Using Physicians’ Affiliations to Build Hospital Networks: Local Clustering and the COVID-19 Pandemic
摘要
The paper studies hospital networks where edges represent physicians’ affiliated with multiple hospitals. Using data from the universe of hospitals in the Brazilian healthcare system (n = 7,837) and their corresponding physicians (n = 623,680), we calculate local clustering using monthly variation from January 2016 to December 2023. We estimate network disruptions induced by the COVID-19 pandemic by comparing local clustering before and after the pandemic. The pandemic caused local connectivity disruptions in neighborhoods where triads are more likely formed under light physician flow. Heavy flow neighborhoods displayed resilience to the pandemic and connectivity was largely unaffected.