OBJECTIVES. To estimate the prevalence of intimate partner violence among parents at a pediatric primary care clinic and to evaluate the stability, sensitivity, specificity, positive and negative predictive values, and likelihood ratios of a very brief screen for intimate partner violence.
METHODS. A total of 200 parents (mostly mothers) bringing in children less than 6 years of age for child health supervision completed the Parent Screening Questionnaire in a primary care clinic. The Parent Screening Questionnaire, a brief screen for psychosocial problems developed for the study, includes 3 questions on intimate partner violence. Mothers then completed the computerized study protocol within 2 months. This included the Parent Screening Questionnaire as well as the Revised Conflict Tactics Scale. Different combinations of the intimate partner violence questions were evaluated against the Revised Conflict Tactics Scale.
RESULTS. A total of 12.0% of the mothers answered “yes” to at least one of the screening questions. On the standardized Revised Conflict Tactics Scale, responses ranged from 9% reporting a physical injury in the past year to 76% reporting psychological aggression. There was moderate stability of the screening questions. A single question, “Have you ever been in a relationship in which you were physically hurt or threatened by a partner?” in relation to the “physically injured” Revised Conflict Tactics Scale subscale was most effective. Sensitivity was 29%, specificity was 92%, positive predictive value was 41%, and negative predictive value was 88%. The positive likelihood ratio was 3.8, and the negative likelihood ratio was 0.77.
CONCLUSIONS. Intimate partner violence is a prevalent problem. A very brief screen can reasonably identify some mothers who could benefit from additional evaluation and possible services. Additional research is needed to find a more sensitive screen and to examine whether identifying intimate partner violence leads to interventions that benefit mothers, families, and children.