Screening Women and Elderly Adults for Family and Intimate Partner Violence: Review of the Evidence. Table 1. Studies of Screening Instruments about Intimate Partner Violence against Women

You Are Here: AHRQ Home > Clinical Information > U.S. Preventive Services Task Force (USPSTF) > Guide to Clinical Preventive Services, 2001-2004 > Screening: Family Violence, Women & Elderly > Review of the Evidence

Table 1. Studies of Screening Instruments about Intimate Partner Violence against Women

Study, Year	Population: N Age Ethnicity Socioeconomic Status Pregnancy Status	Setting	Screening Instrument(s)	Findings	Quality Rating Comments
Comparison of Screening Instruments
Coker et al., 2001⁵⁶	N: 1,152 age: mean 38 (range 18-65) ethnicity: 62% African American, 38% White socioeconomic status: 100% insured (medicaid or managed care), 89% high school graduate or greater pregnancy status: NR	2 university-affiliated family practice clinics	1) WEB, 10 items 2) ISA-P, 15 items All participants screened with both instruments	Higher detection rate with WEB scale (16%) than ISA-P (10%)	Fair Questions asked by graduate students (not health care professionals); used modified version of reference standard; administered verbally although designed as written questionnaires
Brown et al., 2000⁵⁷	N: 307 age: mean 46 (range 18-86) ethnicity: 98% White socioeconomic status: 59% employed, 59% with annual household income > $30,000, 45% with post-secondary education pregnancy status: NR	20 family practice offices	1) WAST, 8 items 2) ARI, self-report, 25 items	1) WAST and ARI results were correlated (r = 0.69; P = 0.01) 2) WAST internally consistent (Cronbach's alpha = 0.75)	Fair An additional question was added to the original 7-item WAST
Sherin et al.,1998⁵⁴	N: 259 all other demographic information: NR	Family practice offices, urban/suburban population	1) HITS, written, 4 items 2) CTS, verbal, 19 items	1) HITS internally consistent (Cronbach's alpha = 0.80) 2) HITS and CTS results were correlated (r = 0.85)	Good
Feldhaus et al., 1997⁵³	N: 322 age: mean 36 ethnicity: 45% White, 19% African American, 30% Hispanic socioeconomic status: 54% uninsured, 49% employed, 64% annual income < $15,000, 67% education level high school or greater pregnancy status: NR	2 urban, hospital-based emergency departments	1) PVS, verbal, 3 items 2) ISA, written, 30 items 3) CTS, verbal, 19 items	PVS had a higher sensitivity and specificity when compared to the ISA (65% and 80%) or CTS (71% and 84%)	Good Screening done by research assistant (not health care professional)
McFarlane et al., 1992⁶²	N: 691 age: 31% teenagers, 57% age 20-29 ethnicity: 39% African American, 34% Hispanic, 27% White socioeconomic status: 95% below poverty level pregnancy status: 100% pregnant	Public prenatal clinics	1) AAS, 3 items 2) ISA 3) CTS 4) DAS	Women identified as abused on the AAS also scored significantly higher on the ISA, CTS, and DAS	Good
Ernst et al., 2002¹⁵	N: 488 age: median 36 ethnicity: 47% White, 26% African American, 11% Hispanic socioeconomic status: NR pregnancy status: NR	Large metropolitan emergency department	1) AAS 2) OAS 3) Single question "Are you presently a victim of IPV?"	The OAS had a sensitivity of 30%, a specificity of 100%, and a positive predictive value of 100%	Good
Comparison of Screening Instrument to Interview
Morrison et al., 2000⁵⁵	N: 1,000 all other demographic information: NR	Charts reviewed in emergency department, tertiary care hospital	1) Emergency Department Domestic Violence Screening Questions, 5 items 2) Standard interview, chart review	1) Retrospective review of charts identified 4 patients (0.4%) as past or present victims of domestic violence 2) Higher detection rate with questionnaire (4% acute, 7% probable, 4% past abuse)	Poor Inappropriate reference standard (interview not defined)
Canterino et al., 1999⁵¹	N: 224 age: mean age 24 ethnicity: 54% African American, 30% White, 11% Hispanic socioeconomic status: 36% employed pregnancy status: 100% pregnant	Prenatal clinic, community-based tertiary care center	1) Domestic Abuse Assessment Questionnaire, self-report, 5 items 2) Directed interview	Self-report questionnaire yielded higher detection rate (85% vs 59%; P = 0.03)	Poor Inappropriate reference standard (interview not defined)
Norton et al., 1995⁵²	N: 334 age: mean 23 ethnicity: 50% White socioeconomic status: 42% uninsured pregnancy status: 100% pregnant	Prenatal visit, interviewed by social services	1) AAS, 5 items 2) Standard interview, chart review	More frequent detection of violence using AAS (41%) compared with interview (14%)	Poor Inappropriate reference standard (interview not defined)
Internal Consistency of Screening Instrument
Pan et al.,1997⁴⁹	N: 90 age: mean 38 ethnicity: 82% White 6% African American, 7% Hispanic, 3% Asian socioeconomic status: 13.7 yrs average education, $32,000 mean annual family income, 38% employed pregnancy status: NR	Suburban family practice clinic, tertiary care university hospital	1) Partner Abuse Interview, 11 items, (modified CTS)	Internally consistent (Cronbach's alpha = 0.82)	Fair Small sample size, inappropriate reference standard (not compared to another method)
Smith and Marth, 1995⁶³	N: 389 age: NR ethnicity: 85% White socioeconomic status: 68% employed, 61% education level high school or greater pregnancy status: NR	Various primary care clinics and community groups	1) WEB Scale, 10 items	High internal consistency (Cronbach's alpha for full sample = 0.99; battered = 0.93; non-battered = 0.86)	Fair Inappropriate reference standard (not compared to another method)
Comparison of Methods of Administration of Screening Instrument
Glass et al., 2001⁴²	N: 4,641 age: 18 and older all other demographic information: NR	Emergency departments at 11 community hospitals	1) AAS as part of intake survey; patients chose whether to self administer or have it read by a nurse interviewer	Prevalence of lifetime and past-year abuse was higher with self-administered questions	Fair Patients self-selected method
Furbee et al., 1998⁵⁸	N: 175 age: mean 34 all other demographic information: NR	Emergency department, rural university-affiliated	1) Face-to-face interview 2) Tape-recorded questionnaire with written answer sheet	Comparable results (16% prevalence of abuse detected with face-to-face interview compared with 15% detected with taped interview)	Fair Narrow spectrum of patients
McFarlane et al., 1991⁵⁰	N: 777 age: 59% in age range 20-29 ethnicity: 47% African American, 34% White, 17% Hispanic socioeconomic status: NR pregnancy status: NR	Planned Parenthood clinic	1) Self-Report, 4 items 2) Interview, 4 items	Higher prevalence of abuse was detected by nurse-conducted interview (29%) than by self-report (7%)	Fair Narrow spectrum of patients
Notes: AAS, Abuse Assessment Screen; ARI, Abuse Risk Inventory; CTS, Conflict Tactics Screen; DAS, Danger Assessment Screen; HITS, Hurt, Insulted, Threatened, and Screamed at; IPV, Intimate Partner Violence; ISA, Index of Spouse Abuse; ISA-P, Index of Spouse Abuse-Physical Scale; NR, not reported; OAS, Ongoing Abuse Screen; PVS, Partner Violence Screen; WAST, Women Abuse Screening Tool; WEB, Women's Experience with Battering Scale.

Return to Document