Severe Intimate Partner Violence,Sources of Stress and the Mental Health of U.S. Black Women

Abstract

Background:

We investigate the mental health risk of U.S. Black women by examining the roles of intimate partner violence (IPV), major discrimination, neighborhood characteristics, and sociodemographic factors using one of the largest and most complete datasets on U.S. Blacks.

Materials and Methods:

The National Survey of American Life (NSAL) used a slightly modified version of the World Health Organization Composite International Diagnostic Interview (WHO-CIDI) with a sample of 6082 participants. We also assess intraracial group differences based on ethnicity and nativity status (U.S.-born African American, U.S.-born Caribbean Black, and foreign-born Caribbean Black).

Results:

The study provides evidence that severe physical intimate partner violence (SPIPV) is a significant threat to the mental health of U.S. Black women. Bivariate and multivariate analyses indicate that those with a history of SPIPV were at greater risk for mental disorders than women who did not experience violence by a spouse or partner. Racial discrimination was associated with higher odds of anxiety and substance disorders, whereas gender discrimination was associated with higher odds of mood disorders. Neighborhood drug problems also increased the odds of mood, anxiety, and substance use disorders. Older age and being an Afro-Caribbean immigrant were associated with lower odds of three of four mental disorders.

Conclusions:

Findings from the study indicate the need for community and clinical interventions aimed at addressing IPV and other community factors that influence Black women's mental health.

Introduction

Physical assault by an intimate partner is a threat to health and can have profound effects on the quality of life for women.^1

–6 Injury outcomes range from the extreme of death, hospitalization, and injuries that require surgical repair to broken bones, eye injuries, dental damage, facial injuries including black eyes and generalized bruises on various parts of the women's bodies.⁷ Intimate partners cause injuries to women by various means including acts such as hitting her with objects, the use of knives or firearms as weapons, as well as hitting, biting, shoving, and kicking their bodies.⁸ Common sequelae of intimate partner violence (IPV) include substance use, depression, anxiety, posttraumatic stress disorder (PTSD), and suicide attempts and ideation.^1,2,4, ^5,9,10

U.S. Black women may especially be prone to these outcomes because of higher rates of violence compared with other racial and ethnic groups (e.g., White and Hispanic). An estimated 41% of Black women have been severely physically assaulted by a spouse or partner over their lifetime.^11,12 This is concerning because despite social, economic, and educational gains, Black women continue to have poorer physical and mental health outcomes, higher rates of illness, and lower life expectancy than nonminority groups. Despite growing public concerns surrounding the possible exacerbation of IPV on the general welfare of women, research on the mental health of Black women survivors of IPV is limited,⁹ an important omission in past research given the unique sources of stress that Black women face in the United States (e.g., racial and gender discrimination and living in vulnerable neighborhoods).

Stressful living conditions such as racism and poverty are not only important contributors to poor health but can also serve as an important precursor to IPV.^9,12
–14 Low-income and ethnic minority women, particularly African Americans, are prone to these conditions because of lower social standing in society and high exposure to discriminatory practices.^15
–17 The combination of extensive victimization and poor social and environmental conditions places Black women at greater risk for various negative mental health outcomes. However, methodological challenges (including small, nonprobability samples) have limited our ability to fully understand the possible effects of IPV in conjunction with other sources of stress on Black women's well-being. This study uses a large nationally representative sample of Blacks to examine the influence of severe physical intimate partner violence (SPIPV) and other sources of stress on the mental health of Black women residing in the United States.

Background

Although most studies focus on the general population, a small but growing body of research has specifically focused on understanding the relationship between IPV and the mental health of U.S. Black women. Among African American women, physical abuse by an intimate partner is associated with mood disorders.^18

–22 Ramos et al.²³ found that lifetime abuse was associated with elevated levels of depression and anxiety among African American women in a primary health care setting. IPV has also been linked to substance abuse^9,24 and suicidality²¹ in Black women. Finally, both cross-sectional^25,26 and prospective studies²⁴ have found an association between IPV and PTSD among African American women. It is important to note that these associations are not only limited to studies that focus on the overall Black racial category or African American women solely; IPV is also associated with poor mental health among Afro-Caribbean women, both in the United States and across the African diaspora.^18,27,28 With some exceptions, very few studies have examined interethnic differences in the association between IPV and mental disorders between African American and Afro-Caribbean women.^7,18,27,28

Nativity status is also imperative to consider when assessing health trends; a host of studies across the globe have documented a trend known as the “healthy immigrant effect.” This term refers to consistent epidemiological evidence that immigrants currently residing in the United States exhibit better health outcomes than their same race/ethnic counterparts who were born in the United States (a phenomenon sometimes called the “immigrant health paradox” given many hail from developing countries and face acculturative stress once in the United States). A common explanation for this phenomenon is that it is because of selection bias whereby those immigrants in the best health (and with more economic resources) are most likely to emigrate, leaving their less healthy and lower socio-economic status (SES) counterparts back in their home country.²⁹

Several comprehensive literature reviews and meta-analyses of both cross-sectional and longitudinal studies consistently find strong linkages between discrimination and mental health, in both regional and national samples.^30

–34 Surprisingly, there is relatively little research using nationally representative samples to focus specifically on discrimination and the mental health of Black women.^35,36 However, important regional studies confirm that unfair treatment is a central determinant of poor mental health outcomes for Black women.^37,38

Research on discrimination attribution is very limited. In terms of mental health, one important study using the National Survey of American Life (NASL) found that Black Americans who attributed unfair treatment to race had higher odds of serious psychological distress than those who attributed it to nonracial causes,³⁹ a finding that has been echoed in research using the same data to examine discrimination attributions and physical health outcomes among older African Americans.⁴⁰ These findings are consistent with sociological research related to the “identity-relevant stress hypothesis,” which posits that stressors related to the aspects of identity that are most central to one's sense of self have more damaging health effects than stressors unrelated to those important aspects of the self.⁴¹ Given the salience of race and gender in Black women's lives, it is likely that racial and gender discrimination—rather than unfair treatment in general—may be especially damaging to Black women's health.⁴² To the best of our knowledge, there is no available research on discrimination attribution among adult Black women.

Although individual socioeconomic status in the form of human capital (e.g., educational attainment, income, and poverty) is a strong determinant of mental health,⁴⁰ community-level influences cannot be ignored because a high percentage of Black women in the United States live in poverty and under-resourced neighborhoods.⁴³ In addition to inadequate resources, residing in these communities also exposes Black women to higher rates of violence and resulting mental health outcomes. One unique and rigorous study in this regard used data from 60 US communities and found that adults who live in high-crime neighborhoods had higher odds of anxiety and major depression than those who did not.⁴⁴

The literature on IPV and sociocontextual factors is relatively distinct. This study seeks to bridge this gap by understanding the relative contributions of severe IPV, social stressors, and sociodemographic characteristics on the mental well-being of U.S. Black women. We draw specifically upon the social ecological theoretical framework of health promotion, one that recognizes the interplay on individual factors and health-promotive or health-deteriorating aspects of both the physical and social environments.⁴⁵ By examining factors beyond the individual, we take a broad socioenvironmental approach to assessing risk factors for mental disorders among Black women currently living in the United States. Our approach was guided by the following research questions:

How is ethnicity/nativity associated with the odds of mental disorders among Black women?;

Are individual-level socioeconomic status measures more predictive of mental disorders among Black women than neighborhood SES measures?;

How are various forms of major discrimination (gender discrimination and racial discrimination) associated with mental health among Black women?; and

What role does IPV play in the mental health of U.S. Black women after controlling for social ecological variables such as discrimination and neighborhood characteristics?

Materials and Methods

Data collections

Data used in this study were from one part of the Collaborative Psychiatric Epidemiological Study (CPES), specifically, the NSAL. Collected between February 2001 and March 2003, the NSAL is the most comprehensive and detailed study of mental disorders and physical health on African American adults over the age of 18 residing in the 48 coterminous states. It also provides information on three ethnic groups: African Americans, Caribbean Blacks, and non-Hispanic Whites residing in areas with at least 10% of the Black population.⁴⁶ Data were collected using a multistage sampling design based on the distribution of the Black population. Face-to-face interview was the main method of data collection with an additional 14% of interviews collected entirely or partly by way of telephone. Institutionalized persons (e.g., prisons, jails, and nursing homes) were not included in the study, nor were individuals who were unable to communicate their survey response in English. In-person interviews were conducted with a computer-assisted device. A total of 6082 participants completed the interview representing 72.3% response rate: 3570 African American; 1621 Caribbean Black; and 891 non-Hispanic White. African Americans comprised individuals who were of African ancestry but did not have Caribbean roots. Caribbean Blacks comprised individuals with African ancestry but had Caribbean heritage.⁴⁶ These participants had to meet several criteria: (1) they were to be of West Indian or Caribbean descent, (2) they were to be from the list of Caribbean countries presented by interviewers, (3) or that their parents or grandparents had to be born in the Caribbean.⁴⁷ Women of African descent were the focus of this study.

Measures

Severe physical intimate partner violence

SPIPV was assessed with a single question that queried participants on whether they had ever “been badly beaten up by a spouse or romantic partner.” We compared the NSAL IPV measure with the National Comorbidity Study Replication (NCS-R) dichotomously defined severe partner violence Conflict Tactic Scale (CTS) within the CPES.^48,49 To address the proposed measure's validity, two tests—probability of agreement (odds ratio [OR] = 4.5, confidence interval [CI] = 1.49–1498; p < 0.001) and area under the curve (≥0.6)^50,51—were conducted and it was found that they have a fair association across different approaches to estimating agreement.³⁶

Mental health

The NSAL used a slightly modified version (see Kessler and Merikangas⁵²; Jackson et al.,⁴⁶ for details) of a structured clinical assessment tool based on the DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, fourth edition) World Health Organization Composite International Diagnostic Interview (WHO CIDI) to address lifetime mental disorders. Mental health conditions included: mood disorders (e.g., major depressive episode, dysthymia, major depressive disorder, bipolar, and any mood disorder); anxiety disorders (e.g., panic, agoraphobia, generalized anxiety disorder (GAD), obsessive–compulsive disorders (OCDs), PTSD, any anxiety; and substance disorders (e.g., alcohol abuse, alcohol dependence, drug abuse, drug dependence, and any substance disorder). Any disorder is a composite of various disorders within specific conditions discussed. Questions surrounding suicide ideation and attempts were also administered to participants. These questions were, “have you ever seriously thought about committing suicide?” and “have you ever attempted suicide?” Response options were yes or no.

Major experiences of discrimination

Nine items were used to gauge lifetime experiences of major discrimination/unfair treatment. Participants were asked to disclose whether or not they ever had any of the following experiences attributed to their race and gender: (1) been unfairly fired; (2) not been hired for a job; (3) unfairly denied a promotion; (4) unfairly stopped, searched, questioned, physically threatened or abused by police; (5) unfairly discouraged by a teacher or advisor from continuing their education; (6) unfairly prevented from moving into a neighborhood because the landlord or a realtor refused to sell or rent you a house or apartment; (7) moved into a neighborhood where neighbors made life difficult for you or your family; (8) unfairly denied a bank loan; and (9) received service from someone such as a plumber or car mechanic that was worse than what other people get. Response options were “yes” or “no.” Discrimination attribution was then measured using a follow-up question for those who responded affirmatively, who were then asked to choose what they deemed to be “the main reason for this experience.” An event that was attributed to one's ancestry/national origins, race, or shade of skin color was counted as an act of racial discrimination, whereas an event attributed to gender was counted as an act of gender discrimination. We then calculated two measures representing a count of the number of acts of racial discrimination and the total number of acts of gender discrimination experienced across the life course.

Neighborhood characteristics

To address neighborhood crime problems, participants were asked, “how often are there problems with muggings, burglaries, assaults, or anything else like that in your neighborhood?” Measured on a Likert scale, response options were as follows: very often, fairly often, not too often, hardly ever, and never. To address the presence of neighborhood drug problems, participants were asked, “how much of a problem is the selling and use of drugs in your neighborhood?” The response options were as follows: very serious, fairly serious, not too serious, and not serious at all.

Controls

The covariates included age (in years), relationship status, educational level, poverty, and ethnicity/nativity. Relationship status was separated by married, partnered, separated or divorced, widowed, and never married. Education level was coded to reflect: less than high school, high school graduate, some college and college educated. The occupational status of participants included employed, unemployed, and not in the labor force. Poverty status is an income-to-poverty ratio consisting of the participant's household income divided by the 2001 US Census poverty threshold for the number of adults and children living in that household. Ratios <1.00 indicate that the income for the participants' household is below the official poverty threshold, whereas a ratio of ≥1.00 indicates income above the poverty level. For example, a ratio of 1.25 indicates that income was 25% above the appropriate poverty threshold.⁵³ Ethnicity/nativity was measured using categories for US-born African Americans, US-born Caribbean Blacks, and foreign-born Caribbean Black.

Analytic strategy

Bivariate analytic procedures were used to address the mental health of women with and without a history of severe physical intimate partner violence. We used stepwise binary logistic regression analysis to address the mental health of U.S. Black women in association with SPIPV, social stressors, and sociodemographic variables in three successive blocks. Block 1 contains sociodemographic variables. In Block 2 are social stressor variables. Block 3, the final block, includes IPV. We specifically focused on a broad class of any disorders (e.g., mood, anxiety, and substance) and suicide ideation. Suicide ideation as opposed to attempts was used for multivariate analysis because of inadequate sample size. Specialized statistical techniques were used to account for the multistage probability sample design of the NSAL. All analyses were conducted using STATA 11, which uses the Taylor expansion approximation technique for calculating the complex designed-based estimates of variance.⁵⁴ Design weights were included in the analysis to adjust for differential sample selection and nonresponse. Standard errors were corrected for weighting, clustering, and stratification. A 0.05 alpha level was set for significance.

Results

Sample description

Descriptive and bivariate statistics of the analytic sample (stratified by IPV history) is given in Table 1. With the exception of age and region, Black women with and without a history of SPIPV were significantly different on all study variables. The average age of participants was 42.5 and 54% resided in the South. In terms of relationship status, more than one-third (36.9%) of women without a history of SPIPV were married or partnered, compared with 30.6% of women with a history of SPIPV (p < 0.001). Women without a history of SPIPV were more likely to have completed a bachelor's degree than those with a history of SPIPV (15.4% vs. 10.8%; p < 0.01). Women with a history of SPIPV were more likely to live at or below the poverty line than those without a SPIPV history (37.8% vs. 26.6%; p < 0.001). Although the majority of women were employed, significantly fewer women with an SPIPV history were employed compared with women who did not report a history of SPIPV (58.8% vs. 65.0%; p = 0.05). Almost half (49.3%) of women without a history of SPIPV owned a home, compared with only roughly one-third (34.5%) of women with a history of SPIPV (p < 0.001).

Table 1.

Univariate and Bivariate Statistics of Sociodemographic Characteristics, Stratified by Severe Physical Intimate Partner Violence History, U.S. Black Women Only, National Survey of American Life (N = 3277)

Characteristics	Total sample, N = 3277 (%)	No severe IPV, n = 2646 (%)	Severe IPV, n = 505 (%)	F-test statistics differences
Age of respondents (mean)	42.5	42.0	42.7	0.46
Marital status
Married	27.4	28.7	20.6
Partnered	8.4	8.2	10.0
Separated or divorced	20.3	17.7	33.6	10.86^***
Widowed	11.5	11.7	8.9
Never married	32.4	33.8	26.9
Education
Less than high school	24.8	23.1	34.1
High school graduate	36.0	36.3	31.1	5.74^**
Some college	24.8	25.2	24.0
Bachelor's degree or more	14.4	15.4	10.8
Poverty status (%)
Above the FPL	71.6	73.4	62.2	14.95^***
At or below the FPL	28.4	26.6	37.8
Occupational status
Employed	63.7	65.0	58.8
Unemployed	11.1	10.2	16.1	6.82^*
Not in labor force	25.2	24.7	25.1
Homeownership
Own	47.1	49.3	34.5	27.76^***
Do not own	52.9	50.8	65.5
Region
Northeast	18.8	18.8	20.7	2.20
Midwest	18.9	16.6	23.3
South	54.0	56.4	46.2
West	8.3	8.2	9.8

p < 0.05; ^** p < 0.01; ^*** p < 0.001.

IPV, intimate partner violence; FPL, federal poverty line.

Bivariate analysis of the prevalence of mental disorders for women with and without a history of SPIPV

Table 2 provides the results from bivariate tests between SPIPV history and mental disorders (broad categories and specific disorders). Except for bipolar disorder, Black women with a history of SPIPV had higher rates of any mood disorders. The rate at which women with a history of severe physical IPV met criteria for any mood disorder was around two times that of women without a history of SPIPV (29.3% vs. 14.1%; p < 0.001) (Table 2). Within specific mood disorders, the percentage of major depressive episode (26.9% vs. 12.1%; p < 0.001), dysthymia (9.5% vs. 3.0%; p < 0.001), and major depressive disorder (26.9% vs. 12.1%; p = 0.000) were more than twice as high for women with a history of SPIPV than those without a history of SPIPV.

Table 2.

Bivariate Analysis of Lifetime Mental Health Conditions Among Black Women With and Without a History of Severe Physical Intimate Partner Violence, National Survey of American Life (N = 3277)

Disorders	No severe IPV (%)	Severe IPV (%)	F-statistics difference
Mood disorders
Any	14.1	29.3	63.71^***
Major depressive episode	12.1	26.9	67.59^***
Dysthymia	3.0	9.5	18.15^**
Major depressive disorder	12.1	26.9	67.59^***
Bipolar	3.1	6.4	7.18
Anxiety disorders
Any	16.3	36.9	110.61^***
Panic disorder	3.0	9.9	28.31^***
Agoraphobia	2.1	7.4	43.76^***
Generalized anxiety disorder	4.7	9.0	11.77^*
Obsessive–compulsive disorder	1.2	3.2	8.11^**
Posttraumatic stress disorder	9.1	25.9	101.18^***
Substance disorder
Any	4.1	15.7	41.23^***
Alcohol abuse	3.1	13.6	60.73^***
Alcohol dependence	1.4	6.7	35.82^***
Drug abuse	2.1	8.2	24.63^***
Drug dependence	0.7	4.2	25.13^***
Suicide
Attempts	3.6	11.1	27.37^***
Ideation	10.2	24.4	56.55^**

p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Similarly, the percentage of all anxiety disorders was higher for women with a history of SPIPV than those without such a history. For example, the rate of any anxiety disorder was more than twice as high for women with a history of SPIPV than those without a history of SPIPV (36.9% vs. 16.3%; p < 0.001). Women with a history of SPIPV compared with women without a history of SPIPV had roughly three times the percentage of panic disorder (9.9% vs. 3.0%; p < 0.001), agoraphobia (7.4% vs. 2.1%; p < 0.001), OCD (3.2% vs. 1.2%; p < 0.01), and PTSD (25.9% vs. 9.1%; p = 0.001). The percentage of GAD was almost twice as high among SPIPV survivors than those without a history of SPIPV (9.0% vs. 4.7%; p = 0.05). It is important to note that, for virtually all mental disorders, the overall prevalence of Black women without a history of severe physical IPV is lower than the overall rate for all women, a paradoxical trend that has been studied extensively by others.

Similar patterns were found for both substance disorder and suicidality. Overall, 15.7% of Black women with a history of SPIPV met criteria for any substance disorder, compared with only 4.1% of Black women without a history of SPIPV (p < 0.001). Regardless of specific disorders, rates of substance use were roughly four times higher among Black women with a history of SPIPV (relative to those without a history of SPIPV). The percentage of suicide attempts were more than three times higher among SPIPV survivors than those without a SPIPV history (11.1% vs. 3.6%; p < 0.001) and SPIPV survivors had more than twice the percentage of suicidal ideation than those without a history of SPIPV (24.4% vs. 10.2%; p < 0.01).

Multivariate results of SPIPV, stressors and sociodemographic on mental disorders

Mood disorder

Table 3 provides the results from a nested binary logistic regression model series predicting the odds of any lifetime mood disorder for U.S. Black women. Model 1 shows that older women (adjusted OR [AOR] = 0.976; CI = 0.963–0.989, p < 0.001) and high school graduates (AOR = 0.716; CI = 0.527–0.973, p < 0.05) had lower odds of any mood disorder (relative to those without a high school diploma); conversely, women who were separated or divorced (AOR = 1.793; CI = 1.261–2.550, p < 0.01) had higher odds of mood disorder than those who were married. No other measures were significant predictors of any mood disorder.

Table 3.

Stepwise Binary Logistic Regression Predicting Any Lifetime Mood Disorder Based on Sociodemographic Characteristics and Sources of Stress, Black Women Only, National Survey of American Life (N = 2940)

Characteristics	Model 1, OR (CI)	Model 2, OR (CI)	Model 3, OR (CI)
Age (continuous)	0.976 (0.963–0.989)^***	0.977 (0.964–0.990)^***	0.976 (0.962–0.989)^***
Ethnicity/nativity
African American	1	1	1
US-born Caribbean Black	1.480 (0.696–3.144)	1.322 (0.604–2.89)	1.333 (0.587–3.027)
Foreign-born Caribbean Black	0.694 (0.459–1.05)	0.825 (0.522–1.303)	0.887 (0.565–1.392)
At/below the FPL (ref = above)	1.302 (0.951–1.783)	1.272 (0.942–1.718)	1.226 (0.888–1.693)
Occupational status
Employed	1	1	1
Unemployed	0.910 (0.587–1.412)	0.864 (0.519–1.359)	0.825 (0.511–1.332)
Not in labor force	0.970 (0.663–1.418)	1.011(0.695–1.472)	1.003 (0.677–1.484)
Education
Less than high school	1	1	1
High school graduate	0.716 (0.527–0.973)^*	0.755 (0.550–1.035)	0.801(0.570–1.124)
Some college	0.989 (0.642–1.523)	0.977 (0.636–0.209)	1.037 (0.661–1.629)
Bachelor's degree or more	1.193 (0.768–1.855)	1.164 (0.728–1.862)	1.290 (0.979–2.088)
Marital status
Married	1	1	1
Partnered	1.346 (0.699–2.594)	1.097 (0.539–2.236)	1.087 (0.546–2.166)
Separated/divorced	1.793 (1.261–2.550)^**	1.464 (1.007–2.130)^*	1.324 (0.596–2.939)
Widowed	1.279 (0.602–2.719)	1.298 (0.591–2.847)	1.324 (0.899–1.950)
Never married	1.001 (0.675–1.486)	0.983 (0.649–1.487)	0.997 (0.652–1.525)
Major discrimination
Racial discrimination	—	1.163 (1.027–1.318)^*	1.130 (0.988–1.292)
Gender discrimination	—	1.622 (1.185–2.229)^**	1.549 (1.120–2.141)^**
Neighborhood characteristics
Crime problem in neighborhood	—	0.989 (0.863–1.132)	0.974 (0.053–1.112)
Drug problem in neighborhood	—	1.176 (1.025–1.350)^*	1.171 (1.112–1.342)^*
History of severe IPV (ref = no)	—	—	2.316 (1.719–3.121)^***
Pseudo R ²	0.0256	0.0382	0.0538
F	4.246	4.296	8.267

p < 0.05; ^** p < 0.01; ^*** p < 0.001.

OR, odds ratio; CI, confidence interval.

After adding sources of stress (racial and gender discrimination, neighborhood crime, and drug problems) in Model 2, education was no longer a significant predictor of mood disorder; all other results remained the same. Each additional instance of racial discrimination was associated with a 16% (CI 1.027–1.318, p < 0.05) increase in the odds of mood disorder, and each additional instance of gender discrimination was associated with a 62.2% (CI 1.622–2.229, p < 0.01) increase in the odds of mood disorder. Women who reported serious drug problems in their neighborhood had 17.6% (CI = 1.025–1.350, p < 0.05) greater odds of meeting the criteria for mood disorder. Neighborhood crime was not a significant predictor of mood disorder in Model 2. When accounting for history of SPIPV in Model 3, marital status and racial discrimination lost significance, although all other results remained similar. Women with a history of SPIPV had 2.32 (CI 1.719–2.121, p < 0.001) times higher odds of meeting criteria for a lifetime mood disorder than women without a SPIPV history.

Anxiety disorder

Table 4 provides the results from a nested binary logistic regression model series predicting the odds of any lifetime anxiety disorder for U.S. Black women. In Model 1, foreign-born Caribbean Black women had 48.7% (CI = 0.368–0.717, p < 0.001) lower odds (than US-born African American women) of meeting criteria for any anxiety disorder. High school graduates (AOR = 0.615; CI = 0.468–0.809, p < 0.001) and those with some college (AOR = 0.650; CI = 0.442–0.957, p < 0.001) had lower odds of any anxiety disorder than those without a high school diploma; those who were separated or divorced (AOR = 1.531; CI 1.080–2.172, p < 0.05) had higher odds of anxiety disorder than those who were married. Increasing age was associated with lower odds of anxiety disorder (AOR = 0.982; CI = 0.970–0.995, p < 0.01). Except for marital status, which lost significance, these results persisted after adding discrimination and neighborhood characteristics in Model 2. Each additional instance of racial discrimination was associated with 27.1% (CI = 1.131–1.428, p < 0.001) greater odds of anxiety disorder, and women who reported serious neighborhood drug problems had 18.6% (CI = 1.027–1.368, p < 0.05) greater odds of anxiety disorder. Although the odds of anxiety disorder were no longer significantly different between women with some college and those without a high school diploma in Model 3, women with a history of SPIPV had 2.61 times (CI = 2.013–3.390, p < 0.001) greater odds of meeting criteria for any anxiety disorder than women without such a history. Neither gender discrimination nor neighborhood crime problems were associated with anxiety disorder in any of the models.

Table 4.

Stepwise Binary Logistic Regression Predicting Any Lifetime Anxiety Disorder Based on Sociodemographic Characteristics and Sources of Stress, Black Women Only, National Survey of American Life (N = 2940)

Characteristics	Model 1, OR (CI)	Model 2, OR (CI)	Model 3, OR (CI)
Age (continuous)	0.982(0.970–0.995)^**	0.982 (0.969–0.994)^**	0.981(0.969–0.993)^**
Ethnicity/nativity
African American	1	1	1
US-born Caribbean Black	1.019 (0.606–1.713)	0.966 (0.570–1.637)	0.971(0.573–1.644)
Foreign-born Caribbean Black	0.513(0.368–0.717)^***	0.635(0.437–0.923)^*	0.686 (0.476–0.988)^*
At/below the FPL (ref = above)	1.240 (0.994–1.549)	1.156 (0.906–1.474)	1.101(0.861–1.408)
Occupational status
Employed	1	1	1
Unemployed	1.067 (0.747–1.526)	1.082 (0.758–1.545)	1.048 (0.723–1.519)
Not in labor force	1.093 (0.759–1.575)	1.121(0.776–1.618)	1.127 (0.767–1.657)
Education
Less than high school	1	1	1
High school graduate	0.615 (0.468–0.809)^***	0.636 (0.487–0.830)^**	0.679 (0.516–0.894)^**
Some college	0.650 (0.442–0.957)^*	0.643(.438–0.943)^*	0.691(0.467–1.164)
Bachelor's degree or more	0.669 (0.418–1.070)	0.638 (0.394–1.033)	0.715 (0.439–1.164)
Marital status
Married	1	1	1
Partnered	1.410 (0.776–2.563)	1.242 (0.662–2.327)	1.229 (0.655–2.306)
Separated/divorced	1.531(1.080–2.172)^*	1.288 (0.887–1.871)	1.127 (0.761–1.670)
Widowed	1.379 (0.869–2.188)	1.460 (0.900–2.368)	1.502 (0.918–2.456)
Never married	0.996 (0.613–1.619)	0.969 (0.595–1.579)	0.984 (0.598–1.617)
Major discrimination
Racial discrimination	—	1.271(1.131–1.428)^***	1.219 (1.079–1.377)^**
Gender discrimination	—	1.237 (0.917–1.668)	1.170 (0.857–1.597)
Neighborhood characteristics
Crime problem in neighborhood	—	1.014 (0.899–1.144)	0.993 (0.877–1.125)
Drug problem in neighborhood	—	1.186 (1.027–1.368)^*	1.182 (1.025–1.362)^*
History of severe IPV (ref = no)	—	—	2.612 (2.013–3.390)^***
Pseudo R ²	0.0233	0.0418	0.0636
F	5.290	4.507	11.11

p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Substance disorder

Table 5 provides the results from a nested binary logistic regression model series predicting the odds of any substance use disorder for U.S. Black women. In Model 1 foreign-born Caribbean Black women had 84.8% (CI = 0.079–0.292, p < 0.001) lower odds of meeting criteria for any substance use disorder, compared with US-born African American women, and higher levels of educational attainment (AOR = 0.347, CI = 0.130–0.924, p < 0.05) was associated with lower odds of substance use disorder. Age, poverty status, employment status, and marital status were not significant predictors of substance use disorder in Model 1. The results remained similar after adding discrimination and neighborhood characteristics in Model 2. Each additional instance of racial discrimination was associated with 20.4% (CI = 1.053–1.377, p < 0.01) higher odds of anxiety disorder and women who reported a serious drug problem in their neighborhood had 33.8% (CI = 1.071–1.673, p < 0.05) higher odds of anxiety disorder. All results remained the same after including severe IPV in Model 3. In this model, women with a history of SPIPV had more than three times higher odds of meeting criteria for a substance use disorder than those without a SPIPV history (AOR = 3.425; CI 2.092–5.609, p < 0.001). Neither gender discrimination nor neighborhood crime problems were associated with substance use disorder in any of the models.

Table 5.

Stepwise Binary Logistic Regression Predicting Any Lifetime Substance Use Disorder Based on Sociodemographic Characteristics and Sources of Stress, Black Women Only, National Survey of American Life (N = 2940)

Characteristics	Model 1 OR (CI)	Model 2 OR (CI)	Model 3 OR (CI)
Age (continuous)	1.006 (0.989–1.022)	1.008 (0.991–1.026)	1.005) 0.987–1.005)
Ethnicity/Nativity
African American	1	1	1
US-born Caribbean Black	0.859 (0.422–1.748)	0.780 (0.365–1.668)	0.771 (0.348–1.708)
Foreign-born Caribbean Black	0.152 (0.079–0.292)^***	0.198 (0.104–0.397)^***	0.220 (0.113–0.427)^***
At/below the FPL (ref = above)	1.545 (0.984–2.425)	1.293 (0.808–2.069)	1.234 (0.779–1.953)
Occupational status
Employed	1	1	1
Unemployed	1.049 (0.581–1.896)	0.980 (0.537–1.787)	0.906 (0.483–1.701)
Not in labor force	0.914 (0.502–1.662)	0.916 (0.491–1.706)	0.881 (0.461–1.685)
Education
Less than high school	1	1	1
High school graduate	0.445 (0.264–0.751)^**	0.488 (0.281–0.847)^*	0.516 (0.293–0.911)^*
Some college	0.713 (0.391–1.300)	0.734 (0.382–1.412)	0.766 (0.394–1.491)
Bachelor's degree or more	0.347 (0.130–9.924)^*	0.367 (0.124–1.084)	0.426 (0.150–1.215)
Marital status
Married	1	1	1
Partnered	1.949 (0.67–4.380)	1.411 (0.570–3.491)	1.339 (0.572–3.132)
Separated/divorced	1.371 (0.773–2.433)	1.101 (0.628–1.929)	0.915 (0.510–1.640)
Widowed	0.677 (0.320–1.431)	0.696 (0.309–1.567)	0.738 (0.334–1.633)
Never married	1.541 (0.800–2.970)	1.402 (0.719–2.735)	1.384 (0.730–2.623)
Major discrimination
Racial discrimination	—	1.204 (1.053–1.377)^**	1.169 (1.021–1.339)^*
Gender discrimination	—	1.120 (0.691–1.815)	0.974 (0.572–1.659)
Neighborhood characteristics
Crime problem in neighborhood	—	1.134 (0.935–1.367)	1.111 (0.899–1.373)
Drug problem in neighborhood	—	1.338 (1.071–1.673)^*	1.319 (1.055–1.649)^*
History of severe IPV (ref = no)	—	—	3.425 (2.092–5.609)^***
Pseudo R ²	0.0203	0.0317	0.0486
F	8.460	7.402	10.32

p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Suicide ideation

Table 6 provides the results from a nested binary logistic regression model series predicting the odds of suicide ideation for U.S. Black women. In Model 1, foreign-born Caribbean Black women had 46.6% (CI = 0.358–0.795, p < 0.01) lower odds of suicidality than US-born African American women. Older women (AOR = 0.978; CI 0.985–0.992, p < 0.01) had lower odds of suicidality and women who earned at least a bachelor's degree (AOR = 0.474; CI = 0.259–0.866, p < 0.05) had lower odds of suicidality than women with less than a high school diploma. These results remained the same after accounting for discrimination and neighborhood problems in Model 2. This model also found that neither discrimination measure nor neighborhood problem measure were significant predictors of suicidality. Although educational attainment lost significance when accounting for severe IPV history in Model 3, all other results remained the same. Model 3 also shows that women with a history of SPIPV had 2.674 (CI = 1.912–3.739, p < 0.001) times greater odds of suicidal ideation than women without a history of SPIPV.

Table 6.

Stepwise Binary Logistic Regression Predicting Lifetime Suicide Ideation on Sociodemographic Characteristics and Sources of Stress, Black Women Only, National Survey of American Life (N = 2798)

Characteristics	Model 1, OR (CI)	Model 2, OR (CI)	Model 3, OR (CI)
Age (continuous)	0.978^** (0.985–0.992)	0.978 (0.964–0.992)^**	0.976 (0.961–0.976)^**
Ethnicity/nativity
African American	1	1	1
US-born Caribbean Black	1.41 (0.799–2.490)	1.353 (0.738–2.483)	1.343 (0.689–2.618)
Foreign-born Caribbean Black	0.534 (0.358–0.795)^**	0.607 (0.377–0.975)^*	0.634 (0.139–0.409)^*
At/below the FPL (ref = above)	1.194 (0.863–1.654)	1.125 (0.796–1.159)	1.045 (0.727–1.502)
Occupational status
Employed	1	1	1
Unemployed	1.286 (0.841–1.967)	1.299 (0.830–2.032)	1.266 (0.799–2.006)
Not in labor force	0.965 (0.666–1.397)	0.970 (0.663–1.420)	0.983 (0.668–1.447)
Education
Less than high school	1	1	1
High school graduate	0.716 (0.478–0.179)	0.735 (0.501–1.079)	0.799 (0.545–1.172)
Some college	0.698 (0.432–1.126)	0.675(0.408–1.115)	0.733 (0.437–1.230)
Bachelor's degree or more	0.474 (0.259–0.866)^*	0.472 (0.255–874)^*	0.537 (0.287–1.007)
Marital status
Married	1	1	1
Partnered	1.268 (0.669–2.403)	1.061 (0.520–1.163)	1.059 (0.509–2.204)
Separated/divorced	1.220 (0.771–1.929)	1.033 (0.633–1.686)	0.876 (0.540–1.419)
Widowed	0.716 (0.350–1.465)	0.765 (0.379–1.544)	0.770 (0.395–1.501)
Never married	0.946 (0.563–1.589)	0.901 (0.534–1.550)	0.910 (0.534–1.550)
Major discrimination
Racial discrimination	—	1.144 (0.996–1.314)	1.092 (0.940–1.269)
Gender discrimination	—	1.417 (0.932–2.155)	1.351 (0.879–2.076)
Neighborhood characteristics
Crime problem in neighborhood	—	0.993 (0.855–1.153)	0.970 (0.840–1.121)
Drug problem in neighborhood	—	1.219 (0.983–1.513)	1.207 (0.982–1.483)
History of severe IPV (ref = no)	—	—	2.674 (1.912–3.739)^***
Pseudo R ²	0.0278	0.0379	0.0548
F	4.601	3.465	6.241

p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Discussion

This study used a social ecological theoretical framework (e.g., Stokols⁴⁵) to investigate the mental health of Black women residing in the United States, with a key focus on severe physical intimate partner violence, racial and gender discrimination, and neighborhood characteristics, using the most complete and detailed national sample of U.S. Blacks. To our knowledge, this is one of the first articles to use this approach. We fulfill a gap in the literature on IPV and mental health among Black women (often based on clinical samples) by adding interpersonal and institutional factors such as racial discrimination, gender discrimination, and neighborhood characteristics. Our research indicates that U.S. Black women, like other women, are vulnerable to poor mental health conditions resulting from IPV.⁹ This was first evident by the significantly higher rates of mental disorders among women victims of severe IPV in comparison with women who did not have such experience.

While confirming the influence of SPIPV on Black women's mental well-being,²⁸ multivariate analysis also sheds light on other important socioecological contributors. Throughout the analysis, major discrimination was shown to have a profound effect on Black women's mental well-being.⁵⁵ Specifically, racial discrimination was associated with higher odds of anxiety and substance disorders, whereas gender discrimination was related to higher odds of mood disorders. By documenting the importance of studying attribution, these findings fill an important gap in the burgeoning discrimination literature. Future research should build upon this work by studying the intersections of these and other various identity characteristics.

In addition to discriminatory practices that can have serious consequences on health and well-being, neighborhood/environmental conditions (specifically, drug problems) was further associated with poor mental health, supporting similar findings where women experiencing IPV while residing in impoverished neighborhood have predisposition for mood disorders.⁵⁶ Illustrated by this study, having drug problems in the neighborhood predisposes women to unhealthy mental well-being.

In light of the findings previously discussed, sociodemographic factors also played a role in Black women's mental health. Our study showed that the odds of mood, anxiety, and suicide ideation decrease with age. These findings may signal greater resilience and resource utilization that comes with age.⁵⁷ Black women may become more adept at dealing or coping with adverse circumstances with age or maturity. Consistent with past research on the social gradient in health,⁵⁸ relative to those who did not complete high school, high school graduates had lower odds of meeting criteria for anxiety and substance disorders.

Finally, our study findings revealed that foreign-born Caribbean women were less likely than African American women to meet criteria for anxiety disorders, substance disorders, and suicidal ideation. These findings support both findings from past research^21,36 and the healthy immigrant effect.²⁹ Moreover, these findings may further have sociocultural undertones. For example, it has been argued that Caribbean-born individuals do not readily self-identify as African Americans, a process that may diminish their sensitivity to tacit discrimination in American society that has been shown to have negative effects on the mental well-being of African Americans.⁵⁹ However, this might change with longer time spent in the country. In light of this, entrenched cultural sentiments and resentment toward drug use may reduce the likelihood of substance disorders among foreign-born individuals. In a similar vein, strong cultural sentiment against taking one's life, as it is viewed as a sign of weakness, may be a factor as to why foreign-born Caribbeans would not endorse suicidal thoughts.

Overall, this study lends support to past findings of the deleterious effects of cumulative stress for ethnic minority populations.⁶⁰ Prolonged exposure to disparate socioeconomic opportunities and perilous socioenvironmental stressors, as well as a history of IPV seems to amplify several negative psychological outcomes for Black women.⁶¹

Limitations

In interpreting these findings, there are shortcomings to the study that should be acknowledged. First, the study was based on cross-sectional data. Although comparative analysis provides certain benefits that allow for drawing inferences about the contribution of IPV to poor health, longitudinal design would better assist in drawing this conclusion. Second, the study focused on individuals who experienced SPIPV at any point over the course of their lives. Therefore, lapses in memory could cause participants to attribute certain conditions to other traumatic experiences (e.g., child abuse) that were not necessarily the result of IPV. Third, the term “badly beaten” may have different meanings for different cultural groups resulting in selection bias; hence, influencing the validity of the study findings. However, single-item measures such as the one used in this study have been consistently used in past research.^36,62 We further addressed the measure's validity by comparing the NSAL IPV measure with the NCS-R dichotomously defined severe partner violence CTS measure within the CPES and it was found to have fair associations across estimates of agreement. Finally, the measure used for this study does not account for other types of violence (e.g., minor, sexual, and psychological/emotional) that can have severe consequences on the health and well-being of women.

Even with the limitations discussed, this study is useful in providing rates of various mental conditions for Black women with and without a history of severe IPV. Through its use of a nationally representative sample, this study is also one of the few studies conducted to assist in understanding the contributions of IPV and other sources of stress to Black women's mental health. The study was based on a sample that used a structured clinical assessment tool that may be more useful in making an accurate determination about the mental health of U.S. Black women.

Conclusions

Therefore, for Black women who endure discrimination and socioeconomic disparities, IPV may further diminish their well-being by intensifying various mental health conditions. According to the study results, SPIPV is significantly and independently associated with mood disorders like major depression and dysthymia. Likewise, anxiety disorders, substance disorders, and suicidal ideation occur at significantly higher rates for women with a history of severe IPV. Increasing our understanding of social factors that exacerbate mental health issues for Black women experiencing SPIPV can assist with developing additional intervention pathways. In addition, because U.S. Black women in this study experienced anxiety disorders, substance abuse, and suicidal ideation at greater rates than foreign-born Caribbean Blacks, research on Black Caribbean immigrant women may provide novel insight to help design interventions to improve the mental health of US-born Black women. Given our findings, this study suggests an urgent need for examining multifactorial origins of SPIPV for Black women with the goal of developing wide-ranging community and clinical interventions aimed at ending IPV among Black women.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the University of Michigan-Dearborn Office of Research and Sponsored Programs and funded by a scholar grant [P/G U059355].

References

Dillion

, Hussain

, Loxton

, Rahman

. Mental and physical health and intimate partner violence against women: A review of the literature. Int J Family Med, 2013; 2013:313909.

Dutton

, Kaltman

, Goodman

, Weinfurt

, Vankos

. Patterns of intimate partner violence: Correlates and outcomes. Violence Vict, 2005; 20:483–497.

Gleason

. Mental disorders in battered women: An empirical study. Violence Vict, 1993; 1:53–68.

Golding

. Intimate partner violence as a risk factor for mental disorders: A meta-analysis. J Fam Violence, 1999; 14:99–132.

Lagdon

, Stringer

. Adult experience of mental health outcomes as a result of intimate partner violence victimisation: A systematic review. Eur J Psychotraumatol, 2014; 12 [Epub ahead of print]; DOI: 10.3402/ejpt.v5.24794.

Plichta

. Intimate partner violence and physical health consequences: Policy and practice implications. J Interpers Violence, 2004; 19:1296–1323.

Anderson

, Stockman

, Sabri

, Campbell

. Injury outcomes in African American and African Caribbean women: The role of intimate partner violence. J Emerg Nurs, 2015; 4136–4142.

Cheng

and Lo , Celia

. Racial disparities in intimate partner violence examined through the multiple disadvantage model. J Interpers Violence, 2016; 31:2026–2051.

Lacey

, McPherson

, Samuel

, Powell Sears

, Head

. The impact of different types of intimate partner violence on the mental and physical health of women in different ethnic groups. J Interpers Violence, 2013; 28:359–385.

10.

Afifi

, MacMillan

, Cox

, Asmundson

GJG

, Stein

, Sareen

. Mental health correlates of intimate partner violence in marital relationships in a nationally representative sample of males and females. J Interpers Violence, 2009; 24:1398–1417.

11.

Black

, Basile

, Breiding

, et al. The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 summary report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2011.

12.

Breiding

, Smith

, Basile

, et al. Prevalence and characteristics of sexual violence, stalking, and intimate partner violence victimization—National Intimate Partner and Sexual Violence Survey, United States, 2011. Atlanta, GA: Centers for Disease Control and Prevention, 2014.

13.

Benson

, Woolredge

, Thistlethwaite

, Fox

. The correlation between race and domestic violence is confounded with community. Soc Probl, 2004; 51:326–342.

14.

Hien

, Ruglass

. Interpersonal partner violence and women in the United States: An overview of prevalence rates, psychiatric correlates and consequences and barriers to help seeking. Int J Law Psychiatry, 2009; 32:48–55.

15.

Adler

, Ostrove

. Socioeconomic status and health: What we know and what we don't. Ann N Y Acad Sci, 1999; 896:3–15.

16.

Williams

. Race, socioeconomic status, and health: The added effects of racism and discrimination. Ann N Y Acad Sci, 1999; 896:173–188.

17.

Williams

, Neighbors

, Jackson

. Racial/ethnic discrimination and health: Findings from community studies. Am J Public Health, 2003; 93:200–208.

18.

Lacey

, Mouzon

. Severe physical intimate partner violence and the mental and physical health of U.S. Caribbean black women. J Womens Health, 2016; 25:920–929.

19.

Hankin

, Smith

, Daugherty

, Houry

. Correlation between intimate partner violence victimization and risk of substance abuse and depression among African-American women in an urban emergency department. West J Emerg Med, 2010; 11:252–256.

20.

Paranjape

, Heron

, Thompson

, Bethea

, Wallace

, Kaslow

. Are alcohol problems linked with an increase in depressive symptoms in abused, inner-city African American women?. Womens Health Issues, 2007; 17:37–43.

21.

Houry

, Kemball

, Rhodes

, Kaslow

. Intimate partner violence and mental health symptoms in African American female ED patients. Am J Emerg Med, 2006; 24:444–450.

22.

Meadows

, Kaslow

, Thompson

, Jurkovic

. Protective factors against suicide attempt risk among African American women experiencing intimate partner violence. Am J Community Psychol, 2005; 36:109–121.

23.

Ramos

, Carlson

, McNutt

. Lifetime abuse, mental health, and African American women. J Fam Violence, 2004; 19:153–164.

24.

Lacey

, Parnell

, Mouzon, et al. The mental health of U.S. Black women: The roles of social context and severe intimate partner violence. BMJ Open, 2015; 5:e008415.

25.

Leiner

, Compton

, Houry

, Kaslow

. Intimate partner violence, psychological distress, and suicidality: A path model using data from African American Women seeking care in an urban emergency department. J Fam Violence, 2008; 23:473–481.

26.

Bradley

, Schwartz

, Kaslow

. Posttraumatic stress disorder symptoms among low-income, African American women with a history of intimate partner violence and suicidal behaviors: Self-esteem, social support, and religious coping. J Traum Stress, 2005; 18:685–696.

27.

Lacey

, Powell Sears

, Govia

, Forsythe-Brown

, Matusko

, Jackson

. Substance use, mental disorders and physical health of Caribbeans at-home compared to those residing in the United States. Int J Environ Res Public Health, 2015; 12:710–734.

28.

Sabri

, Stockman

, Bertrand

, Campbell

, Callwood

, Campbell

. Victimization experiences, substance misuse, and mental health problems in relation to risk for lethality among African American and African Caribbean women. J Interpers Violence, 2013; 28:3223–3241.

29.

Markides

, Rote

. Immigrant health paradox. In: Emerging trends in the social and behavioral sciences. New York: John Wiley and Sons, 2015:1–15.

30.

Lewis

, Van Dyke

. Discrimination and the health of African Americans: The potential importance of intersectionalities. Curr Dir Psychol Sci, 2018; 27:176–182.

31.

Pascoe

, Richman

. Perceived discrimination and death: A meta-analytic review. Psychol Bull, 2009; 135:531–554.

32.

Vines

, Ward

, Cordoba

, Black

. Perceived racial/ethnic discrimination and mental health: A review and future directions for social epidemiology. Curr Epidemiol Rep, 2017; 4:156–165.

33.

Williams

, Mohammed

. Discrimination and racial disparities in health: Evidence and needed research. J Behav Med, 2009; 32:20–47.

34.

Williams

, Mohammed

. Racism and health 1: Pathways and scientific evidence. Am Behav Sci, 2013; 57:1152–1173.

35.

Keith

, Lincoln

, Taylor

, Jackson

. Discriminatory experiences and depressive symptoms among African American women: Do skin tone and mastery matter?. Sex Roles, 2010; 62:48–59.

36.

Lacey

, Sears

, Matusko

, Jackson

. Severe physical violence and Black women's health and well-being. Am J Public Health, 2015; 105:719–724.

37.

Schulz

, Gravlee

, Williams

, Israel

, Mentz

, Rowe

. Discrimination, symptoms of depression, and self-rated health among African American women in Detroit: Results from a longitudinal analysis. Am J Public Health, 2006; 96:1265–1270.

38.

Kwate

, Oyo

, Valdimarsdottir

, Guevarra

, Bovbjerg

. Experiences of racist events are associated with negative health consequences for African American women. J Natl Med Assoc, 2003; 95:450–460.

39.

Chae

, Lincoln

, Jackson

. Discrimination, attribution, and racial group identification: Implications for psychological distress among Black Americans in the National Survey of American Life (2001–2003). Am J Orthopsychiatry, 2011; 81:498–506.

40.

Mouzon

, Taylor

, Woodward

, Chatters

. Everyday racial discrimination, everyday non-racial discrimination, and physical health among African Americans. J Ethn Cult Divers Soc Work, 2017; 26:68–80.

41.

Eaton

, Muntaner

, Sapag

. Socioeconomic stratification and mental disorder. In: Scheid

, Brown

, eds. A handbook for the study of mental health. Cambridge, United Kingdom: Cambridge University Press, 2009.

42.

DuMonthier

, Childers

, Milli

. The Status of Black Women in the United States. Washington, DC: Institute for Women's Policy Research, 2017.

43.

Thoits

. Self, identity, stress, and mental health. In Aneshensel CS, Phelan JC, Bierman A (eds.): Handbook of the Sociology of Mental Health. Dordrecht: Springer, 2013; pp. 357–377.

44.

Stockdale

, Wells

, Tang

, Belin

, Zhang

, Sherbourne

. The importance of social context: Neighborhood stressors, stress-buffering mechanisms, and alcohol, drug, and mental health disorders. Soc Sci Med, 2007; 65:1867–1881.

45.

Stokols

. Establishing and maintaining healthy environments: Toward a social ecology of health promotion. Am Psychol, 1992; 47:6–22.

46.

Jackson

, Torres

, Caldwell

, et al. The National Survey of American Life: A study of racial, ethnic and cultural influences on mental disorders and mental health. Int J Methods Psychiatr Res. 2004; 13:196–207.

47.

Griffith

, Johnson

, Zhang

, Neighbors

, Jackson

. Ethnicity, nativity, and the health of American blacks. J Health Care Poor Underserved, 2011; 22:142–156.

48.

Straus

. Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. J Marriage Fam, 1979; 41:75–88.

49.

Kessler

, Molnar

, Feurer

. Applebaum M. Patterns and mental health predictors of domestic violence in the United States: Results from the National Comorbidity Study. Int J Law Psychiatry, 2001; 24:487–508.

50.

Kessler

, Aguilar-Gaxiola

, Alonso

, et al. The global burden of mental disorders: An update from the WHO World Mental Health (WMH) surveys. Epidemiol Psychiatr Soc, 2009; 18:23–33.

51.

Hanley

, McNeil

. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology, 1982; 143:29–36.

52.

Kessler

, Merikangas

. The National Comorbidity Survey Replication (NCS-R): Background and aims. Int J Methods Psychiatr Res, 2004; 13:60–68.

53.

Proctor

, Dalaker

. Economics statistics administration. Poverty in the United States 2001. Current Population Reports. S.l. Distributed by ERIC Clearinghouse. Washington, DC: Bureau of the Census,, 2002:60–219.

54.

StataCorp. Stata 11 base reference manual. College Station, TX: Stata Press, 2009.

55.

Lipsky

, Caetano

. Impact of intimate partner violence on unmet need for mental health care: Results form the NSDUH. Psychiatr Serv, 2007; 58:822–829.

56.

Mugoya

, Witte

, Bolland

, et al. Depression and intimate partner violence among African American women living in impoverished inner-city neighborhoods. J Interpers Violence, 2020; 35:899–923.

57.

Stockman

, Lucea

, Draughon

, et al. Intimate partner violence and HIV risk factors among African-American and African-Caribbean women in clinic-based settings. AIDS Care, 2013; 25:472–480.

58.

Muntaner

, Ng

, Vanroelen

, Christ

, Eaton

. Social stratification, social closure, and social class as determinants of mental health disparities. In: Aneshensel

, Phelan

JoC

, Bierman

, eds. Handbook of the sociology of mental health, 2nd ed. Dordrecht, Netherlands: Springer, 2013:205–227.

59.

Deaux

, Bikmen

, Gilkes

, et al. Becoming American: Stereotype threat effects in Afro-Caribbean immigrant groups. Soc Psychol Q, 2007; 70:384–404.

60.

Geronimus

, Hick

, Keene

, Bound

. Weathering and age patterns of allostatic load scores among blacks and whites in the United States. Am J Public Health, 2006; 96:826–833.

61.

Carlson

, McNutt

, Choi

, Rose

. Intimate partner abuse and mental health: The role of social support and other protective factors. Violence Against Women, 2002; 8:720–745.

62.

Flake

, Forste

. Fighting families: Family characteristics associated with domestic violence in five latin American countries. J Fam Violence, 2006; 21:19–29.