Injury Prevention
Do responses to an intimate partner violence screen predict scores on a comprehensive measure of intimate partner violence in low-income black women?

https://doi.org/10.1067/S0196-0644(03)00718-2Get rights and content

Abstract

Study objectives

Screening for intimate partner violence has been advocated as an emergency department (ED) procedure. This study aimed to ascertain whether a 5-item intimate partner violence screening questionnaire could be used effectively in the ED with low-income black women to accurately predict partner abuse status.

Methods

Data were collected from 200 black women who answered in the affirmative to at least 1 item on the intimate partner violence screener questionnaire, the Universal Violence Prevention Screening Protocol. The women completed a comprehensive battery of measures, including the Index of Spouse Abuse, a commonly used and psychometrically sound measure of intimate partner violence.

Results

Bivariate logistic regression analyses revealed that, compared with women below the physical–intimate partner violence cut point on the Index of Spouse Abuse, women above the cut point on physical–intimate partner violence on the Index of Spouse Abuse were more likely to answer yes to Universal Violence Prevention Screening Protocol screening questions related to physical, sexual, and emotional abuse; threats to be harmed physically; and being afraid. Compared with women below the nonphysical–intimate partner violence cut point on the Index of Spouse Abuse, women above the cut point on nonphysical intimate partner violence on the Index of Spouse Abuse were more likely to answer yes to each screening question on the Universal Violence Prevention Screening Protocol. The 2 Universal Violence Prevention Screening Protocol screening items related to physical abuse best predicted the 2 Index of Spouse Abuse scales. Accurate prediction of physical and nonphysical abuse on the Index of Spouse Abuse required affirmative responses to 4 or more screening questions on the Universal Violence Prevention Screening Protocol.

Conclusion

A brief intimate partner violence screening device in the emergency care setting can identify abused, low-income, black women. The study is limited by the fact that universal screening was not conducted, the inclusion of only women who acknowledged some form of intimate partner violence, a reliance on retrospective self-reports, and the questionable generalizability of the findings to groups other than low-income black women.

Introduction

Intimate partner violence is a pattern of coercive behavior that includes physical, sexual, and psychologic abuse of one partner by the other partner in the context of a current or past intimate relationship. Data reveal that women accounted for 85% of the more than 790,000 victims of intimate partner violence in 1999, and intimate partners murdered more than 1,200 women in that same year.1 Data are mixed about racial differences in rates of intimate partner violence, with some researchers finding comparable rates between groups,2 others reporting higher rates among black than white females,3 and yet other researchers indicating that between-group differences are no longer present after social class is controlled for.4 A recent study conducted by the National Institute of Justice found that black women experience intimate partner violence at rates similar to those of white women, except for black women aged 20 to 24 years, who experience significantly higher rates of intimate partner violence.1 There is also evidence that low income is related to intimate partner violence.3, 5 Intimate partner violence has already been recognized as a major public health problem for all women, including black women, and warrants special attention among low-income black women, a potentially doubly at-risk population.6, 7, 8, 9

Although intimate partner violence is associated with significant morbidity and mortality, the majority of abused women have never informed a health care provider that they have experienced coercive physical, sexual, or emotional abuse in their intimate relationships10 or do not seek medical care for injuries related to the abuse.11, 12 The reasons victims fail to report their intimate partner violence status to their health care providers are multifaceted and often include fear, humiliation, social isolation, the perception that the health care professional will not care, and the belief that the health care system will not respond.13 Fear of reporting of intimate partner violence may be particularly prominent within the black community because of the black population's general distrust of the health care system, because black women often feel the need to protect their male partners because of the discrimination often exhibited by the legal system toward black men, and because of the value placed within the black community on privacy as it pertains to family matters.14, 15, 16 In a survey of 90 primarily black women, racism and discrimination because of poverty were perceived as significant barriers to seeking help.17

The emergency department (ED) has gained increased attention as a venue for identifying victims of intimate partner violence and initiating interventions to prevent further abuse.18, 19, 20, 21, 22 A commonly held belief is that 17% to 35% of women who present to EDs are there for the medical and psychologic sequelae of intimate partner violence.18, 20, 21, 22, 23 However, when screeners assess intimate partner violence in women in EDs, prevalence rates fall below the 17% to 35% range and range from 3% to 14% for acute episodes.18, 20, 24, 25, 26 Unfortunately, physicians in the ED detect few cases of intimate partner violence, which is particularly problematic because only a small percentage of women voluntarily share their abuse status in this setting.18

One of the major issues related to screening for intimate partner violence and associated prevalence rates concerns the measurement tools to be used. Feldhaus et al25 developed a 3-item screening tool for detecting intimate partner violence in the ED, the Partner Violence Screen. Compared with more comprehensive measures of intimate partner violence, including the Index of Spouse Abuse27 and the Conflict Tactics Scale,28 the Partner Violence Screen correctly identified women in 64.5% and 71% of the cases, respectively. No information could be located on the sensitivity and specificity metrics of the Partner Violence Screen. Another screening measure, the Universal Violence Prevention Screening Protocol, was developed at George Washington University Medical Center and included 5 questions related to intimate partner violence for the previous 12 months and the past month.21 This measure was chosen as the screening tool for the current study because it is more comprehensive than the Partner Violence Screen.25 The Universal Violence Prevention Screening Protocol was adapted and modified for the current study because an item related to emotional abuse was added to address the broader spectrum of intimate partner violence–related behaviors.

The overarching objective of this study was to ascertain whether an expanded version of the Universal Violence Prevention Screening Protocol can serve as a useful screening tool for intimate partner violence in the ED with low-income black women. More specifically, to evaluate the characteristics of the Universal Violence Prevention Screening Protocol, this study was designed for the following: (1) to determine whether low-income black women who scored above the clinical cut points on the physical subscale of the Index of Spouse Abuse would be more likely than those who scored below the cut point to respond in the affirmative to each of the screening questions on the Universal Violence Prevention Screening Protocol; (2) to ascertain whether low-income black women who scored above the clinical cut point on the nonphysical subscale of the Index of Spouse Abuse would be more likely to respond in the affirmative to each of the screening questions on the Universal Violence Prevention Screening Protocol than those who scored below the cut point; (3) to discover which items on the screening measure have the highest positive predictive value and sensitivity value for the 2 subscales of the Index of Spouse Abuse; and (4) to determine the number of affirmative responses on the Universal Violence Prevention Screening Protocol that are needed for the health care provider to have confidence that intimate partner violence has been adequately assessed. We do not address the specificity of the measure because we are more interested in capturing all potential intimate partner violence cases and less concerned about obtaining false positives that would be reflected in low specificity.

Section snippets

Methods

The protocol was approved by the university and hospital's institutional review boards. Two hundred black women who reported a history of partner abuse during the previous year were recruited from the ED of a large, inner-city, Level I trauma hospital and interviewed, after providing written informed consent, as part of a larger study. Because the original intent of the larger study was to compare abused black women who attempted suicide with abused black women who had never attempted suicide,29

Results

Women in both groups ranged in age from 18 to 59 years, with the mean age of the sample being 31.96 years. The sample was low income as evidenced by the fact that only 41% of the women were employed, only 27% of the women reported a monthly household income higher than $1,000, only 56% graduated from high school, and half of the sample could read at the 9th-grade level or above. χ2 Analyses on nominal data and analyses of variance on continuous data revealed no significant between-group

Discussion

Results from this study revealed that responses to each of the 5 screening items were significantly associated with scores on a more comprehensive and psychometrically sound measure of physical and nonphysical intimate partner violence. The data gleaned provided support for the construct validity of the Index of Spouse Abuse, a more comprehensive questionnaire than the Universal Violence Prevention Screening Protocol. In addition, our findings suggested that 1 physical and 1 sexual item on the

References (45)

  • MA Straus et al.

    Physical Violence in American Families: Risk Factors and Adaption to Violence in 8,145 Families

    (1990)
  • JA Grisso et al.

    Violent injuries among women in an urban area

    New Engl J Med.

    (1999)
  • J Joseph

    Women battering: a comparative analysis of black and white women

  • PS Plass

    African American family homicide: patterns in partner, parent, and child victimization 1985-1987

    J Black Studies

    (1993)
  • P Tjaden et al.

    Prevalence, Incidence, and Consequences of Violence Against Women: Findings From the National Violence Against Women Survey

    (1998)
  • D Brookoff et al.

    Characteristics of participants in domestic violence

    JAMA

    (1997)
  • Centers for Disease Control and Prevention

    Lifetime and annual incidence of intimate partner violence and resulting injuries: Georgia, 1995

    MMWR Morb Mortal Wkly Rep.

    (1998)
  • LC Jackson et al.

    Psychotherapy With African American Women: Innovations in Psychodynamic Perspectives and Practice

    (2000)
  • B Gerbert et al.

    Experiences of battered women in healthcare settings: a qualitative study

    Women Health

    (1996)
  • Cited by (50)

    • Perception of victimization in dating in Spanish teen and young

      2015, Revista Iberoamericana de Psicologia y Salud
    • Managing intimate partner violence in the emergency department

      2015, Annals of Emergency Medicine
      Citation Excerpt :

      Despite the challenge of demonstrating improved patient-centered outcomes from emergency department (ED) screening, it can be accomplished safely and effectively without endangering patients7 and has the potential to identify patients at future risk for violence.8 A sample of brief validated screening tools appropriate for use in the ED is available9-12 (Figure 1); complete information about screening instruments is available from the Centers for Disease Control and Prevention (CDC) (Figure 2). Suggested language to use during screening is shown in Figure 3, including informing patients about the limits of confidentiality and using a brief normalizing statement.

    • Postpartum depression and intimate partner violence in urban mothers: Co-occurrence and child healthcare utilization

      2012, Journal of Pediatrics
      Citation Excerpt :

      In this clinic, universal screening for postpartum depression and intimate partner violence began in February 2008 with the implementation of a New Mothers Survey, a paper-based, self-administered 10-item tool given to mothers during all newborn, 2-month, 4-month, and 6-month well-baby visits. The survey included the US Preventive Services Task Force 2-item depression questionnaire22 and the 5-item Universal Violence Prevention Screening Protocol,23 which assesses for emotional and physical threats, physical injury, and forced sex. Both screening tools have been validated previously in similar patient populations.22,23

    • Intimate Partner Violence Screening Tools. A Systematic Review

      2009, American Journal of Preventive Medicine
      Citation Excerpt :

      Test–retest reliability was acceptable in one study.37 See Appendix A, available online at www.ajpm-online.net, for a summary of the content and quality of the 33 included studies.13–45 The 33 articles evaluated a total of 21 IPV screening tools.

    View all citing articles on Scopus

    Author contributions: SLH, NJK, and MPT contributed to writing of the manuscript. NJK, MPT, and EJ participated in the study's conception and design. NJK obtained research funding. NJK and EJ supervised the study. MPT managed and analyzed the data. SLH, NJK, and MPT edited multiple drafts of the manuscript and take responsibility for the paper as a whole.

    Presented as a poster at the Southeast Society for Academic Emergency Medicine conference, Jacksonville, FL, April 2002.

    Supported by the Association of Schools of Public Health/Centers for Disease Control and Prevention/Agency for Toxic Substances and Disease Registry grant “Why does interpersonal violence lead to suicidality in women?”

    Reprints not available from the authors.

    View full text