent meta-analysis, depression or anxiety during pregnancy, prior mental health history, stressful life events, and lack of adequate social support were identified as moderate to strong predictors of elevated risk for postpartum depression (Robertson, Grace, Wallington, & Stewart, 2004). Several specific forms of stress have been identified as potentially affecting the emotional well-being of women during and around pregnancy, including perinatal loss (Price, 2008), family violence and abuse (Martin et al., 2006), and substance abuse (Rosen et al., 2004).
DEPRESSION AND SERVICE UTILIZATION
Other avenues of research have examined patterns of service use that may be influenced by demography, symptom severity, and compounding psychosocial factors. Low-income women with two or more co-occurring mental health conditions are reported to be more likely to seek treatment than those with depression alone (Rosen et al., 2004). Rosen, Warner, and Tolman (2006) also found that less than one-fourth of either a low-income or a general population sample of women with a mental health disorder had received treatment during the past year. Compounding low rates of use are racial and ethnic disparities in access, quality of care, and overall levels of unmet need. For example, Alegria et al. (2002) reported that African Americans and Latinos have significantly lower overall rates of specialty sector service use, even when symptom severity, demographic characteristics, and insurance status are adjusted for. In a study specifically of low-income pregnant and postpartum women, African American women were less likely to use mental health services than were white women (Song et al., 2004).
Both tangible and intangible barriers to care also affect service use. Women experiencing tangible barriers to care--in such areas as transportation, child care, and adequate insurance reimbursement--are significantly less likely to receive treatment for their mental health concerns (Rosen et al., 2004). However, in the same study, over 25% of respondents who wanted treatment reported that fear was also a significant barrier to receiving care (Rosen et al., 2004). Qualitative studies have further illustrated intangible barriers to care for low-income and ethnic minority women, including lack of trust, judgment and stigma from others, system dissatisfaction, not wanting to receive help, perceived inability of psychiatric services to alter life situations, and fear of being perceived as a less than adequate parent (Anderson et al., 2006; Jesse, Dolbier, & Blanchard, 2008).
Using Regier's framework (Regier et al., 1993), mental health services in the United States are described across four de facto sectors of care: specialty mental health, general medical/primary care, human service professions, and voluntary self-help/peer support. Depression severity has been associated with help seeking in the specialty care sector, but not in overall reported service use (Mojtabai, Olfson, & Mechanic, 2002).Women may also choose to use services in general medical settings (including ob-gyn providers) where they are already receiving care from trusted, accessible providers (Brown, Abe-Kim, & Barrio, 2003); several studies have demonstrated the potential efficacy of both direct intervention and colocated engagement for treatment within this sector of care (Grote, Zuckoff, Swartz, Bledsoe, & Geibel, 2007; Miranda et al., 2003). In a similar manner, the peer support sector of care may provide a viable service delivery option, particularly for groups who may feel marginalized or distrustful of specialty mental health services (Segal, Hardiman, & Hodges, 2002).
SOCIAL WORK'S ROLE
Social work practitioners can help people to navigate the complexity of decision making and barriers to access and offer linkages with accessible, affordable, quality services through case management and engagement with treatment (Grote et al., 2007). In the present study, social workers specifically provided up to three supportive home visits focused on overcoming tangible and intangible access barriers to care, linking women with community services and offering psychosocial support to women during "wait list" periods before community services could be initiated.
COMMUNITY CONTEXT OF THE STUDY
This study was housed within a Healthy Start perinatal depression project funded through the Health Resources and Services Administration, the Maternal and Child Health Bureau, and the Missouri Foundation for Health. The project was located in a rural, five-county area of the Midwest that encompassed a small part of a sixth county. The region is populated by an estimated 135,000 residents, and the target population for this project, women ages 16 to 44, constituted approximately 21% of the overall population. In 2007, an estimated 20% of all individuals and 28% of children under the age of 17 lived...