Title: The Diffusion of Health Care Fraud: A Network Analysis

Abstract:

Many studies have examined the diffusion of health care innovations but less is known about the diffusion of health care fraud. This webinar considers the diffusion of potentially fraudulent Medicare home health care billing in the United States during 2002-16, and the extent to which such diffusion might be explained by informal networks of units in the healthcare system. In the first part of the webinar, an overview of the construction of shared-patient networks from patient-physician encounter data will be provided. Then, in response to an a priori hypothesis about mechanisms enabling rapid diffusion of fraudulent strategies, a novel bipartite mixture (or BMIX) network index for patient-sharing in networks across home health care agencies will be developed. The HRR-level BMIX will be shown to be a strong predictor of above-average home care expenditures across HRRs, including in regions when expenditures have increased the most, and that it was much more predictive than any other network measure. Within HRRs analyses show that agencies sharing more patients with other agencies were associated with increases spending the following year, HRR-level analyses show that the BMIX index was also strongly associated with longer-term subsequent changes in home health billing, and additional analyses rule out reverse causation. These results highlight the importance of bipartite network structure in diffusion and in infection models more generally.

Acknowledgements: Joint work with Jon Skinner and Tom Bulbolz

Bio of the speaker:

James O’Malley, Ph.D., is the Peggy Y. Thomson Professor of the Evaluative Clinical Sciences in The Dartmouth Institute of Health Policy and Clinical Practice, a Professor of Biomedical Data Science at the Geisel School of Medicine at Dartmouth, and an Adjunct Professor in the Departments of Computer Science and Mathematics at Dartmouth College. He is a statistician with methodological interests encompassing statistical inference for social networks, comparative effectiveness research including causal inference for both randomized and observational studies, multivariate hierarchical models, and Bayesian statistics. His subject matter research interests include the relationship between health and social networks, evaluation/estimation of variations in health quality and outcomes, vascular surgery, cardiology, and evaluation of medical devices. He has published over 220 peer-reviewed research papers and given numerous invited talks at conferences and academic institutions. He was Director of the Program in Quantitative Biomedical Sciences (QBS) from 2018 - 2020 and currently is Co-Director of the Ph.D. Program in Health Policy and Clinical Practice. In 2011 he received the Mid-career Excellence award from the Health Policy Section of the ASA, in 2012 was elected to be a fellow of the ASA, and in 2019 the ISPOR (International Society of Pharmacoeconomics and Outcomes Research) Health Economics and Outcomes Research Excellence in Methodology Award.

** Registrants will receive a zoom link one day before the webinar.