Effect of Community-based Opinion Leaders on Guideline Dissemination in Large-Scale Physician Networks

Despite continuous innovation and progress in the medical field, many treatments and interventions fail to translate to practice. Studies have shown that healthcare providers are slow to adopt new medical guidelines due to various reasons (slower knowledge dissemination, implementation challenges, etc.). Furthermore, there is a lack of computational modeling approaches to analyze and understand physician guideline adoption behaviors in real-world scenarios. Professional network characteristics and local opinion leaders play a vital role in dissemination and adoption of medical guidelines in physician communities. In this work, we provide a systematic approach to identify opinion leaders (OLs) based on physician community characteristics. The proposed approach will leverage our previous work in Culturally Infused Agent Based Modeling Framework that can capture physician decision-making and guideline adoption behavior in real-world settings. Using large physician datasets such as the Physician Compare and physician share datasets, we demonstrate the utility and scalability of our approach. By comparing with various strategies to select OLs, we show that our community-based OL detection method can capture the trade-off between increasing reach and rate of spread.