Older, less useful tools
The Conflicts Tactics Scale was one of the first instruments to identify partner violence by measuring interpersonal aggression. The original screen consisted of 19 questions.42 The Index of Spouse Abuse is a 30-item self-report scale designed to measure the severity or magnitude of physical and nonphysical abuse inflicted on a woman by her male partner.43 Detailed independent evaluations by experienced therapists to determine whether an individual is a victim of partner abuse, considered to be the gold standard, have been used to validate the Index of Spouse Abuse. However, the Index of Spouse Abuse and Conflicts Tactics Scale are impractical for routine use in the office due to their length and complexity. Table 4 compares these screening tests.
TABLE 4
Performance characteristics of domestic violence screening instruments
Test | LOE | Sn, %* | Sp, %* | LR+, % | LR–, % | PV+, % † | PV–, % † |
---|---|---|---|---|---|---|---|
ISA-P 28 ∂1 | 1b | 90.7 | 92.2 | 11.4 | 0.1 | 72 | 2.15 |
ISA-NP 28 ∂1 | 1b | 90.7 | 90.6 | 10.1 | 0.1 | 69 | 2.15 |
WEB 32 ∂2 | 1b | 86 | 91 | 9.56 | 0.15 | 67.8 | 3.2 |
HITS 29 ∂3 | 3b | 96 | 91 | 10.7 | 0.04 | 70.2 | 0.87 |
WAST 30 ∂3 | 3b | 83 | 75 | 3.32 | 0.23 | 42.2 | 4.82 |
*Sensitivity and specificity summarized as reported in individual studies. | |||||||
† Posttest probability was calculated assuming a pretest probability of 18%. | |||||||
Sn, sensitivity; Sp, specificity; LR+, positive likelihood ratio; LR–, negative likelihood ratio; PV+, probability of disease given a positive test; PV–, probability of disease given a negative test; ∂ , reference standard; ∂1, detailed interview; ∂2, ∂3, Index of Spouse Abuse self-identified abuse victims; ISA-P, Index of Spouse Abuse scale measuring the severity or magnitude of physical abuse inflicted on a woman by her male partner; ISA-NP, Index of Spouse Abuse scale measuring the severity or magnitude of nonphysical abuse inflicted on a woman by her male partner; WEB, Women’s Experience with Battery Scale; HITS, hurt, insult, threaten, scream; WAST, Woman Abuse Screening Tool |
How physicians can help ensure safety
Table 5 shows the strength of recommendation supporting different aspects of treatment. The care of the abused woman requires a multidisciplinary team approach involving institutional and community services.28 The literature suggests that once a victim of abuse is identified in an office setting, a primary care physician can improve outcome by caring for acute injuries,28 offering support, and making appropriate referrals.
A physician can help ensure safety by:
- Assessing immediate risk. Has the violence increased in frequency or severity over the past year? Has your partner threatened to kill you or your children? Are there weapons in the house? Does your partner know that you are planning to leave? (LOE: 5)44
- Discussing safety behaviors. This includes advice on self-protection (ie, removal of weapons from the home) and planning for leaving safely in a threatening situation. One study of abused pregnant mothers found that receiving a safety intervention protocol significantly increased the safety behaviors adopted during and after pregnancy (from 47.6% at visit 1 to 78.1% at visit 6; P.≤001), preventing further abuse and increasing the safety and well-being of mother and baby (LOE: 2c)8
- Helping the patient obtain a civil protection order. This can be obtained with the assistance of the police or community advocacy services. Women with permanent protection orders are less likely than those without orders to be physically abused (risk ratio in 12 months, 0.2; 95% confidence interval, 0.1–0.8; LOE: 2b).46
- A trusting relationship with the patient can help her break the cycle of abuse and enable her to change her circumstances (LOE: 4).47 A qualitative study showed that battered women have rated the following behaviors highly desirable in their physicians (LOE: 4).10
- Initially validates their experiences with compassionate messages and emphasizes their worth as human beings
- Clearly labels the abuse as wrong and criminal
- Listens in a careful, nonjudgmental manner.
Having someone to confide in and having told someone about the abuse were factors associated with diminished abuse at 3 months in one study (P=.001 and .023, respectively) (LOE: 2c).48
TABLE 5
Evidence supporting interventions for domestic violence
SOR | Treatment |
---|---|
A | Community-based advocacy intervention programs40 |
B | Safety intervention protocols35 |
B | Civil protection order36 |
B | Telling or confiding in someone39 |
B | Contact with community resources on domestic violence39 |
B | On-site advocacy programs41 |
C | Validating the patient’s experience38 |
C | Assessing immediate safety and emphasizing potential for lethal outcome33,34 |
SOR, strength of recommendation. For an explanation of the recommendations. |
Referral to community resources
A randomized controlled trial with 2-year follow-up investigated community-based advocacy for abused women who were leaving a shelter program. This study found that advocacy services led to significantly greater effectiveness in obtaining resources, a decrease in physical violence, a decrease in depression, and an improved quality of life and social support at 10 weeks post-shelter. At 2 years, advocacy services led to reduced physical violence (11% vs 24%, P..05, number needed to treat=7.7), increased likelihood of leaving the abusive relationship (96% vs 87%, number needed to treat=11, P.<.03), and improved quality of life (P..01) (LOE: 1b).9