Who is Identified by Screening for Intimate Partner Violence?

Background

Intimate partner violence (IPV) against women is prevalent and has significant physical and mental health consequences; accurate identification of IPV in health settings can be an important first step in appropriate response and referral to services for women.

Methods

As part of a randomized controlled trial assessing IPV screening, we assessed exposure to IPV in the past year in 5,607 women visiting one of 26 health care sites across Ontario, Canada, between August 2005 and December 2006. Women completed both the brief (8-item) Woman Abuse Screening Tool (WAST) and the longer (30-item) Composite Abuse Scale (CAS), which served as the criterion standard. This paper describes the agreement between these 2 instruments, and identifies covariates associated with being positive on both the screen and the criterion standard versus positive on the screen only.

Results

The WAST identified 22.1% of women as experiencing past year abuse, in contrast with the CAS, which identified 14.4% (κ = .63; standard error [SE], .01). Women were more likely to have the following characteristics when identified as IPV positive on both the WAST and CAS than on the WAST alone: being married (odds ratio [OR], 2.7; 95% confidence interval [CI], 1.3–5.5; p = .009), having a mental health issue (OR, 2.3; 95% CI, 1.3–4.0; p = .002), having a drug problem (OR, 1.7; 95% CI, 1.1–2.9; p = .036), and having a partner with a substance problem (OR, 2.0; 95% CI, 1.2–3.2; p = .006).

Conclusion

Screening in health care settings may overidentify IPV and care needs to be taken in decisions regarding how abuse is identified. However, screening alone may underidentify specific characteristics of women, partners, and relationships that could enable more accurate identification of abuse and specific mental health concerns through clinical case finding.