Presentation (CTI)

OBJECTIVES: Real-world evidence (RWE) studies frequently attribute patients to the healthcare organization (HCO) where they are diagnosed, even though diagnosis and treatment may occur at different institutions. This distinction may become increasingly important as oral systemic-therapies shift treatment into community-based settings. Using incident breast cancer, the most commonly diagnosed cancer among US women - as a case study, we quantified concordance between the attributed diagnosing and treating HCO and evaluated whether concordance differed by treatment route of administration (RoA).
METHODS: Incident breast cancer patients were identified from linked US EHR and open claims (>=2 C50.x diagnoses on distinct dates; index 2022-2024; 12-month treatment-free lookback; >=1 systemic-therapy post-index). The diagnosing HCO was taken from the index diagnosis record; the treating HCO from the provider on the earliest systemic-therapy. Both diagnosing and treating providers were mapped to parent HCOs, allowing assessment of whether diagnosis-based attribution reflects the organization delivering systemic-therapy; concordance defined as the same HCO for diagnosis and treatment. Patients stratified by RoA and class; analyses were descriptive and unadjusted.
RESULTS: Among 264,743 incident patients, 64.3% received systemic-therapy at a different HCO than where they were diagnosed; only 34.6% remained within the diagnosing organization. Same-HCO concordance was modestly higher for infused than oral first therapy (40.6% vs. 31.1%); conversely, 68.9% of oral-treated and 59.4% of infused-treated patients were treated at a different HCO.
CONCLUSIONS: Diagnosis and systemic treatment occurred within the same HCO for only about one-third of incident breast cancer patients. Concordance was associated with route of administration: infused therapies were more often delivered within the diagnosing HCO, oral therapies at a different one. This suggests that attribution relying on the diagnosing organization may misclassify the treating organization in most patients, increasingly so as care shifts toward community-based oral treatment. Future work should evaluate these findings using adjusted-analyses and health-system-level attribution.