The Adoption of Electronic Medical Records by U.S. Hospitals: An Exploration of Network Methods and Models A Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at George Mason University by Yinyue Hu Master of Arts Georgetown University, 2010 Bachelor of Arts University of Shanghai for Science and Technology, 2008 Director: Laurie Schintler, Associate Professor School of Policy, Government and International Affairs Spring Semester 2016 George Mason University Fairfax, VA This work is licensed under a creative commons attribution-noderivs 3.0 unported license. ii DEDICATION This dissertation is dedicated to my grandfather, YIN Guangqi, who had always inspired me to never stop learning. iii ACKNOWLEDGEMENTS First and foremost, I would like to express my deep gratitude and appreciation to my committee members, without whose persistent help this dissertation would not have been possible. I owe a debt of gratitude to my committee chair Professor Laurie Schintler for her guidance and extreme patience from the very early stage of my doctoral study through the completion of this dissertation. I would like to thank my committee member Professor Robert Axtell, who introduced me to agent based modeling and complexity theories, for his friendly guidance and thought-provoking suggestions. My earnest thank you goes to my committee member and supervisor for nearly six years, Professor Arnauld Nicogossian, for his mentorship throughout my doctoral study. I am beyond grateful to have had the opportunity to work with him. His continued direction and support is what makes me survive the doctoral study. He has taught me what a true researcher should be in innumerable ways. Also, I would like to thank my external reader Professor Azhar Rafiq from Virginia Commonwealth University for offering his help to review my dissertation. I would also like to recognize Professors Naoru Koizumi, Phil Auerswald and Ray Sommer for their input during my years of study at George Mason University. I would like to thank my friends and colleagues at George Mason University for their feedback, support and iv friendship. Also, a thank you to Professor Charles Doarn from University of Cincinnati for all the constructive discussions throughout the years. This dissertation would not be possible without the Dissertation Completion Grant from the Office of the Provost at GMU during my last semester of the study. In a similar vein, I would like to thank the School of Policy, Government and International Affairs for the financial support granted at various stages of this dissertation. I would like to acknowledge the HIMSS Foundation for granting me the access to the research data. I would like to thank my parents for continuously encouraging me in all my pursuits. They always put education at the first priority and have faith in me. Finally, I would like to thank my husband, Kai He, who supported me in every possible way to see the completion of this work. Thank you for being such a great life and intellectual partner. v TABLE OF CONTENTS Page List of Tables ................................................................................................................... viii List of Figures .................................................................................................................... ix List of Abbreviations .......................................................................................................... x Abstract .............................................................................................................................. xi Chapter 1 INTRODUCTION .............................................................................................. 1 1.1 ELECTRONIC MEDICAL RECORDS ................................................................... 2 1.2 RESEARCH QUESTIONS ....................................................................................... 5 1.3 RESEARCH DATA AND METHODOLOGY ........................................................ 8 1.4 ROADMAP OF THE DISSERTATION ................................................................ 12 Chapter 2 LITERATURE REVIEW ................................................................................. 13 2.1 EMPIRICAL STUDIES ON EMR ADOPTION .................................................... 13 2.2 NETWORK ANALYSIS ........................................................................................ 16 2.2.1 Network Measures and Models ........................................................................ 16 2.2.2 Network Analysis in Health and Medical Studies ............................................ 24 2.3 DIFFUSION OF INNOVATION ........................................................................... 26 2.4 AGENT BASED MODELING ............................................................................... 31 2.4.1 Agent Based Models of the Diffusion of Innovation ....................................... 31 2.4.2 Implications from Agent Based Models ........................................................... 33 2.5 SUMMARY ............................................................................................................ 39 Chapter 3 NETWORK ANALYSIS ................................................................................. 41 3.1 METHODOLOGY .................................................................................................. 42 3.1.1 Spatial Clustering ............................................................................................. 42 3.1.3 Network Construction....................................................................................... 46 3.1.3 Network Analysis ............................................................................................. 47 3.2 RESULTS................................................................................................................ 48 3.3 SUMMARY ............................................................................................................ 61 vi Chapter 4 EVENT HISTORY MODEL ........................................................................... 63 4.1 METHODOLOGY .................................................................................................. 64 4.1.1 Cox Proportional Hazard Model....................................................................... 64 4.1.2 Model and Variable Specification .................................................................... 65 4.2 RESULTS................................................................................................................ 68 4.2.1 Descriptive Statistics ........................................................................................ 68 4.2.2 Preliminary Regression Analysis...................................................................... 75 4.2.3 Cox Proportional Hazard Model....................................................................... 79 4.3 SUMMARY ............................................................................................................ 86 Chapter 5 AGENT BASED MODELING ........................................................................ 87 5.1 METHODOLOGY .................................................................................................. 89 5.1.1 Model ................................................................................................................ 89 5.1.2 Parameter Setting .............................................................................................. 90 5.2 RESULTS................................................................................................................ 92 5.2.1 Baseline Model ................................................................................................. 93 5.2.2 Individual Interventions .................................................................................... 95 5.2.2 Alteration Interventions .................................................................................. 101 5.3 SUMMARY .......................................................................................................... 108 Chapter 6 CONCLUSION .............................................................................................. 110 6.1 SUMMARY OF FINDINGS ................................................................................ 111 6.2 LIMITATIONS OF THIS STUDY ....................................................................... 113 6.3 POLICY IMPLICATIONS ................................................................................... 115 6.4 IMPLICATIONS FOR FUTURE RESEARCH ................................................... 117 Appendix ......................................................................................................................... 119 References ....................................................................................................................... 138 vii LIST OF TABLES Table Page Table 1-1 HIMSS EMR Adoption Model ........................................................................... 3 Table 1-2 Summary of Data .............................................................................................. 10 Table 2-1 Network Measures and Methods ...................................................................... 18 Table 3-1 Number of Hospitals Which Underwent Management or Structural Changes 42 Table 3-2 Sample Result of DBSCAN Sensitivity Test (2005 data) ................................ 45 Table 3-3 Summary of Network Data ............................................................................... 52 Table 3-4 Summary Statistics of Simulated Random Network and Real Network .......... 61 Table 4-1 Basic EMR........................................................................................................ 69 Table 4-2 Summary Statistics of Number of Beds ........................................................... 73 Table 4-3 Multicollinearity Test ....................................................................................... 76 Table 4-4 Logistic Regression Results for Hospital EMR Adoption ............................... 78 Table 4-5 Cox Model for Hospital EMR Adoption .......................................................... 81 Table 4-6 Cox Model for Hospital EMR Adoption, Reduced Dummies ......................... 85 Table 5-1 Parameter Setting.............................................................................................. 92 Table 5-2 Network Properties, Real vs. Ideal ................................................................. 106 viii LIST OF FIGURES Figure Page Figure 1-1 Summary of EMR Adoption Score of U.S. hospitals, 2008-2015. ................... 4 Figure 1-2 Flowchart of Methodology .............................................................................. 12 Figure 3-1 Illustration of DBSCAN (Ester et al. 1996) .................................................... 44 Figure 3-2 Illustration of Network Construction .............................................................. 47 Figure 3-3 Illustration of Levels of Adjacent Vertices ..................................................... 48 Figure 3-4 Sample Node Subgraphs, 1.5 Levels of Adjacent Vertices ............................ 50 Figure 3-5 Sample Node Subgraphs, 2 Levels of Adjacent Vertices ............................... 51 Figure 3-6 Network Visualization, 2005 .......................................................................... 53 Figure 3-7 Network Visualization, 2013 .......................................................................... 54 Figure 3-8 Distribution of Top 200 Hospitals by Degree Centrality, 2005 ...................... 55 Figure 3-9 Distribution of Top 200 Hospitals by Degree Centrality, 2013 ...................... 56 Figure 3-10 Histogram of Degree Centrality .................................................................... 58 Figure 3-11 Degree Distribution ....................................................................................... 59 Figure 3-12 Degree Distribution with Power Law Fit, 2013 ............................................ 60 Figure 4-1 Summary of EMR Adoption, 2005-2013 ........................................................ 70 Figure 4-2 Distribution of Hospital by Specialty and Category ....................................... 71 Figure 4-3 Hospital Ownership Status .............................................................................. 72 Figure 4-4 Number of Beds, Boxplot ............................................................................... 73 Figure 4-5 Degree Centrality, Boxplot ............................................................................. 74 Figure 4-6 Direct Network Exposure, Barplot .................................................................. 75 Figure 4-7 Estimated Survival Function ........................................................................... 83 Figure 5-1 HIT Legislations and Policy Interventions ..................................................... 88 Figure 5-2 Baseline Model, Scale Free Network .............................................................. 94 Figure 5-3 Baseline Model, Real Network ....................................................................... 94 Figure 5-4 Individual Intervention, Real Network ........................................................... 97 Figure 5-5 Individual Intervention, Scale Free Network .................................................. 98 Figure 5-6 Final Uptake of Individual Interventions, Real Network ................................ 99 Figure 5-7 Final Uptake of Individual Interventions, Scale Free Network .................... 100 Figure 5-8 Uptake Comparison, Individual Interventions .............................................. 101 Figure 5-9 Alteration Interventions, Real Network ........................................................ 103 Figure 5-10 Alteration Interventions, Scale Free Network ............................................. 104 Figure 5-11 Final Uptake of Alteration Interventions, Real Network ............................ 105 Figure 5-12 Final Uptake of Alteration Interventions, Scale Free Network................... 105 Figure 5-13 Uptake Comparison, Alteration Interventions ............................................ 107 ix LIST OF ABBREVIATIONS Electronic Medical Records .......................................................................................... EMR Agent Based Modeling ................................................................................................ ABM Health Information Technology ...................................................................................... HIT Healthcare Information and Management Systems Society ..................................... HIMSS American Recovery and Reinvestment Act ............................................................... ARRA Health Information Technology for Economic and Clinical Health ....................... HITECH Clinical Data Repository ............................................................................................... CDR Nursing Documentation ................................................................................................ DOC Electronic Medication Administration Records......................................................... EMAR Clinical Decision Support .............................................................................................. CDS Computerized Physician Order Entry ......................................................................... CPOE Full-time Equivalent ...................................................................................................... FTE American Hospital Association .................................................................................... AHA Distance-based Spatial Clustering of Application with Noise .............................. DBSCAN x