Abstract
BACKGROUND: Intimate partner violence (IPV) is associated with adverse health effects and increased healthcare utilization. Systems-level interventions have been shown to be effective in identifying and referring survivors but little is known about how these strategies impact future utilization. The objective of this study is to examine the impact of a systems-level response on healthcare utilization among patients screening positive for IPV from November 2016 to February 2019 in a large multi-specialty outpatient health system in the Midwest. METHODS: Using electronic health record (EHR) data, we identified patients who screened positive for IPV (N = 756) and categorized their response as accepted printed material (N = 116), accepted direct referrals (N = 85), declined both (N = 271), or missing (N = 255). We used negative binomial models to model post-period utilization as a function of decision group, pre-period utilization, and clinical and demographic factors. RESULTS: After controlling for demographic characteristics and baseline utilization, the printed materials and direct referral groups had higher utilization rates than those who declined printed materials and direct referral during the post-period for every type of service. However, these differences were only statistically significant for outpatient, behavioral health, and social work services. Specifically, the visit rate for patients receiving printed materials was two times higher (rate ratio: 2.18; 95% CI: 1.21, 3.94) for behavioral health services and three times higher (rate ratio: 3.33; 95% CI: 1.3, 8.52) for social work services compared to those who refused printed material and direct referral. For those opting for a direct referral, the visit rate was two times higher for outpatient services (rate ratio: 1.97; 95% CI: 1.13, 3.42) compared to those who refused. CONCLUSIONS: Patients receiving printed materials or direct referrals had more social work and behavioral health visits, highlighting an important outcome of the protocol. However, higher utilization rates among outpatient services and a trend toward higher utilization of other services, including the emergency department, suggest greater health service utilization is not diminished by the systems level response-at least not within a two-year time frame.