Intimate partner violence and depression among Whites, Blacks, and Hispanics

Intimate partner violence and depression among Whites, Blacks, and Hispanics

Intimate Partner Violence and Depression among Whites, Blacks, and Hispanics RAUL CAETANO, MD, PHD, AND CAROL CUNRADI, PHD, MPH PURPOSE: To examine t...

96KB Sizes 24 Downloads 151 Views

Intimate Partner Violence and Depression among Whites, Blacks, and Hispanics RAUL CAETANO, MD, PHD, AND CAROL CUNRADI, PHD, MPH

PURPOSE: To examine the relationship between intimate partner violence and depression. METHODS: A household probability sample of Whites (n ⫽ 616), Blacks (n ⫽ 377), and Hispanics (n ⫽ 592) age 18 or older was interviewed in 1995. The response rate was 85%. Logistic analysis is used to identify predictors of depression. RESULTS: Among men, Black (OR ⫽ .29; 95% CI, 0.13–.65) and Hispanic (OR ⫽ 0.4; 95% CI, 0.2–0.8) ethnicity were protective against depression. Factors of risk for men included victimization by female to male partner violence (OR ⫽ 4.04; 95% CI, 1.15–14.11), unemployment (OR ⫽ 7.65; 95% CI, 1.59–16.39), and living in a high-unemployment neighborhood (OR ⫽ 4.6; 95% CI, 1.86–11.37). Among women, the predictors are perpetration of moderate (OR ⫽ 4.08; 95% CI, 1.33–12.47) or severe (OR ⫽ 6.57; 95% CI, 1.76–24.52) female to male partner violence, and impulsivity (OR ⫽ 1.82; 95% CI, 3.87–20.71). CONCLUSIONS: Knowledge from surveys using general population samples is important for developing prevention interventions in the community. Because predictors of depression in these samples are both individual and contextual at neighborhood level, prevention interventions to be effective must address not only individual factors of risk but also structural conditions in the environment where individuals live. Ann Epidemiol 2003;13:661–665. 쑕 2003 Elsevier Inc. All rights reserved. KEY WORDS:

Ethnicity, Domestic Violence, Depression.

INTRODUCTION Intimate partner violence (IPV) is a major public health problem in the US. Recent reports based on the data being analyzed in this article showed that the rate of male-tofemale partner violence (MFPV) and female-to-male partner violence (FMPV) among US couples was 13.6% and 21.4%, respectively (1). Higher rates of IPV have also been found among certain ethnic groups (2, 3). Cross-ethnic analyses of the data in this article indicated that the rate of MFPV was 23% for Black, 17% for Hispanic, and 12% for White couples (4). Rates of FMPV were 30% for Black, 21% for Hispanic, and 16% for White couples (4). The differences between Blacks and Whites remained after controlling for socioeconomic factors. Existing evidence indicates a strong and consistent association between psychological distress or depression and IPV. For instance, the prevalence of IPV among women diagnosed with depression is twice that of the general population (5), and physical abuse has been identified as one of the most important risk factors for suicide among women

From the Dallas Regional Campus, University of Texas School of Public Health, Dallas, Texas. Address correspondence to: Raul Caetano, M.D., Ph.D., Dallas Regional Campus, University of Texas School of Public Health, 5323 Harry Hines Boulevard, Room v8.112, Dallas, TX 75390-9128. Tel.: (214) 648-1080; Fax: (214) 648-1081. E-mail: [email protected] Received June 24, 2002; accepted January 3, 2003. 쑕 2003 Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010

(6, 7). Women reporting IPV are two to three times more likely to be depressed than women without history of victimization by violence (8, 9). Previous research also indicates an increased prevalence of IPV among depressed men (10). Drinking and alcohol-related problems have also figured prominently as a risk factor for IPV in research over the last 30 years (4, 11, 12). In addition to individual-level risk factors, recent research suggests that neighborhood or socioenvironmental factors may also be associated with increased risk for depression. For example, Yen and Kaplan (13) reported a twofold increase for depression among residents of poverty areas characterized by high unemployment, crime, and family problems in the Alameda County Study. Indeed, many of the factors of risk for depression, such as unemployment, frequent residence moves, and multiple stressful life events (14–16) characterize the life of residents of poor and ethnic minority neighborhoods. With this background, this article examines whether depression is an outcome of the presence of IPV in an intimate relationship. Drinking problems and sociodemographic, psychological, and contextual variables representing neighborhood characteristics are also considered in the analysis. Based on our own and others’ research, we hypothesize that IPV, alcohol problems, and contextual factors, such as neighborhood poverty, will be associated with an increased risk of depression. A potential common thread linking all these areas is stress. IPV, problematic drinking, 1047-2797/03/$–see front matter doi:10.1016/S1047-2797(03)00296-5

662 Caetano and Cunradi

AEP Vol. 13, No. 10 November 2003; 661–665

INTIMATE PARTNER VIOLENCE AND DEPRESSION

Selected Abbreviations and Acronyms OR ⫽ odds ratio CI ⫽ confidence interval IPV ⫽ intimate partner violence MFPV ⫽ male to female partner violence FMPV ⫽ female to male partner violence CES-D ⫽ Center for Epidemiologic Studies Depression Scale SUDAAN ⫽ Software for the Statistical Analysis of Correlated Data

and depression have all been recognized as abnormal responses to life under stressful conditions (17). Life in neighborhoods characterized by poverty, high unemployment rates, lack of economic opportunities and breakdown of social networks has been identified as stressful, thus creating a context in which these abnormal responses develop side by side (17–20).

METHODS Sample and Data Collection Subjects constituted a probability sample of married and cohabiting couples 18 years of age and older living in households in the 48 contiguous United States. A total of 1635 couples participated in the study for a response rate of 85%. Both members of the couple were interviewed independently in face-to-face interviews. The present analysis is not focused on couples but individuals. This is because the depression scale (CES-D) was given only to one of the partners. Measurements Intimate partner violence: Violence was considered as present (past 12 months) when at least one of the partners reported a violent incident. Using the Conflict Tactics Scale, Form R (21) and in accordance with previous reports (22, 23) moderate violence consisted of any of the following: threw something at partner; pushed, shoved or grabbed; slapped. Severe violence was represented in the scale by kicked, bit or hit; hit or tried to hit with something; beat up; choked; burned or scalded; forced sex; threatened with a knife or gun; used a knife or gun. Depression: Assessed with the Center for Epidemiologic Studies Depression Scale (CES-D) (24). The scale has a high level of internal consistency (24), and a good level of agreement between the cut off point of 16 or more (used in this article) and a clinical diagnosis of depression (25–27). Alcohol consumption: The respondent’s frequency and quantity of drinking over the 12-month period prior to the survey was combined and used to estimate the average

number of drinks (a four-ounce glass of wine, a twelveounce can of beer or one ounce of spirits) of alcohol consumed weekly. Alcohol problems: The items address 14 specific problem areas covering the 12 months prior to the survey: salience of drinking, impaired control, withdrawal, relief drinking, tolerance, binge drinking, belligerence, accidents, health-related problems, work-related problems, financial problems, problems with the police, problems with the spouse, and problems with persons other than the spouse. Sociodemographic: See Caetano (4) for a detailed description of these variables. Briefly, ethnic identification was based on self-identification as Hispanic (Black or White), Black not of Hispanic origin, and White not of Hispanic origin. The age of respondents was measured continuously in years. Household income was analyzed as a continuous variable. Male respondents were categorized into 4 employment categories: retired; unemployed; other (disabled, in school, volunteer, in job training program); and employed, which was the reference group. Female respondents had one additional category, homemaker. Individuals’ marital status was classified as either married or cohabiting (reference group). Each respondent was asked about the highest grade or year in school that he/she completed. Neighborhood characteristics: These data are from the 1990 US Census (28). Undereducation. Following Krieger’s (29) census-based methodology, tracts where at least 25% of those 25 years and older did not have a high school diploma were characterized as “undereducated.” Unemployment. Categorized as high (⬎10%), medium (5–10%) and low (⬍ 5%), with the last category serving as the comparison group. Working-class composition. Working class if 66% or more of the employed persons age 16 years and older were classified as being in working class occupations. Poverty. Coded as “1” if 20% or more of the tract’s population was below the poverty line ($12,700 annual income for a family of four) (30), and “0” if otherwise. Psychological: Childhood violence victimization. Positive reports of having been hit with something, beaten up, choked, burned or scalded, threatened with knife or gun, had a knife or gun used against them. Exposure to violence between parents. A positive response to observation of threats of violence or actual violence between parents/guardians during childhood or adolescence. A dichotomous variable was created based on any positive response to this item. Data Analysis All analyses used the Software for Survey Data Analysis (SUDAAN) statistical package (31). Analyses were conducted on data weighted to correct for probability of selection into the sample and non-response rates. The regression analyses were conducted using a two-stage process. The first

AEP Vol. 13, No. 10 November 2003; 661–665

663

Caetano and Cunradi INTIMATE PARTNER VIOLENCE AND DEPRESSION

stage consisted of cross-tabulations examining the association between neighborhood-level variables and depression. Variables with statistically significant associations (p ⬍ .05) were retained. Among men these included neighborhoods with high unemployment and those with high poverty. For women, the variables were working-class neighborhood, neighborhoods with high undereducation, and impoverished neighborhoods. The second stage in the analysis consisted of developing separate gender-specific logistic models predicting depression. All variables were entered in the model as a block.

not reporting this type of violence (Table 2a), independent of ethnicity. Comparison across genders shows that among White women the prevalence of depression is equal to the prevalence among White men. Among Blacks and Hispanics, however, cross-gender comparisons within MFPV and no-MFPV groups show that the prevalence of depression is higher among women than among men. The prevalence of depression for FMPV has the same magnitude and show the same relationship as the data for MFPV, with the exception that the prevalence is higher among White men reporting FMPV than among White men reporting MFPV (Table 2b). Multivariate Regressions Predicting Depression

RESULTS Prevalence of Depression and IPV The prevalence of depression among men is similar across ethnic groups and higher in the violence than in the noviolence group (Table 1). Data for women also show that the prevalence of depression is higher in the violence than in the no-violence group. However, the cross-ethnic variation in prevalence is different among women than among men. Among women, prevalence is higher among Hispanics and Blacks than Whites. Also, the prevalence of depression for Black and Hispanic women in the “no violence” group is relatively high, more so than the prevalence for men and for White women also in the no violence group. Prevalence of Depression by Type of Violence (MFPV and FMPV), Gender, and Ethnicity

Among men, results show that Black and Hispanic ethnicity are protective factors against depression, but unemployment, “Other” employment, FMPV (moderate) and living in a neighborhood with a high unemployment level are

TABLE 2. Association between depression and male to female or female to male partner violence by race/ethnicity and gender a) Male-to-female partner violence (MFPV) Men depressed White

MFPV No MFPV χ2

Violence No violence χ2

20% 7%

Black

(71) (263)

%

6.53a

(79) (141)

%

(100) (258) 6.00a

Women (CES-D Score ⬎ 16)

Violence No violence χ2

Black

%

(N)

16% 8%

(70) (288)

0.73

(N)

%

(N)

%

(N)

20% 6%

(58) (224)

30% 16%

(58) (99)

38% 15%

(70) (164)

Black (N)

%

(N)

%

(N)

14% 9%

(40) (242)

31% 18%

(41) (116)

48% 15%

(45) (189)

0.67

4.13a

2.21

White

FMPV No FMPV χ2

Black (N)

%

(N)

%

(N)

23% 7%

(63) (271)

18% 4%

(71) (149)

17% 7%

(84) (274)

7.30b

4.89a

White

b

a

3.34

Hispanic

%

5.00a

Women depressed

a

p ⬍ 0.01. p ⬍ 0.05. *Here and in Table 2 Ns are the bases for the percentages.

Hispanic

%

Men depressed

FMPV No FMPV χ2

3.14

4.54a

3.84

Hispanic

%

3.78b

(N) (54) (166)

(N)

17% 8%

5.02a

White

% 18% 5%

Hispanic (N)

17% 4%

(N) (42) (292)

b) Female-to-male partner violence (FMPV)

Men (CES-D Score ⬎ 16)

(N)*

% 14% 9%

White

MFPV No MFPV χ2

TABLE 1. Prevalence of depression by violence status, ethnicity, and gender

%

Hispanic

Women depressed

Considering MFPV first, the prevalence of depression is higher among those who report MFPV than among those

White

Black

p ⬍ 0.05. p ⬍ 0.01.

b

Black

Hispanic

%

(N)

%

(N)

%

(N)

26% 6%

(47) (235)

34% 16%

(52) (105)

43% 15%

(56) (178)

4.97a

4.73a

4.69a

664 Caetano and Cunradi INTIMATE PARTNER VIOLENCE AND DEPRESSION

factors of risk for depression (Table 3). Among women, FMPV (both moderate and severe) and an impulsive personality are factors of risk for depression. The effects of potential interactions between gender and IPV and between ethnicity and IPV on depression were also tested in this analysis. Each interaction term was added to each genderspecific model at a time, and improvement in model fit was assessed by comparing chi-square for the model with and without the interaction effect (df ⫽ 1). Addition of the interaction terms did not show a significant improvement in model fit in any case.

DISCUSSION The Association between IPV and Depression The prevalence of depression is higher among those groups reporting violence than in the other groups. The prevalence of depression is particularly high among minority women, and this is independent of whether these women report MFPV or FMPV. There are a number of reasons why violence could affect minority women more than it affects TABLE 3. Predictors of depression among white, black, and hispanics:Odds ratios and (95% confidence intervals) Men OR (95% CI) Race/ethnicity Black Hispanic Employment Unemployed Other MFPV Moderate Severe FMPV Moderate Severe Neighborhood unemployment Higha Neighborhood working– class compositiona Impulsivity Childhood physical abuse Exposure to parental violence Any alcohol problems past 12 mos. Change of 5 drinks in weekly volume a

Women OR (95% CI)

0.29 (0.13–0.65)c 0.40 (0.20–0.81)b

1.71 (.73–4.01) 1.36 (.58–3.22)

7.65 (2.04–28.73)c 5.11 (1.59–16.39)c

3.03 (.32–28.75) 0.80 (0.32–1.97

0.49 (0.15–1.63) 1.47 (0.39–5.55) 4.04 (1.15–14.11)b 2.65 (0.81–8.66)

4.60 (1.86–11.37)c

.59 (.21–1.64) .94 (.23–3.81) 4.08 (1.33–12.47)c 6.57 (1.76–24.52)c



– 1.24 (0.71–2.17) 0.72 (0.31–1.60)

1.27 (.42–3.80) 1.85 (1.19–2.86)b 1.14 (0.45–2.86)

0.99 (0.52–1.91)

1.27 (0.53–3.01)

0.88 (0.18–4.25)

2.88 (0.76–10.90)

0.99 (0.88–1.12)

0.76 (0.33–1.20)

Odds ratios not shown because model building procedure did not support the inclusion of the variable. p ⬍ 0.05. c p ⬍ 0.01. b

AEP Vol. 13, No. 10 November 2003; 661–665

White women. First, minority women have a higher prevalence of depression than White women do, independent of violence. This is seen in Table 2, where the prevalence among Black and Hispanic women who do not report IPV is two and a half to three times higher than among their White counterparts. Studies of depression in the general population confirm this finding (32). Second, the frequency and severity of IPV might be higher among Black and Hispanic women than among White women. Data on frequency of violence are not available in this study, and thus this hypothesis cannot be tested. Previous analysis of this data set showed that rates of major violence (e.g., beat up, choked, forced sex, threatened with a knife or gun) are similar across women in the three ethnic groups (4). In the regression analysis, only FMPV is an important risk factor for depression, independent of gender. This strong association of FMPV with depression may be a result of individuals’ reactions to an event that they see as unusual, i.e., that women, who are usually associated with nurturing roles, would perpetrate violence against their partners. Although not statistically significant, the odds ratio indicates that men involved in a relationship with severe FMPV are 2.6 times more likely than men not involved in a violent relationship to be depressed. Neither alcohol consumption nor drinking problems were associated with depression. Yet, clinical data in the literature as well as general population data (33–36) show a consistent relationship between these variables and depression. There are a number of potential reasons to explain this negative result. First, the sample under analysis is a sample of couples in the general population, which is not representative of the population as a whole. Second, the association between drinking and depression has been consistently reported in the clinical literature but not in the general population. Clinical samples have much more severe levels of depression as well as higher levels of alcohol consumption and more severe drinking problems. Finally, one of the variables representing neighborhood characteristics predicts depression—high unemployment is associated with depression among men. It is not difficult to understand why unemployment would affect men more than women. In spite of increasing participation of women in the workforce, societal and gender roles still make employment much more of a part of men’s identity than women’s. Thus, men who live in neighborhoods with a high level of unemployment are more likely to be depressed than other men independent of their own employment status.

Work on this paper was supported by a grant (R37-AA10908) from the National Institute on Alcohol Abuse and Alcoholism to The University of Texas School of Public Health, Houston, Texas. Data collection also had the support of a National Alcohol Research Center grant (AA05595) from the National Institute on Alcohol Abuse and Alcoholism to the Alcohol Research Group, Public Health Institute, Berkeley, California.

AEP Vol. 13, No. 10 November 2003; 661–665

REFERENCES 1. Schafer J, Caetano R, Clark C. Rates of intimate partner violence in the United States. Am J Public Health. 1998;88:1702–1704. 2. Kantor GK. Refining the brushstrokes in portraits of alcohol and wife assaults. In: Martin SE, ed. Alcohol and Interpersonal Violence: Fostering Multidisciplinary Perspectives. Rockville, MD: US Department of Health and Human Services; 1993:281–290. 3. Straus MA, Smith C. Violence in Hispanic families in the United States: Incidence rates and structural interpretations. In: Straus MA, Gelles RJ, eds. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 families. New Brunswick, NJ: Transaction Books; 1990:341–363. 4. Caetano R, Cunradi C, Schafer J, Clark C. Intimate partner violence and drinking among white, black and Hispanic couples in the US. J Subst Abuse. 2000;11:123–138. 5. Dienemann J, Boyle E, Baker D, Resnick W, Wiederhorn N, Campbell J. Intimate partner abuse among women diagnosed with depression. Issues Ment Health Nurs. 2000;21:499–513. 6. Bergman B, Brismar B. Suicide attempts by battered wives. Acta Psychiatr Scand. 1991;83:1486–1489. 7. Stark E, Flitcraft A. Killing the beast within: Women battering and female suicidality. Int J Health Serv. 1995;25:43–64. 8. Petersen R, Gazmararian J, Clark KA. Partner violence: Implications for health and community settings. Womens Health Issues. 2001;11:116–125. 9. Bauer HM, Rodriguez MA, Stable EP. Prevalence and determinants of intimate partner abuse among public hospital primary care patients. J Gen Intern Med. 2000;15:811–817. 10. Feldbau-Kohn S, Heyman RE, O’Leary KD. Major depressive disorder and depressive symptomology as predictors of husband to wife physical aggression. Violence Vict. 1998;13:347–360. 11. Caetano R, Schafer J, Cunradi C. Alcohol-related intimate partner violence among White, Black, and Hispanic couples in the United States. Alcohol Health Res World. 2001;25:58–65. 12. Cunradi C, Caetano R, Clark C, Schafer J. Alcohol-related problems and intimate partner violence among White, Black, and Hispanic couples in the US. Alcohol Clin Exp Res. 1999;23:1492–1501.

Caetano and Cunradi INTIMATE PARTNER VIOLENCE AND DEPRESSION

665

18. Marmot M. Multilevel approaches to understanding social determinants. In: Berkman LF, Kawachi I, eds. Social Epidemiology. New York: Oxford University Press; 2000:349–367. 19. Berkman LF, Glass T. From social integration to health: Durkheim in the new millennium. Soc Sci Med. 2000;51:137–173. 20. Brunner R, Parzer P, Schuld V, Resch F. Dissociative symptomatology and traumatogenic factors in adolescent psychiatric patients. J Nerv Ment Dis. 2000;188:306–331. 21. Straus MA. The conflict tactics scales and its critics: An evaluation and new data on validity and reliability. In: Straus MA, Gelles RJ, eds. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 families. New Brunswick, NJ: Transaction Books; 1990:49–73. 22. Forjuoh SN, Coben JH, Gondolf EW. Correlates of injury to women with partners in batterer treatment programs. Am J Public Health. 1998;88: 1705–1708. 23. Leonard KE, Quigley BM. Drinking and marital aggression in newlyweds: An event-based analysis of drinking and the occurrence of husband marital aggression. J Stud Alcohol. 1999;60:537–545. 24. Radloff L. The CES-D Scale: A self-report depression scale for research in the general population. Appl Psych Meas. 1977;1:385–401. 25. Radloff L, Locke BZ. The community mental health survey and the CESD scale. In: Weissman M, Myers J, Ross C, eds. Community Surveys of Psychiatric Disorders. Surveys in Psychosocial Epidemiology. New York: Rutgers University Press, 1986: 177–189. New York: Prodist; 1983. 26. Weissman MM, Sholomskas D, Pottenger M, Prusoff V, Locke B. Assessing depressive symptoms in five psychiatric populations: A validation study. Am J Epidemiol. 1977;106:203–214. 27. Vega WA, Kolody B, Valle R, Hough R. Depressive symptoms and their correlates among immigrant Mexican women in the United States. Soc Sci Med. 1986;22:645–652. 28. United States Bureau of the Census. United States Census 1990. US Department of Commerce, Economics, and Statistics. Washington D.C.: United States Bureau of the Census; 1992. 29. Krieger N. Overcoming the absence of socioeconomic data in medical records: Validation and application of a census-based methodology. Am J Public Health. 1992;82:703–710. 30. Federal Register. Annual update of the poverty income guidelines. Washington, D.C.: U.S. Government Printing Office, 1990.

13. Yen IH, Kaplan GA. Poverty area residence and changes in depression and perceived health status: Evidence from the Alameda County Study. Int J Epidemiol. 1999;28:90–94.

31. Research Triangle Institute. Software for survey data analysis (SUDAAN). Version 6.30. Research Triangle Park, NC: Research Triangle Institute; 1994.

14. Gutierrez-Lobos K, Woelfl G, Scherer M, Anderer P, Schmidl-Mohl B. The gender gap in depression reconsidered: The influence of marital and employment status on the female/male ratio of treated incidence rates. Soc Psychiatry Psychiatr Epidemiol. 2000;35:202–210.

32. Jones-Webb RJ, Snowden LR. Symptoms of depression among Blacks and Whites: Differential patterns of risk. Am J Public Health. 1993;83:240–244.

15. Brown RL, Rounds LA. Conjoint screening questionnaires for alcohol and other drug abuse: Criterion validity in a primary care practice. Wis Med J. 1995;94:135–140. 16. Gonzalez HM, Haan MN, Hinton L. Acculturation and the prevalence of depression in older Mexican Americans: Baseline results of the Sacramento Area Latino Study on Aging. J Am Geriatr Soc. 2001;49:948–953. 17. Cazenave NA, Straus MA. Race, class, network embeddedness, and family violence: A search for potent support systems. In: Straus MA, Gelles RJ, eds. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 families. New Brunswick, NJ: Transaction Publishers; 1995:321–335.

33. Brennan PL, Moos RH, Mertens JR. Personal and environmental risk factors as predictors of alcohol use, depression, and treatment-seeking: A longitudinal analysis of late-life problem drinkers. J Subst Abuse. 1994; 6:191–208. 34. Brown SA, Schuckit MA. Changes in depression among abstinent alcoholics. J Stud Alcohol. 1988;49:412–417. 35. Charney D, Paraherakis A, Negrete J, Gill K. The impact of depression on the outcome of addictions treatment. J Subst Abuse Treat. 1998; 15:123–130. 36. Jones RJ. The Socio-economic Context of Alcohol Use and Depression: Results from a National Survey of Black and White Adults in the General Population. Berkeley, CA: School of Public Health, University of California, Berkeley; 1989.