Abstract
IMPORTANCE: Although patients with brain metastases receive interdisciplinary and multi-institutional care, the association between neuro-oncologic care networks and patient outcomes remains unknown. As patients often interact with multiple facilities, quantifying this association across a network of hospitals is critical to capture the complexity of the health care journey for patients with brain metastases. OBJECTIVE: To evaluate how statewide health care network metrics are associated with inpatient mortality and hospital length of stay (LOS) for patients with brain metastases. DESIGN, SETTING, AND PARTICIPANTS: This multicenter, statewide cohort study used data from the 2018 to 2019 Healthcare Cost and Utilization Project State Inpatient and Emergency Department Databases. Primary analyses were completed by August 2023. Participants included adults with a brain metastases receiving care in Massachusetts. EXPOSURE: All inpatient and emergency department visits mapped for patients following the first diagnosis of brain metastasis. MAIN OUTCOMES AND MEASURES: Inpatient mortality and hospital LOS were the main outcomes assessed. Hospital interdependence in brain metastases care was calculated using a connectedness score (weighted degree: weighted sum of ties to other care facilities). The association between hospital connectedness and clinical outcomes was analyzed using mixed-effects logistic and linear regression models, adjusting for hospital-level features. RESULTS: In this cohort study, 4679 patients with brain metastases were identified with inpatient or ED encounters in Massachusetts (from 2018 to 2019). The median (IQR) age was 64 (57-73) years, and 2559 (55%) were female. There was interdependence in brain metastases care, with 993 patients (21%) visiting 2 or more unique hospitals. Highly connected hospitals were heterogeneous, with many being small and one-half lacking subspecialty neuro-oncologic care or teaching status. Increased hospital connectedness was significantly associated with improved inpatient mortality for patients with brain metastases, with the lowest connectedness quartile associated with more than double the risk of mortality compared with the highest quartile (odds ratio, 2.34; 95% CI, 1.33-4.11; P = .003). A stepwise increase in inpatient mortality risk was observed as hospital connectedness decreased, independently of hospital volume. Furthermore, intermediate hospital connectedness was associated with increased hospital LOS (coefficient, 1.08; 95% CI, 0.17-1.95; P = .006). CONCLUSIONS AND RELEVANCE: This study found that hospital-to-hospital interconnectedness was significantly associated with improved clinical outcomes for patients with brain metastases. The salience of network metrics highlights their potential role alongside other patient-level and hospital-level variables to evaluate and improve oncology care delivery.