Evaluating the Performance of the Patient Health Questionnaire-2 as a Depression Screener in Women With Histories of Intimate Partner Violence With and Without Related Traumatic Brain Injuries.

Intimate partner violence (IPV) is a known risk factor for depression. The Patient Health Questionnaire-2 (PHQ-2) is widely used as an initial depression screener before the administration of the full Patient Health Questionnaire-9 (PHQ-9), yet its screening performance among women with histories of IPV, either with or without traumatic brain injuries (TBIs), has not been systematically evaluated. This study examined the performance of the PHQ-2 in identifying such women who meet criteria for moderate-to-severe depressive symptoms. Women aged 18 years and older (N = 216) with self-reported lifetime physical IPV experience completed the PHQ-9. Moderate-to-severe depressive symptoms were defined as PHQ-9 ≥10, a validated symptom threshold used in many screening evaluation studies. TBI was assessed with the Ohio State University ID Method and the Brain Injury Severity Assessment. Receiver operating characteristic analyses estimated the PHQ-2's area under the curve, sensitivity, specificity, and standard error. Cohen's κ coefficients quantified concordance between PHQ-2 and PHQ-9 classifications. Analyses were repeated among women reporting IPV within the past 12 months to assess consistency by recency of IPV experience and were additionally stratified by presence or absence of IPV-related TBI. Across all women with IPV experience, the PHQ-2 demonstrated high overall accuracy but limited sensitivity and only moderate agreement with the PHQ-9, indicating that many women who would have screened positive on the PHQ-9 were not identified by the PHQ-2. In addition, sensitivity and concordance were lowest among women with IPV-related TBIs. Limited PHQ-2 sensitivity suggests that reliance on the PHQ-2 alone could prevent many women with histories of IPV from advancing to the next level of clinical assessment for depression, potentially delaying care.