Effect of hospital referral networks on appropriateness of hospitalization decisions: An HLM analysis

The purpose of this study was to investigate how a hospital’s position in its collaborative network influences patient outcomes. To overcome the limitations of existing research, we applied social network analysis techniques to investigate the inter-organizational networks formed by patient transfers across hospitals (Iwashina et al., 2009; Mascia & Di Vincenzo, 2011; Mascia et al., 2012). This social network perspective allowed us to evaluate the relationship between healthcare networks and patient outcomes in a multifold manner. First, we identified networks as emerging from actual patient flows to conceptualize healthcare system variables in a way that can be replicated across domains and geographical areas, thus overcoming the problematic use of taxonomies. Second, this study measured the variance in hospitals’ network membership by observing the different structural positions of hospitals within their own networks of patient referrals. Consequently, by combining social network theory, strategic management perspectives, and health service research, we propose a multi-level model, which explores how the position within a referral network influences both organizational performance and, importantly, patient-level outcomes.