Social Determinants of MRI-driven Management Changes in Infective Endocarditis: The Role of Insurance, Homelessness, and Intravenous Drug Use
Meha Prabhu1, Shreya Tripathy1, Steven Feske2
1Boston University Chobanian and Avedisian School of Medicine, 2Boston Medical Center, Neurology Department
Objective:
We hypothesized that insurance status, homelessness, and intravenous drug use (IVDU) significantly influence the likelihood of diagnostic and management changes following brain MRI or MRA in patients with infective endocarditis (IE) presenting with neurologic complications.
Background:
IE often affects socially and economically marginalized populations with homelessness and IVDU being major risk factors. Neurologic complications occur in up to 55% of cases and worsen outcomes, yet current guidelines do not consistently recommend neuroimaging. Limited evidence exists on how social determinants affect the utilization and clinical impact of MRI/MRA in IE.
Design/Methods:
We retrospectively analyzed the EMR data of 374 IE patients (2014–23) at a large urban safety-net hospital. Among them, 162 underwent brain MRI or MRA. Provider documentation before and after MRI was reviewed to identify changes in diagnosis, medication, or surgical management. Logistic regression assessed associations between these outcomes and homelessness, IVDU, and insurance type (α = 0.05).
Results:
MRI findings led to diagnostic modifications in 78.9% of cases, medical management changes in 54.9%, and surgical changes in 16.6%. Uninsured patients had greater odds of diagnostic modification (OR = 7.66, p = 0.05), and privately insured patients had higher odds of medical management change (OR = 4.92, p = 0.045). IVDU was associated with a trend toward increased surgical intervention (OR = 2.26, p = 0.085). Homelessness showed no significant association with diagnostic or management changes.
Conclusions:
Insurance type and IVDU status are strongly correlated with how imaging findings are translated into medical or surgical interventions. Future studies should explore the possible reasons for these association.
10.1212/WNL.0000000000217505
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.