Abstracts

Social network analysis to measure epilepsy coordination of care—a mixed methods study

Abstract number : 2.398
Submission category : 13. Health Services (Delivery of Care, Access to Care, Health Care Models)
Year : 2017
Submission ID : 349495
Source : www.aesnet.org
Presentation date : 12/3/2017 3:07:12 PM
Published date : Nov 20, 2017, 11:02 AM

Authors :
Hamada Hamid. Altalib, Yale University; Holly Lanham, University of Texas Health Science Center at San Antonio; Katharine McMillan, University of Texas Health Science Center at San Antonio; Kei Cheung, VA Connecticut Healthcare System; Brenda Fenton, VA C

Rationale: The study aims to explore the strengths and weakness of Social Network Analysis (SNA), using the number of patients shared among physicians, to measure coordination of care in a large healthcare system. SNA is particularly valuable for quantifying physician connectivity in large administrative healthcare databases. In contrast, Relational Coordination (RC) measures quality of communication and coordination at a micro-system level.  Methods: In a parallel mixed-methods approach, we correlated SNA and RC scores of 57 provider teams caring for Veterans with epilepsy, within the Veterans Health Administration (VA), nationwide. Key informant interviews of 80 epilepsy care team members were conducted concurrently to describe the quality of epilepsy care coordinating in the VA healthcare system.  Results: RC scores ranged from 2.43-5.00 (with the maximum score of 5). RC scores correlated well with SNA scores, specifically neurologists’ average node degree, with a Spearman correlation = 0.356 (p-value < 0.05).  Conclusions: Key informants described the mechanisms of developing and sustaining professional networks as well as the limitations of both SNA and RC measures in capturing care coordination. The VA healthcare system has established many services based on the hub and spoke model to improve coordination of care across facilities. SNA may be a powerful tool to measure and quantify the effectiveness of these models.  Funding: U.S. Department of Veterans Affairs (IIR-062) Pugh (PI)
Health Services