|Home | About | Journals | Submit | Contact Us | Français|
To examine perceived need for care for mental health problems as a possible contributor to ethnic disparities in receiving care among low-income depressed women.
The role of ethnicity, somatization, and stigma as they relate to perceived need for care is examined. Participants were 1577 low-income women who met criteria for depression.
Compared with U.S.-born depressed white women, most depressed ethnic minority women were less likely to perceive a need for mental health care (black immigrants: OR 0.30, p<0.001; U.S.-born blacks: OR 0.43, p<0.001; immigrant Latinas: OR 0.52, p<0.01). Stigma-related concerns decreased the likelihood of perceiving a need for mental health care (OR 0.80, p<0.05). Having multiple somatic symptoms (OR 1.57, p<0.001) increased the likelihood of endorsing perceived need.
Findings suggest that there are ethnic differences in perceived need for mental healthcare that may partially account for the low rates of care for depression among low-income and minority women. The relations among stigma, somatization, and perceived need were strikingly similar across ethnic groups.
Poor young women are at high risk for depression,1 a disorder that causes significant decrements in functioning when left untreated.2 Minority women are overrepresented among the poor3 and are particularly unlikely to receive mental healthcare in nationally representative studies.4–7 Because untreated depression in young women also results in poor outcomes for their children8 and because treatment of maternal depression can also improve child outcomes,9 efforts to understand the underutilization of mental health services by women are important. The extent to which depressed young women perceive that they need help with a mental health issue may be an important factor in their actually receiving needed care.
Perceived need may facilitate depressed low-income women's efforts to seek mental healthcare. Perceived need is integrated into behavioral models predicting access to health services.10 As would be expected, people with psychiatric disorders are more likely to perceive a need for mental healthcare than are those without such disorders.11,12 Although many individuals who perceive need for help with emotional issues may seek help through family, friends, clergy, and other informal sources, findings from studies in different countries (United States, Canada, Australia, and Puerto Rico) indicate that perceived need for help with an emotional problem is one of the strongest predictors of actually getting mental healthcare from health and mental health professionals.12–14
With few exceptions,12,15 very little attention has been paid to ethnic and immigrant group differences in self-reported perceived need, particularly among depressed, low-income ethnic minority and immigrant women. One small study conducted with low-income women suggests that white women are more likely than minority women to recognize themselves as depressed, but the study failed to find significant group differences.15 In addition, this study did not specifically examine immigrants and nonimmigrants separately.
Stigma and somatization may facilitate or impede the likelihood of perceiving a need for help with a mental health problem. Although study results are mixed with regard to stigma, some evidence indicates that minority groups do exhibit higher levels of stigma about seeking help for emotional problems than do U.S.-born whites.16,17 Studies of Caribbean immigrants to the U.K. and Africans in Nigeria are suggestive of similar patterns, with women indicating concerns about being perceived as “crazy” or holding negative attitudes toward those with mental illness.18–21 Although the link between stigma and accessing care is not strong,22,23 a separate analysis of the larger dataset used in this paper showed that stigma has been associated with decreased interest in mental health treatment among low-income women,24 and we might expect that stigma concerns will influence acknowledgment that one even has an emotional or mental health problem.
Individuals who report multiple somatic symptoms when depressed may be less likely to seek mental health treatment than those who do not report such symptoms. Prior research examining treatment use among people with psychiatric disorders demonstrates that individuals with both mental health difficulties and significant somatic symptoms are more likely to seek medical treatments than psychiatric treatments25 and may, therefore, be less likely to perceive a need for help with mental health issues than do those without such symptoms. Somatization may also impact perceived need differently across ethnic groups, considering that minority groups are thought to somatize more than do U.S.-born whites.26 For example, African women have been found to report somatic symptoms that may actually eclipse primary mood symptoms.27,28 Other research suggests, however, that somatization may be ubiquitous across cultures26,29 and, therefore, does not influence perceived need differently across ethnic groups.
Using a large and diverse sample of depressed women, the present study builds on the literature by examining if the proportion of depressed women who say they perceived a need for help with an emotional problem varies between U.S.-born white women and U.S.-born and immigrant black and Latina women. Based on trends found in previous research15 and the fact that ethnic minorities are less likely to receive care than are white Americans,4 we expect that ethnic minorities will be less likely than whites to report that they need help with an emotional problem. We also examine the association among demographics, stigma, and somatization with perceived need for mental healthcare in this sample of depressed women. Specifically, we expect that women with stigma will be less likely than women without stigma to perceive a need for mental healthcare. Similarly, we predict that women who somatize will be less likely than those who do not to perceive a need for help. Finally, based on research suggesting that stigma and somatization may be more common among some of the ethnic minority groups, we explore whether the relations between perceived need and stigma and perceived need and somatization vary by ethnic group by looking at interaction effects.
Participants in this study of perceived need for mental health care were 1,577 depressed young women. These women are a subset of 15,383 low-income women who were screened in the Women Entering Care (WE Care) depression treatment study.30,31 We chose to examine the depressed subsample because they are in need of care. Further details on the screening methods are described in a study of the full sample by Nadeem et al.24 Participating women were screened primarily in the Women Infants and Children (WIC) program, a program that provides nutritional services to low-income pregnant and postpartum women and their children (up to age 5). Women were also screened in county-run Title X family planning clinics, which provide comprehensive family planning services for low-income women. A small number were screened in pediatric clinics and in a subsidized housing project. Of all women approached for screening, only 8.8% declined participation, indicating that the vast majority of women served in these settings were screened. All participants provided written informed consent in either English or Spanish. Immigrant black women served in these settings were English speaking. Immigrant Latina women were primarily Spanish speaking. Seventy-five percent of the immigrant Latina women in the present study were from Central America (e.g., El Salvador, Guatemala, Nicaragua), 19% were from South America (e.g., Chile, Bolivia), and the rest were from the Caribbean or Mexico. Specific country of origin information was not available for immigrant black women. Research procedures were approved by the institutional review boards of Georgetown University, the State of Maryland, and the University of California, Los Angeles.
All measures were administered in the participants' preferred language (English or Spanish) in a personal interview format by WE Care project staff at the different service settings.
The full version of the Primary Care Evaluation of Mental Disorders (PRIME-MD)32 was used to identify women who were depressed. The PRIME-MD's questions are based on DSM-IV criteria, and the measure demonstrates good agreement with independent psychiatric diagnoses made using a structured interview (92%). The PRIME-MD has also been used successfully with ethnic minority populations.32
Perceived need for care for a mental health problem was assessed by asking women to answer yes or no to an item used by Yokopenic et al.33 in a study of depression and help-seeking; In the past month, have you had severe enough personal, emotional, behavior, or mental problems that you needed help? and who also responded yes to one or more descriptions of these problems: depression, feeling low, anxiety, panic, nerves. Women with a perceived need were coded with a 1, and those without a perceived need were coded 0. Although this item has not been validated, it is very similar to items used in the National Comorbidity Study (NCS) and other psychiatric epidemiological studies.12,34 In the NCS, 32% of people who had disorders perceived a need for help, and perceived need mediated differences in service use.12
A stigma variable was created from three yes/no items in response to the query: Would any of the following keep you from getting professional help? Participants were categorized as having stigma-related concerns if they endorsed one or more of the following: being embarrassed to talk about personal matters with others, being afraid of what others might think, and family members might not approve. If they did not endorse any of these items, they were categorized as not having stigma concerns. Items for stigma-related concerns were taken from the work of Sussman et al.23 and were selected based on their relevance to issues around seeking care.
Participants were classified as somatizers if they reported six or more DSM-IV symptoms of somatization disorder, based on the abridged somatization construct developed by Escobar et al.25 The measure was tested in a mixed ethnicity population, was shown to discriminate well between those with somatization and those with other problems, and had good construct validity.25 Physical health status was not available for the current study; however, our sample was young and would not be expected to have significant physical health problems.
Participants provided demographic information, including age, marital status, employment status, insurance status, number of children, education, and years in the United States. Information about ethnicity and country of origin was obtained via the following two items: What is your cultural or ethnic identity? Where were you born?
Simple descriptive statistics were used to characterize the samples. In order to examine the relations among sociodemographic characteristics, perceived need, stigma, and somatization, we conducted a logistic regression using stigma and somatization and the interactions between ethnicity and each of these variables to predict perceived need, controlling for sociodemographic variables. Interpretation of odds ratios (ORs) in these models assumes all other variables in the model are held constant.
Sample characteristics are presented in Table 1. The sample consists of 26 (1.6%) U.S.-born Latina, 598 (37.9%) immigrant Latina, 802 (50.9%) U.S.-born black, 49 (3.1%) immigrant black (African and Caribbean), and 102 (6.5%) U.S.-born white women. The women are relatively young (M 28.9, SD 8.5). Almost half were married or living with a partner, and about 35% did not complete high school. About 8% of the women graduated from college, 57% of women are uninsured, and 21% have medical assistance. On average, the immigrants had been in the United States just over 8 years. Overall, 55% (n=863) of our depressed sample perceived a need for help with an emotional problem, suggesting that failure to perceive need for care may be an important factor in underutilization of mental healthcare by poor women. For comparison, we note that of the 13,769 women from the larger screening sample who did not have depression, only 8.64% perceived a need for mental healthcare. In addition, in this depressed sample, 34% (n=537) had stigma-related barriers to care, and 67% (n=1053) reported significant somatic complaints.
The ethnic groups differed across a variety of demographic factors for which we control in subsequent analyses. Of note is the lack of ethnic group differences in somatization. In addition, African immigrants were more likely to endorse stigma-related barriers to care than were U.S.-born whites; however, differences in stigma between whites and other minority groups were not as pronounced. Very few of these depressed women (8%, n=129) were in treatment, and whites were more likely to be in treatment than each of the ethnic minority groups.
Table 2 depicts the logistic regression predicting the endorsement of perceived need by ethnicity, stigma, and somatization. We also controlled for demographics because many of these variables differed across ethnic groups. There was an overall effect of ethnicity (p<0.01). With the exception of U.S.-born Latinas, all the minority groups were less likely to perceive a need for mental healthcare compared with U.S.-born white women (black immigrants: OR 0.30, p<0.001; U.S.-born blacks: OR 0.43, p<0.001; immigrant Latinas: OR 0.52, p<0.01). The trend for the U.S.-born Latinas, our smallest group, was in the same direction. In addition, those with stigma-related concerns were less likely to perceive a need for mental healthcare (OR 0.80, p<0.05) than those without stigma-related concerns. Being a somatizer (OR 1.57, p<0.001) increased the likelihood of endorsing perceived need compared with those with few somatic complaints.
The interactions tested were not significantly related to perceived need and are not included in Table 2 (p=0.54 for stigma by ethnicity, and p=0.97 for somatization by ethnicity). Moreover, none of the individual contrasts between the ethnic minority groups and the whites were significant, suggesting that stigma and somatization are related to perceived need in a similar manner across ethnic groups.
The current study sought to explore ethnic differences in perceived need for treatment among a sample of depressed low-income U.S.-born and immigrant women from various ethnic backgrounds. Many of these women in the study did not perceive a need for help. Consistent with the work of others,15 about 56% of our sample of depressed women perceived a need for help with an emotional problem, leaving a sizable proportion (44%) of depressed women who did not report needing help. The finding that nearly half of these depressed women did not identify themselves as needing mental healthcare is consistent with our clinical work with this population. Many women are aware that they feel sad, but they frequently attribute these feelings to external forces. For example, women report that they are unhappy with their lives (e.g., work, significant relationships) and perceive these issues as the problem, not their current mood. We have found that these women are open to learning that their sadness could be helped.
When we examined correlates of perceived need for care, we found significant differences between the different ethnic and immigrant groups. In general, the ethnic minority groups were less likely to perceive a need for help with an emotional problem than the U.S.-born white women, even after controlling for sociodemographic variables. Given the strong link that has been drawn between perceived need for help and actually accessing mental health treatment,12 this finding has implications for how we understand treatment-seeking processes in ethnic minority groups. The most pronounced group difference was that immigrant blacks were far less likely to perceive a need for help than U.S.-born whites. U.S.-born Latinas were most similar to the U.S.-born whites. Although there was not a significant difference between the U.S. born Latinas and the U.S.-born whites, it is important to note that the sample size for the U.S. born Latinas was relatively small, making it important to use caution in interpreting the findings for this group. Nonetheless, the trend was in the same direction as that of the U.S.-born whites. These findings suggest that perceived need does vary across different ethnic groups. Programs aimed at educating these women about mental healthcare and depression that are conducted in the public service settings they frequent could potentially help to overcome disparities in care for depression among low-income and minority women.
Overall, women who reported stigma concerns about seeking treatment were less likely to say they perceived a need for care. This finding has implications for the way we conceptualize the role of stigma in theoretical models of access to care. Stigma appears to mask identification of need for care. Efforts to educate women about need for care may also need to address stigma, helping women to feel more comfortable identifying their troubles as depression.
Somatization increased the odds that depressed women in our study perceived a need for help. This finding is in contrast to the notion that those who somatize are less likely to assume their problems are emotional. In contrast, the experience of multiple somatic complaints may signify greater general distress and serves as another indication that they need help with their problems. This pattern is similar to that found in research showing that those with psychiatric comorbidities are more likely to perceive a need for mental healthcare.11
Although stigma and somatization were significantly associated with perceived need, they do not appear to function differently for the various ethnic groups. In fact, we found no significant interactions and there were more ethnic group similarities than there were differences in the overall levels of stigma and somatization. This was most striking for somatization, a finding that supports the notion that somatization is more ubiquitous across cultures than it is often thought to be.29 With regard to stigma, we found that black immigrants from Africa and the Caribbean were more likely than whites to report stigma-related concerns. This is consistent with findings suggesting that stigma and negative attitudes toward the mentally ill are more severe in African than in Western cultures18 and with research suggestive of heightened stigma among Caribbean immigrants to the U.K. and Canada.19,35 At the same time, there were no stark contrasts among depressed U.S.-born whites, U.S.-born blacks, U.S.-born Latinas, and immigrant Latinas in their endorsement of stigma-related concerns about seeking care.
What then does account for ethnic differences in perceived need? Although we are not able to pinpoint factors that account for ethnic differences in perceived need for mental healthcare, cultural beliefs and contextual factors could certainly account for this difference. The tendency to perceive a problem as something that could be helped by mental healthcare is likely shaped by cultural beliefs and exposure to mental healthcare. Similarly, minority women may have less exposure to and knowledge about care because fewer of their friends, family members, and work associates may have gotten and benefited from care. These cultural and exposure differences may influence minority women's perceptions of their own need for care.
Several limitations to the present study should be taken into consideration when interpreting the findings. The study is cross-sectional, so we cannot specify the direction of the effects reported here. Second, because so few of the women in our study were in treatment, particularly among the minorities, it was not possible to examine perceived need in relation to actual treatment use. This is an important topic for future research and will be challenging considering the low rates at which impoverished, ethnic minority women are in treatment. Third, there are some limitations to our measures. Our measure of somatization is effective at identifying somatic symptomatology; however, we do not have additional physical health measures on our sample. Related, the research on stigma is not well developed at this time, and efforts should be made to construct well-validated measures of different aspects of stigma. This study's measure of stigma assesses perceived stigma from others and does not fully address all of ways in which stigma could influence the reporting of emotional concerns.
Finally, there are important considerations in interpreting our data related to the sample. Our sample was primarily impoverished. The minority women may be relatively representative of immigrant minorities in general because of the high rates of poverty in those populations. On the other hand, fewer white women are impoverished, so the poor women in our sample are likely less representative of white women in general. Our ethnic comparisons should be interpreted only as differences between poor women. Our study did not specify relations between variables based on countries of origin in part because of limited sample size, and it is important to reiterate that not all Latino cultures or all African or Caribbean cultures are the same. However, research conducted with various Latino cultural groups, for example, reveals similarities in key values, such as reliance on family,36 that influence help-seeking processes. Additionally, our study used immigrant status as the primary method to distinguish between those who had greater or less exposure to mainstream U.S. culture and services systems. We recognize the importance of viewing acculturation and immigration as multifaceted processes and advocate that future research should focus on these issues and specific cultural constructs that are relevant to the help-seeking process. Nonetheless, the present study is the first of its kind to include a sample rich in ethnic diversity and to explore differences among immigrant and nonimmigrant women of Latina, African, and Caribbean backgrounds.
In summary, our findings suggest that failure to perceive a need for care may partially account for the low rates of care for depression among low-income and minority women. In addition, there are ethnic differences in overall levels of perceived need for help with emotional problems across depressed women of different ethnic groups, with U.S.-born white women being more likely than immigrant black and Latina women and US.-born black women to perceive the need for help. This finding has important implications, given the strong link between perceived need and receiving care in the literature, for educating women about depression and depression care. Study results also suggest that stigma may reduce the likelihood that low-income depressed women, regardless of ethnicity or immigrant status, are willing to state that they have a problem for which they need help. Women who somatize, on the other hand, are more likely than those who do not to perceive a need for care. Interestingly, the relations among stigma, somatization, and perceived need were strikingly similar across ethnic groups.
This research was funded by National Institute of Mental Health grants MHR01070260 and MH56864. Writing of this paper was funded through three centers: Resource Centers for Minority Aging Research/Center for Health Improvement of Minority Elderly (RCMAR/CHIME) funded by National Institutes of Health/National Institute on Aging (3P03AG021684), UCLA/Drew Project EXPORT funded by the National Institutes of Health/National Center for Minority Health and Health Disparities (1P20MD00148-01), and UCLA-RAND Center for Research on Quality in Managed Care (MH068639-01) and the John D. and Catherine T. MacArthur Foundation.
The authors have no conflicts of interest to report.