Project Overview and Description

CHOT Spring Meeting April 16 & 17, 2015 April 16 8:30-9:00 Breakfast buffet 9:00-9:30 Welcome: Eva K. Lee LIFE Forms Review and NSF Update: Craig Scott State of the Center: Bita A. Kash Meeting Overview: Lesley Tomaszewski 9:30-9:45 Presentation – Florida Atlantic University Ankur Agarwal, Ravi Behara, & Gulcin Gumus 9:45-10:15 Patient-Centered Care Cluster Chair: Eva K. Lee, Georgia Institute of Technology 1.1 Improving Health Promotion: Leveraging Statistical Learning and Electronic Medical Records for Healthcare Market Segmentation 1.2 Understanding the “White Space” of Where Patients Go After They Leave the Hospital 1.3 User Perception and Use of Technology and the Association of Technology with Hospital Readmissions and ER Visits in Home Health Care 1.4 Healthcare system Optimization: advancing delivery timeliness, quality and effectiveness 1.5 Personalized Medicine 10:15-10:45 Industry Member Discussant Panel + LIFE Forms: Jamey Gigliotti (IAB chair) 10:45-11:00 Coffee Break 11:00-11:30 Quality and Safety Cluster Chair: Bita A. Kash, Texas A&M University 2.1 Improvements to Root Cause Analysis of Patient Safety Events 2.2 Examining How Lean Six Sigma Processes Reduce Hospital-Acquired Conditions 2.3 Does the Casework Design Affect the Patient Room Cleanliness and Healthcare-Associated Infection (HAI) Rates? 11:30-11:45 Industry Member Discussant Panel + LIFE Forms: Jamey Gigliotti (IAB chair) 11:45-12:00 Business Canvas Model – Overview and Feedback 12:00-1:15 Lunch 1:15:1:30 Business Canvas Model - Discussion 1:30-2:00 Enabling HIT and Care Coordination Cluster Chair: Rob Weech-Maldonado, University of Alabama at Birmingham 3.1 An Immersive Virtual Reality Approach for Real-Time, Scalable Learning in Healthcare 3.2 Assessment of a Tele health Device in Promoting Heart Failure Patient Engagement and Self-Care in Rural Areas 3.3 Investigating the Impacts of a Patient’s Social Network in Achieving Gamification Solutions in Personalized Wellness Management 3.4 Reducing Readmission after Hip Surgery using Statistical Process Control and Smart Home Care 2:00-2:15 Industry Member Discussant Panel + LIFE Forms: Jamey Gigliotti (IAB chair) 2:15-3:00 Access & Efficiency Cluster Chair: Deirdre McCaughey, Pennsylvania State University 4.1 Ebola epidemic regional and facility response models 4.2 Identifying and Utilizing Inexpensive Technologies to Manage Patient Populations 4.3 Robust healthcare staff scheduling 4.4 Challenges in telemedicine – a systematic review and engagement with rural communities 3:00-3:15 Industry Member Discussant Panel + LIFE Forms: Jamey Gigliotti (IAB chair) 3:15-3:30 Break with light refreshments 3:30-4:15 Macro/Policy Cluster Chair: Jim Benneyan, Northeastern University 5.1 Understanding Group Practice Trends in 2015 and into the Future 5.2 Surgical Care Trends and the Future Role of Hospitals 5.3 Modeling ACOs as macro systems of care 5.4 Patient Flow in Children’s Hospitals: Research-Informed Strategies to Influence Discharge Time and Capacity 5.5 Translating UBRICA’s Vision for Kenya to Evidence-based Strategy and Funding 5.6 Hospital acquired conditions - systematic and adaptive approach 4:15-4:30 Industry Member Discussant Panel + LIFE Forms: Jamey Gigliotti (IAB chair) 4:30-4:45 Debrief – End of Day One 4:45-6:00 Networking Reception April 17 8:30-9:00 Breakfast buffet 9:00-9:15 Day 2 Introductions 9:15-9:45 Collaborative Research Proposals Chair: Harriet B. Nembhard, Pennsylvania State University 6.1 Replicating a Study of the Efficacy of Quality Improvement Processes in Reducing Hospital Acquired Conditions 6.2 Technology Trends and Smart Interventions to Mitigating Patient Risk at Critical Transitions for Total Joint Arthroplasty (TJA) 6.3 Social Network Analysis: Examining Interactions among Providers at the Network Level 6.4 Choosing wisely and reducing practice variation 9:45-10:00 Industry Member Discussant Panel + LIFE Forms: Jamey Gigliotti (IAB chair) 10:00-10:30 Coffee Break 10:30-12:00 IAB Members: IAB Business Meeting Non-IAB member/guests: Q&A with Lesley Tomaszewski Graduate students: Graduate student business meeting 12:00 PM Adjourn Box lunch ready for pick-up CHOT IAB Spring Meeting 2015 List of attendees First Name Last Name Email Address Position Organization Ankur Agarwal ankur@cse.fau.edu Associate Professor Florida Atlantic University Ali Almusawi ali.n.mosawi@gatech.edu Student Georgia Institute of Technology Osama Alotaik oaa127@psu.edu Graduate Student Pennsylvania State University Lavannya Atri latri3@gatech.edu Undergraduate Georgia Institute of Technology Nathaniel Bastian ndbastian@psu.edu Ph.D. Candidate Pennsylvania State University Ravi Behara rbehara@fau.edu Faculty Florida Atlantic University James Benneyan benneyan@coe.neu.edu CHOT Site Director Northeastern University Nancy Borkowski nborkows@uab.edu Professor University of Alabama at Birmingham Chris Brainard cbrainard@uabmc.edu Director University of Alabama at Birmingham Health System Dominic Breuer dbreuer@coe.neu.edu Graduate student Northeastern University Barbara Brown bbrown@gaanes.com HR Manager Northside Anesthesiology Consultants Darrell Burke deburke@uab.edu Associate Professor University of Alabama at Birmingham Yu Cao ycao98@gatech.edu Graduate Student Georgia Institute of Technology Kayla Cline kcline@sph.tamhsc.edu Graduate Student Texas A&M University Mary Ellen DeBardeleben mary.debardeleben@healthsouth. com Associate Director, Quality HealthSouth James Gigliotti james.gigliotti@highmark.com Director Highmark Abigail Gonzalez abigail.gonzalez@sph.tamhsc.edu REU Texas A&M University Ajay Karthic Gopinath Bharathi abg167@psu.edu Graduate Student Pennsylvania State University Gulcin Gumus ggumus@fau.edu Assistant Professor Florida Atlantic University Nick Halzack n.halzack@asahq.org Health Policy Research Analyst American Society of Anesthesiologists James Hury jfhury@texaschildrens.org Director of Finance Texas Children's Hospital Zulqar Islam zulqarislam@gatech.edu Undergraduate Georgia Institute of Technology Jean Kang jkang70@gatech.edu Undergraduate Georgia Institute of Technology Michael Karuu michael.karuu@ubrica.com VP Ubrica Bita Kash bakash@sph.tamhsc.edu CHOT Director Texas A&M University Lisa Korman lgk117@psu.edu Assistant Director, CIHDS Pennsylvania State University Arthur Lambert a.lambert@neu.edu Adjunct Faculty Northeastern University Amy Landry akyarb@uab.edu Faculty University of Alabama at Birmingham Eva Lee eva.lee@gatech.edu CHOT Site Director Georgia Institute of Technology Jinha Lee jlee68@mail.gatech.edu Graduate student & investigator Georgia Institute of Technology Cynthia LeRouge lerouge@uw.edu CHOT Site Director University of Washington Jennifer Lingenfelter jennifer.lingenfelter@piedmont.or g Executive Director Piedmont Healthcare Yifan Liu yifanliu@gatech.edu Graduate student & investigator Georgia Institute of Technology Kathy Lomaskin katherine.lomaskin@siemens.com Program Director Siemens Linlin Ma lxm1009@psu.edu PhD student Pennsylvania State University Christina Mastrangelo mastr@uw.edu Professor University of Washington CHOT IAB Spring Meeting 2015 List of attendees First Name Last Name Email Address Position Organization Deirdre McCaughey mccaughey@psu.edu Associate Professor Pennsylvania State University Lauren McManemin Lim5137@psu.edu Research Assistant Pennsylvania State University Amanda Mewborn amanda.mewborn@piedmont.org Executive Director Piedmont Healthcare Hande Musdal h.musdal@neu.edu Post-doc Research Associate Northeastern University Harriet Nembhard hbn2@psu.edu CHOT Site Director Pennsylvania State University Andrew Norton allank@mlhs.org Chief Medical Officer Main Line Health Jihwan Oh Jihwan.Oh@gatech.edu Undergraduate Georgia Institute of Technology Michael O'Toole michael.otoole@piedmont.org Executive Director Piedmont Healthcare Sang Wook Park sangwpark90@gmail.com Undergraduate Georgia Institute of Technology Peter Preziosi peter.preziosi@verizon.com Managing Principal Verizon Midge Ray midgeray@uab.edu Associate Professor University of Alabama at Birmingham Craig Scott scottcs@u.washington.edu Center Evaluator NSF Prabhu Shankar prshank@emory.edu Asst. Professor Emory University Andriy Shapoval ashapoval3@gatech.edu Postdoctoral Fellow Georgia Institute of Technology Harold Simon hsimon@emory.edu Professor and Vice Chair Department of Pediatrics Children's Healthcare of Atlanta/Emory University Abhinav Singh aus370@psu.edu Student Pennsylvania State University Mary Ellen Skeens maryellen.skeens@philips.com Director of Solutions Consulting Philips Wes Smith wsmith@aqaf.com CEO Alabama Quality Assurance Foundation Cory Stasko corystasko@gmail.com Graduate student Northeastern University Eric Swenson ers187@psu.edu Student Pennsylvania State University Prashant Tailor ptailor3@gatech.edu Undergraduate Researcher Georgia Institute of Technology Debra Tan dtan@tamhsc.edu Graduate Research Assistant Texas A&M University Pavan Thaker pavan.thaker@gmail.com Graduate Student Georgia Institute of Technology Haozheng Tian tianhzh@gatech.edu Graduate Student & Investigator Georgia Institute of Technology Lesley Tomaszewski tomaszewski@sph.tamu.edu Managing Director Texas A&M University Conrad Tucker ctucker4@psu.edu Assistant Professor Pennsylvania State University Karan Uppal kuppal3gt@gmail.com Informaticist Morehouse School of Medicine Samuel Wachira wachisam@yahoo.co.uk Administrator Ubrica Yuanbo Wang wangc@mail.gatech.edu Graduate Student & Investigator Georgia Institute of Technology Zixing Wang zwang411@gatech.edu Graduate student Georgia Institute of Technology Robert WeechMaldonado rweech@uab.edu CHOT Site Director University of Alabama at Birmingham Xin Wei xwei36@gatech.edu Graduate Student & investigator Georgia Institute of Technology Alacare Home Health and Hospice Samika Williams samika.williams@alacare.com VP Strategic Network Partnerships Hank Williams hwilliams33@gatech.edu Undergraduate (GT/8901) Georgia Institute of Technology Jade Wronowski jmwronowski@gmail.com Research assistant Pennsylvania State University Yifeng Yu yiy5058@psu.edu PhD Student with Proposal Pennsylvania State University Ferhat Zengul ferhat@uab.edu Assistant Professor University of Alabama at Birmingham NSF Center for Health Organization Transformation (CHOT) Vision www.chotnsf.org State of the Center CHOT IAB Spring Meeting April 16-17, 2015 Bita A. Kash, PhD, MBA, FACHE Center Director | Associate Professor | Texas A&M University CHOT History 2007 2008 2010 2011 2012 2013 2014 2015 • NSF Planning Grant awarded to Texas A&M University CHOT: NSF IUCRC model http://www.nsf.gov/eng/iip/iucrc/iucrc_video.jsp • February: CHOT hosts inaugural IAB Meeting in Houston, TX • August: CHOT receives NSF Award, Texas A&M University to serve as lead site, Georgia Tech as first University Partner • Northeastern University joins as CHOT university partner • Penn State University joins CHOT • Rush University and University of Michigan are approved by IAB to apply for planning grants to join Center (not awarded planning grant) • April: University of Alabama-Birmingham and University of Washington are approved by IAB to apply for planning grants to join Center • April: Texas A&M and Georgia Tech are awarded Phase II and UAB is awarded planning grant • August: Texas A&M awarded IMD Award • April: University of Washington awarded planning grant Future CHOT Sites NSF Planning Grant awarded Spring 2015: Potential new U.S. CHOT site for IAB vote Spring 2015: Potential international CHOT site for IAB vote Fall 2015: CHOT Accomplishments • 19 students engaged in CHOT projects • 18 publications in peer-reviewed journals • CHOT project named 2014 INFORMS Franz Edelman finalist, 2nd prize in INFORMS Wagner Excellence in Practice • One high impact publication in the Milbank Quarterly (ranked #1 in Health Policy & Services) • CHOT research featured in International Innovation “A Model to Healthcare”, Becker’s Hospital Review, NewsMedical, and EndoNurse • CHOT Quarterly Newsletter Spring 2015 Webinars IAB Defines CHOT Research Agenda Available to Members Only at www.CHOTNSF.org DATE TITLE February 11 International Comparison of Preoperative Testing and Assessment Protocols and Best Practices to Reduce Surgical Care Costs: A Systematic Literature Review March 4 The Pediatric Medical Home: Results From A Systematic Literature Review April 1 Implementing a Medical Screening Tool for Rural Hospital Emergency Departments April 22 The Identification and Management of Information Problems During Morning Rounds April 29 Results from the Implementation of a Medical Screening Tool for Rural Hospital Emergency Departments May 6 Optimizing Hospital Safety Culture CHOT Site Texas A&M University Request for Proposals (RFPs): Texas A&M University Texas A&M University The Pennsylvania State University Texas A&M University The Pennsylvania State University • • • • • Remote Health & Tele-health Organization & System Design Transitions of Care Human Technology Patient Behavior & Self-care RFP Process Implemented Fall 2014 RFP: Organization & System Design RFP: Remote Health & Tele-health Purpose/Topics Addressed: • Identify the most appropriate care at the most appropriate place • Getting patient care that’s needed at the right time • Developing novel technologies to facilitate community • Barriers include: Interstate variability; scalability for adoption; infrastructure; reimbursement; cost and usability for clinicians and patients; policy compliance/privacy concerns (e.g., HIPPA compliance and stringent privacy guidelines may stand in the way of building simple/convenient telehealth applications) Eligibility Requirements • Demonstrate what is needed to prove out novel remote health technologies • Leverage national scale, not local Deliverables • Framework to understand variability • Framework to show what could be implemented to facilitate telehealth Publications Lee, E.K., Mejia, A.F., Senior, T., & Jose, J. (2010). Improving patient safety through medical alert management: an automated decision tool to reduce alert fatigue. Paper presented at the American Medical Informatics Association Symposium, Washington D.C. Bennett-Millburn, A., Griffin, P., Hewitt, M., & Savelsbergn, M. The value of remote monitoring systems for treatment of chronic disease. To appear in IIE Transactions on Healthcare Systems Engineering, 2014. CHOT CONFIDENTIAL Purpose/Topics Addressed: • Effective patient placement • Effective models of care delivery and coordination • Macro-level strategic planning • Successful change implementation and organizational capacity for change • Care process design • Facility Design • International studies, patient surveys • ED/OR throughput processes & systems Eligibility Requirements • Data availability assured by industry partner prior to study • Pre-existing knowledge of data, what it is, form, utilization capacity etc. • Access to data manager/processor • Identify if project is a “data/outcomes” project or a “process/system” study. Deliverables • Pilot studies to set ground for two-phase research work: • Phase 1 being a quantitative data analysis project • Phase 2 being qualitative type deep dive into findings from phase 1 Publications Kash, B.A., Spaulding A., Johnson, C.E., and Gamm, L.D. (2014). “Success Factors for Strategic Change Initiatives: A Qualitative Study of Healthcare Administrators' Perspectives,” Journal of Healthcare Management 59(1):65-82. Lee, E.K., Yuan, F., Zhou, R.L., Lahlou, S., Post, E., Wright, M., Atallah, H., Haley, L.L. Modeling and optimizing emergency department workflow of large urban public hospital. To appear in Interfaces, 2014. CHOT CONFIDENTIAL RFP: Transitions of Care Purpose/Topics Addressed: • To study the processes, procedures, and best practices of care coordination models currently used in health care settings to identify key factors and characteristics of successful transitions of care. • Identify key points from both the existing literature and existing care transition and coordination models • Identify success factors and barriers (reimbursement, clinical, social, etc.) using a mixed method approach • Multiple care settings/levels of care Eligibility Requirements • Each university partner selects a health care setting via IAB members pertaining to the transition of postoperative care for hip or knee surgical patients (i.e., joint replacement) Deliverables • Survey of current best practices, success/barrier analysis across health care setting (white paper) • Cross-university initiative (survey design, implementation, post-hoc analyses) RFP: Human-Technology Interaction Purpose/Topics Addressed: • Barriers to the use of technology • How to better leverage technology • How to better manage technology evolution • Technology barriers evidenced across the healthcare ecosystem • How technology impacts the workflow • Use of technology for obtaining patient information prior to the patient-clinician encounter Eligibility Requirements • Study the use of technology (tablets and smart phones) in a clinical setting to evaluate improvement of clinical outcomes and patient and clinical satisfaction • Impact on clinicians on the use of new technology in the clinical workflow – the right balance on usability Deliverables • A comprehensive study plan • Clinicians’ perspectives on emerging technology – relevant information associated with the symptoms of the patient • Patients’ perspectives on emerging technology Publications Kash, B.A., Zhang, Y., Cline, K., Menser, T., Miller, T.R. (2014). “The Perioperative Surgical Home (PSH): A Comprehensive Review of U.S. and NonU.S. Studies Shows Predominantly Positive Quality and Cost Outcomes,” The Milbank Quarterly 92(4). Gregory, S.T., Tan, D. Tilrico, M., Edwardson, N., & Gamm, L. Bedside shift report: A systematic literature review. Journal of Nursing Administration, 44(10), 541-545. CHOT CONFIDENTIAL Center for Health Organization Tansformation Publications Lee, E.K. & Cha, K. (Mar, 2010). Automated data collection and integration for cancer treatment design and clinical quality evaluation investigations. 2010 AMIA Summit on Clinical Research Informatics, San Francisco, CA. Musdal, H., Shiner, B., Chen, T., Ceyhan, M.E., Waatts, B.V., & Benneyan, J. (2014).In-person and videobased post-traumatic stress disorder treatment for veterans: A location-allocation model. Military Medicine, 179(2), 150-156. CHOT CONFIDENTIAL RFP: Patient Behavior & Self-Care Fall CHOT IAB Meeting Purpose/Topics Addressed: • Identify the “change in the slope” associated with the following self-care behaviors: - Health literacy - Health promotion - Chronic disease management • For segmentation strategies – quantified self and predictive analytics • For persistent engagement – gamification • Habit loops – starting new habits, replacing old habits with new ones, stopping old habits • How to innovate motivation Research project Presentation IAB members share research ideas and questions Eligibility Requirements • How to innovate technology to create persistent engagement over time • How do we get people to engage in health promotion, chronic disease management, and health literacy • We need to collect insights from consumers in multiple places -> settings/institutions • Understanding market segmentation • Literature review, existing CHOT facilities doing this work • Understand neuroscience dynamics of habit formation Following the Spring Meeting CHOT sites conduct research projects Deliverables • Define preferences for engagement tools informed by data analytics, literature review, patient groups, and other qualitative market research that will need to be tested on market segments in year 2. Annual Research Cycle Spring CHOT IAB Meeting Publications Present Research Proposals Kraschnewski, J.L., Sciamanna, C., Stuckey, H.L., Chuang, C.H., Lehman, E.B., Hwang, K.O., Sherwood, L.L., & Nembhard, H.B. (2013). A silent response to the obesity epidemic: Decline in US physician weight counseling. Medical Care, 51(2), 186-192. Vest, J. R., Gamm, L. D., Oxford, B. A., Gonzalez, M. I., & Slawson, K. M. (2010).Determinants of preventable readmissions in the United States: a systematic review. Implementation Science, 5(1), 1-27. IAB provides feedback & ranks research proposals CHOT CONFIDENTIAL Center Finances Each industry member contributes $50k per year to its respective University Partner - 10% overhead fixed Members sign a 3-year contract - Can be cancelled after one year with 90-day notice NSF contributes $55k to each University Partner - Texas A&M receives an additional $20k as lead site TL5 - All NSF funding is reduced by full university overhead Slide 15 TL5 Do we need to have this here? Is there another way we can say this? Tomaszewski, Lesley, 4/7/2015 Following the Fall meeting Research proposals developed with IAB input CHOT sites facilitate collaborative research project CHOT Industry Advisory Board (IAB) Industry Members delegate representatives to serve on the CHOT IAB. IAB contributes to the Center’s strategic direction and advises on projects, new university partners, industry members, and project voting and selection. IAB meets twice per year to conduct CHOT business. IAB members are invited to participate in any additional meetings and research webinars during the year. IAB elects Chairperson for two year term - James Gigliotti was elected IAB Chair in the fall of 2013 Phase II Direction 1. Further organization of CHOT operations and the development of a suitable project management approach 2. Focused growth based on defined primary customer segments and corresponding value propositions by segment Opportunities for Engagement Opportunities for Value Creation Opportunities for Improvement 3. A system designed to facilitate high-impact multi-disciplinary research across university partners and industry members Spring meeting attendance 74 Total Registered Attendees – As of April 10 Industry (guest), 10 NSF, 1 Faculty, 19 Industry (member), 10 Staff, 3 Undergraduate students, 8 Graduate students, 23 Review of Agenda Contribute greatly and enjoy immensely so that we all may learn lots! Feel free to call Bita: 979‐575‐6768 Business Model Canvas Key Partners University CHOT Sites CHOT Co-directors Faculty Graduate Students Industry Members National Science Foundation Key Activities  Implementation of research     studies Application/dissemination of research findings Member retention/client relations External/supplemental grant writing Collaboration among universities and industry members Key Resources Human Resources  Experienced/well known researchers  Graduate students  Undergraduate students Financial Resources  Administrative infrastructure  NSF funding  Industry membership fees Designed for: Center for Health Organization Transformation (CHOT)     Variable costs (staff, principal investigators, students) Travel Materials and supplies Marketing Iteration: 2 Value Propositions Customer Relationships Customer Segments Research:  Gain a competitive advantage on evidence-based innovations in healthcare delivery specific and relevant to industry  Engage in a multidisciplinary approach to research  Have a valid, neutral third-party perspective by using university researchers to collect and analyze data  Pre-publication access to CHOT research findings at least two years ahead of publication  Access to CHOT university sites’ resources and facilities  Leverage credibility of the NSF CHOT research methodology and rigor to engage physician leaders Individual Level  Regularly scheduled communication between PIs and industry member  Regularly scheduled communication between co-directors  Regularly scheduled communication between CHOT and NSF Community Level  Bi-annual meetings (fall and spring)  Bi-annual conference calls (winter and summer)       Networking:  Work along with other industry members to develop CHOT’s research agenda  Develop working relationships with leading researchers from internationally recognized academic institutions  Access to top PhD, MHA, and MPH students from CHOT university sites  Display the NSF CHOT logo and Member seal (brand) as member of this unique IUCRC Professional Development:  Participate in CHOT Webinars (used for employee training, organization-wide learning networks or institutes for learning)  Co-author peer-reviewed articles with CHOT researchers  Co-present at professional conferences with CHOT researchers Channels  Technical/trade publications & presentations  Peer-reviewed publications and      presentations Monthly/bi-weekly webinars CHOT listserv CHOT website CHOT LinkedIn page CHOT quarterly newsletter Academic Publications  Publications in peer-reviewed journals  Presentations at academic conferences Cost Structure Designed on: March 2015 Revenue Streams      National Science Foundation core funds Membership fees Usage fees for patents and registered products Practice improvement services Healthcare executive training program Associations Government Health Systems Pharma – company/retail Retail Vendor (tech, consult, etc.) FAU Team • Collaboration – College of Engineering and Computer Science Ankur Agarwal Ravi Behara Gulcin Gumus Florida Atlantic University Boca Raton, Florida • Ankur Agarwal (PI) • Abhijit Pandya (Investigator) – College of Business • Ravi Behara (CO-PI) • Gulcin Gumus (CO-PI) Gulcin Gumus, Ph.D. Ravi Behara: Key Studies Assistant Professor, Health Administration Program, FAU College of Business Research Fellow, IZA, Institute for the Study of Labor, Bonn, Germany Research Affiliate, Health Economics Research Group, University of Miami • Research focus on health insurance, managed care, and safety-net hospitals. Examples include: – “Social Network Analysis of Provider Networks in the Workers’ Compensation System” Behara (FAU), Gumus (FAU), Borkowski (UAB), and Schmidt (FIU) – “Modifying Physician Behavior to Improve Cost Efficiency in Safety-Net Ambulatory Settings” Borkowski (UAB), Gumus (FAU), and Deckard (FIU) Transition of Care in Emergency Departments Analytic Approach to Kidney Allocation Patient Response to Pain Management Patient Comment Categories • Other research interests: effects of health insurance factors on labor markets, Patient Protection and Affordable Care Act (ACA), and traffic safety. Communication (Physician, Nurse, Administration, Staff) • Currently also serving as a consultant on a research project funded by the Agency for Health Care Administration (AHCA). This project involves both quantitative and qualitative data analyses to deter Medicaid fraud and abuse in the state of Florida. Treatment Protocol Personnel (Nurse, Staff) Care Process (wait) Investment in HIE Security Ankur Agarwal • Funding  CO-PI I/UCRC – CAKE FAU Site  NSF FRP  NSF PFI – Commercialization Grant  NSF RAPID • Research Accomplishment & Acknowledgement  Published as a breakthrough in I/UCRC NSF Compendium of Research 2014  Two Patent Disclosures  One licensed by a company • Faculty  Five Faculty Members Engaged in Research  College of Engineering + College of Business + Community Physicians • Students  Seven Students Graduated • Publications  Total of 14 Publications • Research Continuation  New Ideas for Field Advancement and Future Research Grants Areas of Interest Clinical Research Information Technology Healthcare Operations Mobile Technology and Healthcare Applications Why FAU …..? • Healthcare Companies in State of Florida • Florida Population and Healthcare Emphasis on healthcare • Research Emphasis by FAU Strategic Plan – Medical and Healthcare – Big Data Analytics • Existing Relationship with Healthcare companies • Prior Experience of Running a Successful I/UCRC – CAKE • Ongoing NSF work in Healthcare Informatics Potential Projects • Analyzing and Avoiding Hospital Readmission for CHF, Diabetes • Data Correlation and Fusion for Behavioral Analytics and Outcome Improvement in Sports Medicine Big Data Analytics in Healthcare Healthcare Information Security Health Economics and Patient Care • Data Analysis for Developing Decision Support System in Dermatology • Personal and Aggregated Community Health Score for Chronic Disease Potential Members • Medical Informatics 1. Quantum Innovation 2. SorenTech 3. Modernizing Medicine • Hospitals & Health Systems 4. Broward Health 5. Boca Raton Regional Hospital 6. Lynn Cancer Institute • Healthcare Security 7. HIPAA HiTech Solutions Collaboration with Companies in FAU Research Park 5 Year Goals for FAU Site 14 12 • 27 companies with 900 high-tech jobs • Incubator with 24 start-up companies 10 8 6 4 2 0 Year 1 Year 2 Companies Year 3 Patent Disclosures Year 4 Students Supplemental Year 5 NSF I/UCRC - CAKE, FAU Site 29 Industry Members                LexisNexis ProntoProgress Wigime, Inc. Relli Technologies SmartVCR, LLC ILS Technology Avocent/Emerson Corp. Jansyl Technologies Tecore Networks Aware Technology Adventure Automation LastBestChance, LLC Hillers Electrical Engineering Mobile Help Video Semantics               Tecore Wireless Systems Soren Technology Motorola Mobility (Google) Omega Optics Plangent Technology JM Family VideaStream M.R. Research Personics Labs C-Capture Florida Solar Energy Bridging Nations Foundation NOA, Inc. Wireless Sensor, LLC 1. Patient-Centered Care Cluster Chair – Eva K. Lee, Georgia Institute of Technology 1.1 Improving Health Promotion: Leveraging Statistical Learning and Electronic Medical Records for Healthcare Market Segmentation RFP: Patient Behavior & Self-care Nathaniel D. Bastian, Osama Alotaik, & Harriet B. Nembhard, Pennsylvania State University 1.2 Understanding the “White Space” of Where Patients Go After They Leave the Hospital RFP: Transitions of Care Amy Y. Landry & Larry R. Hearld, University of Alabama at Birmingham 1.3 User Perception and Use of Technology and the Association of Technology with Hospital Readmissions and ER Visits in Home Health Care RFP: Remote Health & Tele-health Darrell Burke & Midge Ray, University of Alabama at Birmingham 1.4 Healthcare system Optimization: advancing delivery timeliness, quality and effectiveness RFP: Organization & System Design Cody Wang, Eva K. Lee, & Matthew Hagen, Georgia Institute of Technology 1.5 Personalized Medicine Eva Lee & Xin Wei, Georgia Institute of Technology NSF IUCRC PROJECT PROPOSAL 1.1 Project Name: Improving Health Promotion: Leveraging Statistical Learning and Electronic Medical Records for Healthcare Market Segmentation Primary Investigator(s): Nathaniel D. Bastian, Osama Alotaik, & Harriet B. Nembhard, Pennsylvania State University Description: Healthcare market segmentation offers insights into healthcare consumers’ behaviors and attitudes, which is critical information in an environment where healthcare is moving rapidly towards patient-centered care. Personalized healthcare considers patient data from the electronic medical record (EMR) to help diagnose diseases, predict their onset, and suggest models for innovative healthcare delivery systems that better utilize resources to treat patients while improving health promotion in the community. Although every patient is unique, there are commonalities among patient characteristics (clinical, diagnostic, demographic, etc.) that can be discovered and leveraged through statistical learning methods to improve health promotion. By learning from patient data found in EMRs, we can identify and target specific types of patients to develop an effective healthcare market segmentation strategy. We seek to improve health promotion by dividing a community into homogenous subsets of patients who have common healthcare needs. Tailored marketing strategies can be designed and implemented to target these unique patient groups to improve health promotion in the community. Also, healthcare organizations are likely to interact with healthcare market segments, so meeting the preferences, needs, and demands of each segment may require innovative and tailored products and services, marketing approaches, business strategies, and new customer service models. Experimental Plan: Objectives are: 1) leverage unsupervised statistical learning methods to learn, explore and analyze underlying patterns in patient data from EMRs to identify healthcare market segments, and 2) leverage supervised statistical learning methods to develop predictive models using these segments to improve health promotion in a community by deploying effective healthcare marketing strategies (i.e. highly-targeted, consumer-oriented products, service offerings, and online support) to reach each type of patient in the population and propel them forward into greater engagement and self-management. The ultimate goal of this project is to achieve more satisfied patients, greater adherence to treatment choices, improved health outcomes, and reduced overall health care spending. How this is different than related research: This research project is different from previous studies which drew on marketing science to highlight the importance of market segmentation and investigate its effects using survey data in health care settings: 1) it has the purpose of improving health promotion in the community, and 2) it uses data from EMRs, which provides more reliability in terms of accuracy and sample size than self-response data. This research also differs in that it integrates unsupervised statistical learning methods with supervised methods to develop predictive models that can help in increasing the effectiveness of healthcare marketing strategies. Milestones & Deliverables: The expected deliverable is a framework leveraging statistical learning and EMRs to aid in healthcare market segmentation decision-making and follow-on health promotion strategy development. Nest, we will request EMR patient data elements from the Penn State Hershey Medical Center (1-5 months), conduct predictive analytics (6-8 months), develop the methodological platform (8-11 months), and disseminate the results (11-12 months). Potential member benefits: Gaining insights into healthcare consumers’ behaviors and attitudes, which is critical as the healthcare industry is moving rapidly towards patient-centered care; providing valuable clues as to how healthcare organizations may more effectively target and personalize products and services for healthcare consumers (focus of effort); managing resources more efficiently due to focus of effort; and enhancing competitiveness of healthcare organizations as better knowledge about patients constitutes a competitive advantage. Estimated Cost: $20,000 Project Overview and Description Improving Health Promotion: Leveraging Statistical Learning and Electronic Medical Records for Healthcare Market Segmentation Nathaniel D. Bastian, MS, MEng Osama Alotaik, MS Harriet B. Nembhard, PhD PI • Objective: improve health promotion in communities. • Healthcare market segmentation offers insights into: – – – • Healthcare consumers’ behaviors and attitudes. Critical in patient‐centered, personalized healthcare. Engender active participation in managing their health. There are commonalities among patient characteristics: – E.g., clinical, diagnostic, demographic, etc. • Discover and leverage using statistical learning methods to improve health promotion. • By learning from patient data found in EMRs: – – The Pennsylvania State University • Identify and target specific types of patients. Develop an effective healthcare market segmentation strategy. Improve health promotion by: – – Dividing a community into homogenous subsets of patients who have common healthcare needs. Tailored marketing strategies can then be designed and implemented to target these groups. Project Deliverables / Benefits Approach • Methodology: 1. 2. Leverage unsupervised statistical learning methods to learn, explore and analyze underlying patterns in patient data from EMRs to identify healthcare market segments. Leverage supervised statistical learning methods to develop predictive models using these segments. • How this is different than related research: – It has the purpose of improving health promotion in the community. – It uses data from EMRs rather than self‐response survey data (more reliable, larger sample size). – Integrated framework combining unsupervised and supervised statistical learning methods. • The expected deliverable is a framework leveraging statistical learning and EMRs to aid in healthcare market segmentation decision‐making • Benefits: – Gaining insights into healthcare consumers’ behaviors and attitudes. – Helping healthcare organizations to more effectively target and personalize products and services for healthcare consumers. – Managing resources more efficiently due to focus of effort. – Enhancing competitiveness of healthcare organizations. EMR Unsupervised Statistical Learning Health Market Segments Supervised Statistical Learning Predictive Models Subject Matter Expertise NSF IUCRC PROJECT PROPOSAL 1.2 Project Name: Understanding the “White Space” of Where Patients Go After They Leave the Hospital Primary Investigator(s): Darrell Burke & Midge Ray, University of Alabama at Birmingham Description: The ACA and value based purchasing have placed increased urgency on providing quality health care services beyond a specific acute care episode. Hospitals are now responsible for preventing avoidable readmissions for particular diagnoses, and will be financially penalized by Medicare for not doing so in an effective manner. Patients, however, are often discharged to a variety of settings and organizations, such as home health, skilled nursing facilities, inpatient rehabilitation, and home with no care, which can influence a hospital’s ability to manage the transition effectively. Understanding the frequency of different transitions of care and how successful these different transitions are in reducing avoidable readmissions for a health system can help decision makers choose the most appropriate care destination for patients and potentially target interventions to improve transitions. The main objectives of the project are to: 1) examine the settings and organizations to which patients are being discharged, and 2) identify whether certain settings and strategies of outplacement are more successful in reducing avoidable readmissions for patients with a particular diagnosis. Experimental Plan: We will conduct an analysis of discharge data for patients with a diagnosis of either COPD of CHF in two Deep South states. Mississippi and Florida were selected for this analysis because of data availability, and their similarity to the states in which our industry partners reside. These conditions were selected because they are both included in Medicare’s Hospital Readmission Penalties Program. The CHF diagnosis has been in the program since its inception, and COPD was added in 2014. Additionally, the southern region of the country has high incidence rates of both CHF and COPD. Investigators will analyze the AHRQ state-inpatient database to identify discharge settings (e.g. to home health, skilled nursing, etc.) for patients with diagnoses of CHF and COPD. Based on the data obtained from this analysis, readmission data for these patients will be gathered and analyzed to identify any differences or patterns in readmission rates based on organizational setting to which patients are discharged. This analysis will allow us to identify the most appropriate hospital outplacement strategy for patients with these two conditions. How this is different than related research: A variety of research is being conducted on clinical strategies to reduce avoidable readmissions. However, this work will evaluate the influence of a particular clinical setting on hospital readmission rates for particular conditions. Milestones & Deliverables: In the first quarter of the study year, data will be prepared for analysis. nd rd In the 2 and 3 quarters analyses will be conducted, and findings will be written up in the final quarter of the study year. We will produce a white paper outlining the findings of this report, and prepare a manuscript for submission to a peer reviewed journal. Potential member benefits: A variety of research is being conducted on clinical strategies to reduce avoidable readmissions. However, this work will evaluate the influence of a particular clinical setting on hospital readmission rates for particular conditions. Understanding which discharge settings minimize the likelihood of hospital readmission for a particular diagnosis is a useful tool for hospitals and health systems, particularly given the changing reimbursement environment. This work will help hospitals more effectively manage transitions of care and create appropriate discharge strategies based on patient diagnosis. Additionally, this work will be beneficial for providers of inpatient rehabilitation and home health care services. This will provide an evidence base for these providers to market their services by demonstrating how an appropriate transition in care for a specific diagnosis can benefit the patient. Estimated Cost: $50,000 Project Overview and Description • Overview: Understanding the “White Space” of Where Patients Go After They Leave the Hospital Amy Yarbrough Landry, PhD Larry R. Hearld – Analysis of how different outplacement strategies influence avoidable readmissions for particular conditions • Description: – Examine the settings and organizations to which patients are being discharged – Identify whether certain settings and strategies of outplacement are more successful in reducing avoidable readmissions for patients with a particular diagnosis. University of Alabama at Birmingham Approach • We will analyze discharge data for patients with a diagnosis of either COPD or CHF in two Deep South states using the AHRQ state inpatient database • Readmission data for these patients will be gathered and analyzed to identify patterns in readmission rates based on discharge setting • This analysis will allow us to identify the most appropriate hospital outplacement strategy for patients with COPD and CHF Project Deliverables / Benefits • Project Deliverables – Q 1: IRB approval; Data preparation – Q 2 and 3: Data analysis – Q 4: Final report preparation; manuscript preparation • Benefits – Results will help hospital partners effectively manage transitions of care and create diagnosis appropriate discharge strategies – Results will provide an evidence base for inpatient rehabilitation and home health partners to promote how an appropriate care transition can benefit the patient NSF IUCRC PROJECT PROPOSAL 1.3 Project Name: User Perception and Use of Technology and the Association of Technology with Hospital Readmissions and ER Visits in Home Health Care Primary Investigator(s): Darrell Burke & Midge Ray, University of Alabama at Birmingham Description: The use of healthcare technologies in home health has been increasing rapidly. There is an increasing availability of remote clinical data capture that can be used to manage the patient care more efficiently. In addition, based on a study of 847 home health agencies, the top three quality initiatives are hospitalizations, oral medication management and emergency care. We propose to identify and measure the utilization of technologies in a large home health agency; explore the association between technology and hospital readmissions and emergency room visits and look at geographic variation. The main objectives of this project are to: 1) measure the association of technology and hospital readmissions during the first 30 days of home health care; 2) measure the association of technology and emergency department use without hospital readmission during the first 30 days of home health; 3) identify predictors and barriers to successful technology use; 4) identify clinical technologies used in home health care; 5) develop survey based on literature review and home health agency observation of issues. Experimental Plan: We propose working with a large home health agency to measure the association of use of technology with the following two variables: hospital readmissions during the first 30 days of home health care and the emergency department use without hospital readmission during the first 30 days of home health care. In addition, we will assess the staff perspective of technology. We will administer a survey via the internet to the home health clinical staff, including RNs, LPNs, PTs and OTs to determine how the staff use the technology, the staff satisfaction with the technology and the staff perspective on patient use of technology. We will use administrative data to identify the hospital readmission and emergency room visits and the area health resource file to look at geographic variation. How this is different than related research: The current related research on the use of and satisfaction with home health technology is based on the perception of the organizational leaders. Our proposal will be the perception of the clinical staff users of technology. Also, with the 2015 quality measures for hospital readmissions and emergency department visits, understanding how technology influences these measures is important. Milestones & Deliverables: This project will be conducted over an 18 month period. During months 1 - 3 we will partner with home health agency, review literature & conduct clinical staff focus group, develop/modify survey, obtain IRB approval. During months 4 - 6 we will pilot test the survey in 1 - 2 branches (recruit and consent participants, collect data, analyze data, & modify protocol). During months 7 - 13 we will recruit and consent staff state-wide, collect data, merge data sources and analyze data. During months 14 - 18 we will prepare a final report on staff utilization of technology & associations, barriers to use and staff satisfaction with technology and the association between technology and hospital admissions/emergency room visits. The final milestone will be dissemination of the study findings, i.e. a publication and/or presentation that promotes translation of the research into practice. Potential member benefits: We plan to explore the association that technology has with hospital readmission and emergency department visits. In addition, we may identify barriers to the use of the technology and opportunities for training to facilitate the use of technology. Knowing that certain technologies may be associated with hospital readmissions and emergency room visits could assist home health care agencies with decision-making. Additionally, findings from the clinical staff survey of technology can serve to improve the training and use. Estimated Cost: $50,000 Project Overview and Description User Perception and Use of Technology and the Association of Technology with Hospital Readmissions and ER Visits in Home Health Care Darrell Burke, PhD Midge Ray, RN, CCS, MSN, PhD University of Alabama at Birmingham • Overview – Identify and measure staff satisfaction and use of technology in home health agency or other facility – Examine the association between technology use and hospital readmissions and ER visits within 30 days of HH care • Description – Identify the clinical technologies purchased and implemented in facility – Conduct online survey of staff perception of design, adoption, training and use of selected HH technology – Data sources include administrative data from HH agency, staff survey and Area Resource Files Project Deliverables / Benefits Approach • The projected 69% increase of home health positions through 2020 is five times that of national average • Increasing availability of remote clinical data capture and use ‐ Limited knowledge of the clinical users’ perspective of technology adoption and use • Proposed value‐based purchasing (CMS) ‐ With 5‐8% variance of CMS payment based on quality performance, the ability to select and use technology will enable more informed decisions Deliverables • Months 1 ‐ 3: Identify specific technologies, review literature, conduct staff focus group • Months 4 ‐ 6: Pilot test with small group of clinical staff • Months 7 ‐ 13: Recruit & consent staff and collect & analyze data • Months 14 ‐ 18: Final report for presentation to partner and begin development of a manuscript for publication Benefits • Identify barriers and facilitators of technology use • Findings will inform decisions regarding technology • Develop better understanding of how technology use may influence quality measures NSF IUCRC PROJECT PROPOSAL 1.4 Project Name: Healthcare system Optimization: advancing delivery timeliness, quality and effectiveness Primary Investigator: Cody Wang, Eva K. Lee, & Matthew Hagen, Georgia Institute of Technology Description: Individual health systems provide various services and allocate different resources for patient care. Healthcare resources including professional and staff time are constrained. Patients are ‘sicker’ often with a combination of chronic diseases. It would already take 16 – 18 hours daily to do everything the guidelines recommend that primary care provide for their patients. Patient lifestyle patterns are mostly suboptimal with adherence with pharmacotherapy is often limited. This study aims to 1) identify critical variables that impact outcomes (e.g. control of risk factors and prevention of hospital/ED admission) and inform allocation of limited time and resources for greater effect; 2) address realistically modifiable social determinants of health that will improve community health; and 3) seek greater use of treatment evidence (e.g. secondary EMR usage, “OMICs” data) to advance quality and effective of care delivery. We aim to increase quality and timeliness of care, maximize financial performance, and decrease practice variability across the organization. Experimental Plan: This is a collaborative project (CHOA, Grady, Morehouse, Northside, CCI). It covers over 2.7 million patient population. We will perform spatial (geocode) mapping of patient data across the care facilities. Big data mining, predictive analytics and systems modeling of key process, geographic, demographic, other predictors of health conditions and resource requirement will be performed. This will also include mining of EMR, laboratory/imaging results, and unstructured doctors’ notes for outcome prediction, across multiple providers and different socio-economic background. Service capabilities will be evaluated. Demand and resource will be aligned and optimized. Initial capacity assessment for change in a small but representative unit of practices will be performed. Outcome measures will be documented to understand potential impacts on change. Plans for system roll-out to multiple units will be designed. How this is different than related research: This study attempts to combine social-economic and demographics demands, hospital resources, and evidence of treatment (including EMR, Omics, and other laboratory data) to redesign the delivery process for quality and effectiveness of healthcare delivery. While efficiency is often performed via process improvement, patient risk factors, disease patterns and treatment characteristics may shed lights on resource needs and care requirement, and provide holistic health systems redesign opportunities for improving care quality and effectiveness. Milestones & Deliverables: This study will produce 1) a demand map by patients, types of services, socio-demographics, and hospital resource usage; 2) treatment outcome evidence of various types of diseases. Initial results will focus on treatment evidence of high-demand services; 3) resource usage, demand gaps, process bottlenecks, and systems capability assessment; 4) Systems redesign, change requirement, and implementation results for some chosen hospital units. The study will answer some fundamental question: “Will systems modeling of critical process variables impacting outcomes inform more efficient practice re-design to deliver effective patient-centered, team-based care and control modifiable risk factors than practice re-design uninformed by modeling alone?” Potential Member Benefits: Improve quality of care; improve efficiency of care; reduce waste, serve more needed patients; improve demand-resource alignment, reduce prolonged LOS (and thus reduce hospital acquired conditions), and improve capability in the event of pandemic or disaster response. From the patient standpoint, it offers timeliness and personalized evidence-based care, and reduces unnecessary hospital stay, associated risks and costs. This work has the potential to reduce healthcare delivery disparities. Estimated Cost: $70,000 Project Overview and Description Healthcare resources including professional and staff time are constrained. Patients are ‘sicker’ with combination of illnesses. Patient lifestyle patterns are mostly suboptimal, adherence with pharmacotherapy is often limited. Objectives • Identify critical variables that impact outcomes and inform allocation of limited time and resources for greater effect • Address realistically modifiable social determinants of health that will improve community health • Seek greater use of treatment evidence (e.g. EMR usage, “OMICs” data, “Precision Medicine”) to advance quality and effective of care delivery. • • • Healthcare System Optimization: Advancing Delivery Timeliness, Quality, and Effectiveness Eva K Lee, Cody Wang, Matthew Hagen Collaborative Effort (CHOA, Grady, Morehouse, Northside): Aim to increase quality and timeliness of care, maximize financial performance, and improve efficiency and effectiveness across the organizations. Georgia Institute of Technology Approach Project Deliverables / Benefits • Resource usage, Process maps, demand gaps, process, Systems simulation workflow, capability assessment Global system optimization GIS, Visualization, System models Complex disease modeling Machine learning Text mining Decision models • Optimal demand‐ resource alignment, timeliness, waste reduction • Socio‐demographics. demand patterns/needs • Personalized evidence‐based treatment models • Treatment outcome evidence & characteristics Multi‐units, multi‐disease, stakeholders, (big) data and evidence‐driven investigation Deliverables • • • • Demand map by patients, types of services, socio‐demographics, and hospital resource usage Treatment outcome evidence: high‐demand services vs costly procedures). Resource usage, demand gaps, process bottlenecks, and systems capability assessment Systems redesign, change requirement, and implementation results for some chosen hospital units Potential Benefits • • • • • • Improve quality and efficiency of care Reduce waste; serve more needed patients Improve demand‐resource alignment Reduce prolonged LOS (and thus reduce hospital acquired conditions), Improve capability in the event of pandemic or disaster response For patients: timeliness and personalized evidence‐based care; reduce unnecessary hospital stay, and associated risks and costs NSF IUCRC PROJECT PROPOSAL 1.5 Project Name: Personalized Medicine Primary Investigator(s): Eva Lee & Xin Wei, Georgia Institute of Technology Description: The term "personalized medicine" is often described as providing "the right patient with the right drug at the right dose at the right time." More broadly, personalized medicine (also known as precision medicine) may be thought of as the tailoring of medical treatment to the individual characteristics, needs, and preferences of a patient during all stages of care, including prevention, diagnosis, treatment, and follow-up. The project focuses on evidence-based approach where treatment design and management is personalized. In the events of multiple conditions, drug-drug interactions and side effects will also be modeled to minimize its negative effect. The objective of this study covers both the clinical visits, and a patient-home-centric approach to optimize the outcome and sustained health of patients. Experimental Plan: This is a collaborative project (Grady, Morehouse, CHOA, VA Atlanta). Our study will focus on broad areas of diseases: cancer, hypertension, cardiovascular, diabetes, obesity, and HIV. We have previously designed targeted personalized treatment plans for cancer patients with outstanding results (improving tumor control probability from 55% to 95%). Optimal treatment plans will be designed based on evidence, disease characteristics and combination, and collaborative decision from the patient (and family) and a care-team of providers. Predictive models will first be designed to uncover evidence of outcome versus treatment modality versus demographics from a cohort of patients. Treatment plan will be personalized and designed for individual patient for his/her conditions, with the goal to maximize the success of treatment outcome while avoiding polypharmacy and minimizing drug-drug interaction and risks. Biological and genomic information will be explored and incorporated along with clinical evidence. How this is different than related research: The project focuses on personalized treatment design and will accommodate potential co-existing multiple conditions, rather than a single disease. Thus, it is more challenging, interesting, and clinically relevant. Evidence will be uncovered from a set of realpatient data to establish the relationship of patient characteristics versus treatment outcome. A quantitative model based on patient characteristics and clinical desirable outcome will reduce the negative effect of individual provider’s subjectivity on decision making process on managing treatments and drug therapy. The project will bring together multi-team of providers to identify guidelines of multiple disease treatment. It will assist doctors to perform patient-centered complex treatment management. Milestones & Deliverables: Personalized treatment planning models for optimal drug therapy decision and intervention plan of patients (with multiple conditions) will be derived. Decision support system for clinicians to prescribe optimal medication will be developed. Financial and outcome evaluation of the usage of mathematical models in drug prescription will be performed. A number of disease conditions will be analyzed to experiment the methodologies’ applicability on real patient cases. Prioritization on the disease choices will be guided by patient needs and demand. Potential member benefits: The study focuses on personalized treatment design. The study will return optimal outcome-driven individualized treatment with lower cost and better control of disease symptoms. The resulting treatment will also use minimum amount of drugs, thus reducing the risk of adverse/side effects and increasing the efficacy of the treatment (more drugs mean high risk of noncompliance). This all will translate to improve the quality of care and quality of life of patients. Estimated Cost: $60,000 Project Overview and Description Personalized “Precision” Medicine Xin Wei, Eva K Lee Georgia Institute of Technology Personalized medicine: provide "the right patient with the right drug at the right dose at the right time. Precision medicine: tailor medical treatment to the individual characteristics, needs, and preferences of a patient during all stages of care, including prevention, diagnosis, treatment, and follow‐up. Motivation • Biological, imaging, genomic advances offer opportunities for novel treatment design • Physicians demand holistic approach to better manage disease conditions (minimize polypharmacy, maximize treatment outcome) • Patients demand better outcome and lower costs • Lifestyle and pro‐active engagement remain key to success • Internet technology is ubiquitous and offers timely opportunity Collaborative effort: Grady, Morehouse, CHOA, Northside, VA Atlanta. Approach Project Deliverables / Benefits Key features: 1. Predictive, Texting Mining, Decision Models • • • Analyze EMR, unstructured clinical and treatment notes Explore biological and imaging findings Explore effective control of multiple conditions. 2. Personalized Treatment Models • Multiple objectives, probabilistic, dynamic, outcome‐ biologic‐driven • Optimize within simulation • Polypharmacy network Deliverables • A predictive model to uncover treatment and outcome patterns • Personalized treatment planning models for optimal drug therapy decision and intervention plans of patients • Decision support system for clinicians to prescribe optimal drug therapy • Financial and outcome evaluation of the usage of mathematical models in treatment design & drug prescription Potential Benefits • • • • • • Understand complex treatment process, understand/quantify tradeoffs Return outcome‐driven treatment with lower cost and better control of conditions Allow clinicians and patients to focus on personalized outcome‐driven treatment Use minimum amount of drugs necessary, minimize risk of side effects Improved quality of care and quality of life of patients Reduce unnecessary side effects 2. Quality & Safety Cluster Chair - Bita A. Kash, Texas A&M University 2.1 Improvements to Root Cause Analysis of Patient Safety Events RFP: Transitions of Care Awatef Ergai & James Benneyan, Northeastern University 2.2 Examining How Lean Six Sigma Processes Reduce Hospital-Acquired Conditions RFP: Organization & System Design Deirdre McCaughey & Maria Hamilton, Pennsylvania State University 2.3 Does the Casework Design Affect the Patient Room Cleanliness and Healthcare-Associated Infection (HAI) Rates? RFP: Organization & System Design Ferhat D. Zengul, University of Alabama at Birmingham NSF IUCRC PROJECT PROPOSAL 2.1 Project Name: Improvements to root cause analysis of patient safety events Primary Investigator(s): Awatef Ergai & James Benneyan, Northeastern University Description: This research project will extend the body of knowledge regarding methods beyond simple root cause analysis (RCA) to analyze and reduce causes of healthcare adverse events (AEs). Patient safety and AEs are a widespread problem across healthcare, with huge cost and health implications, and have been the focus of significant improvement focus for over 20-30 years. Widespread progress on patient safety, however, on average has been slow and frustrating, and thus there is increasing interest in looking beyond the basic tools used to-date for new methods that might have value. In addition to RCAs, in other industries (such as aviation) other methods have been developed to help better classify and study adverse events so as to better prevent future occurrences. Examples include the Human Factors Analysis and Classification System (HFACS), System-Theoretic Accident Model and Processes (STAMPS), and others. This project therefore will adapt these other methods for healthcare application, refine them iteratively through use, and study their relative advantages and disadvantages versus RCAs. Northeastern works with several dozen health systems (both as part of CHOT and through our 3 other centers, healthcare systems engineering institute, and various grants) and as such has real-time access to hundreds of retrospective and prospective AE data – including central line blood stream infections, falls, employee injury, catheter associated urinary tract infections, surgical site infections, ventilator associated pneumonia, Clostridium difficile infections, and others. Experimental Plan: In this study, we will 1) adapt the above methods to healthcare, develop training materials, design pilot study, and process IRB research/ethics approval, 2) conduct RCA, HFACS, STAMPS, and other methods on error events that occur and analyze any relative advantage through identification of more failure reasons, prevention and sustainment of more error, associated costs and care improvements, and qualitative surveys, and 3) document their performance comparisons to RCAs. How this is different than related research: Almost all retrospective analysis of safety events in healthcare is done via the gold standard of root cause analysis, which has had some value but also some limitations. The above methods have been developed in other industries but rarely been used in healthcare and would need some adaptation to do so. Milestones & Deliverables: 1) Adaptation of methods beyond simple RCA to healthcare and development of training materials and pilot studies, 2) Illustration of the use of RCA, HFACS, STAMPS, and other methods on healthcare adverse events and documentation of their advantages and disadvantages, 3) Development of expository and research papers on the methods and their performance comparisons to RCAs. Potential member benefits: More informative analysis of their patient safety events and reduction in adverse events and associated costs. Estimated Cost: $42,000 Project Overview and Description Rationale HFACS Framework • Extend knowledge on methods beyond root cause analysis to analyze and reduce causes of healthcare adverse events Improvements to Root Cause Analysis of Patient Safety Events Awatef Ergai, PhD James Benneyan, PhD Northeastern University • Methods used in high‐risk industries are rarely used in healthcare, such as – Human Factors Analysis and Classification System (HFACS) – System‐Theoretic Accident Model and Processes (STAMPS) – Rasmussen’s risk management framework (Accimap) Relevance • Patient safety and AEs are a widespread problem in healthcare with slow progress • Impact on patient safety with huge cost savings Approach Project Deliverables / Benefits Phase 1 (months 1‐6) Milestones / Deliverables • Adapt mentioned methods to healthcare and refine iteratively through use • Develop training materials • Design pilot study • Process IRB research/ethics approval • Adaptation of methods to healthcare and development of training materials and pilot studies • Illustrate how to use these methods on healthcare adverse events • Document their performance comparisons to RCA • CHOT report, member webinar, and journal‐ready publication(s) to disseminate findings Phase 2 (months 7‐12) • Conduct RCA, HFACS, STAMPS, Accimaps, and others on error events • Analyze and document their advantages and disadvantages through – – – – Identification of additional failure reasons Prevention and sustainment of more errors Associated costs and care improvements Qualitative surveys Potential Member Benefits • More informative analysis of patient safety events • Reduction in adverse events and associated costs NSF IUCRC PROJECT PROPOSAL 2.2 Project Name: Examining How Lean Six Sigma Processes Reduce Hospital-Acquired Conditions Primary Investigator: Deirdre McCaughey & Maria Hamilton, Pennsylvania State University Description: The Hospital-Acquired Condition (HAC) Reduction Program, implemented by the Centers for Medicare & Medicaid Services (CMS), serves the purpose to achieve better patient outcomes while slowing health care cost growth. The program targets largely preventable conditions that patients did not have upon admission to a hospital, but which developed during the hospital stay. Hospital performance under the HAC Reduction Program is determined based on a hospital’s total HAC score and all hospitals that rank in the worst quartile of HAC scores will receive a payment reduction of one percent for all CMS services. With the average American hospital earning approximately 5 % margin on, a loss of 1% revenue has the potential to be a significantly negative effect on the financial viability of some hospitals. Experimental Plan: The project seeks to extend the previous work conducted by the coinvestigators examining the efficacy of Lean Six Sigma processes to examine the processes and related to sources of system breakdowns that result in HACs occurring. The project has 3 phases. In the first phase the research team will we will conduct a retrospective review of the HAC events (Patient Safety Indicators) from 2012 to 2014 at Hershey Medical Center. We will examine the data and through data analysis, we will identify the relevant antecedents to HAC occurrences as well as the effect of HAC events on incremental hospital costs and length of stay indicators. In the second phase, Lean Six Sigma methodology will be utilized to identify root cause factors that have contributed to HAC events. We will conduct a rapid improvement event with all stakeholders to document current process, confirm the identified root causes, and develop action plans. In the third phase, we will compare findings from this extension study with our first study (2014) to identify the HAC root causes and process deficiencies that have occurred in both studies. How this is different than related research: Limited research exists that examines the efficacy of Lean Six Sigma processes (e.g. rapid improvement events) on quality improvement in healthcare organizations. Research evidence is needed that explores and identifies how using process improvement methodologies positively impacts HAC performance. Utilizing an extension of our first study to test the effect of rapid improvement event on HAC frequencies will serve as a unique validation of our initial finding and provide an evidence-based foundation from which this methodology can be utilized to improve hospital HAC performance. Milestones & Deliverables: Acquisition, coding, and cleaning data will occur in months 1-3. Analysis of data and identifying emerging process issues will take place in months 4-5. Rapid improvement event will be held in month 6 and process improvement strategies will be developed in months 6. Months 7-9 will serve to monitor the effect of the rapid improvement event and serve to prepare final reports for presentation to both practitioner and academic audiences as well disseminate finding to participating hospital stakeholders. Potential Member Benefits: The results of this research will assist all hospitals in better utilization of Lean Six Sigma methodologies to examine deficient hospitals processes that result in HACs. Further, extending our previous research (2014), the project will offer hospitals a critical evidencebased “next - step” in utilizing the study results to improve patient care thereby fostering greater utilization of this research. Estimated Cost: $20,000 Project Overview and Description Examining How Lean Six Sigma Processes Reduce Hospital Acquired Conditions On any given day, 1 in 25 USA hospital patients has at least one healthcare‐associated infection (HAI) & 75,000 hospital patients with HAIs died during their hospitalizations1. Two CMS Programs: 1. IPPS Preventable HAC Program (12 HACs) 2. Hospital Acquired Conditions (HAC) Reduction Program Deirdre McCaughey, PhD,MBA Maria Hamilton, MBA, BSIE, CSSBB The Pennsylvania State University – New for 2015 (8 HACs) – Improve quality of inpatient care by providing a negative financial incentive to hospitals who do not reduce HACs – Will result in 1% penalty on reimbursement – Teaching hospitals will be disproportionately affected by HAC reduction program due to total payments being reduced, including add‐ons (IME and DSH) 1CDC HAI Prevalence Survey, 2011 Approach • Research question: How can previous Lean Six Sigma methodology research be extended to identify and improve system breakdowns contributing to HACs occurrences across various organization system areas? • Using HAC occurrence data at HMC from UHC for 2014‐2015, the project will: – Identify relevant antecedents to HACs and the impact on hospital costs and length of stay – Identify most frequently occurring HACs – Use Lean Six Sigma processes (root cause analyses & implementation improvements) to identify HAC event sources and system breakdowns – Conduct a RIE with all stakeholders of the selected root cause to document the current process, identify pain points, waste, and rework, and develop action plans – Compare & contrast findings of this replication study with original study (CHOT 2014‐2015) to further validate the efficacy of Lean Six Sigma in reducing HACs Project Deliverables / Benefits • Data analysis: – Acquisition, cleaning & coding of data – Identification of emerging process issues – Process improvement strategies developed • Action step: – Rapid Improvement Event to validate findings • Dissemination: – Stakeholder presentations & feedback • Implementation: – Incorporate RIE action items & monitor results • Benefit: – Validate additional mechanisms of Lean Six Sigma methodology to reduce HACs – Demonstrate continued efficacy of previous research findings in reducing HAC frequencies (CHOT 2014‐2015) NSF IUCRC PROJECT PROPOSAL 2.3 Project Name: Does the Casework Design Affect the Patient Room Cleanliness and HealthcareAssociated Infection (HAI) Rates? Primary Investigator(s): Ferhat D. Zengul, University of Alabama at Birmingham Description: U.S. hospitals utilize strategies to reduce healthcare-associated infections (HAIs), which can be very costly in regard to patient lives, hospitals’ financial bottom-line and public image. Cleanliness in hospitals, specifically surface cleanliness, has been identified as a major prevention method for infections. Recently, some hospitals have started adopting new casework designs that incorporate wall mounted casework, and faucets. These designs eliminate sink desks and provide solid surface tops, backsplash and offset drain. However there is little evidence-based information on the relationship between these new casework designs, and improvement in cleanliness and reduction in infection rates. This project’s aim is to evaluate these new casework designs and their impact on patient room cleanliness and HAIs. Experimental Plan: We will work with hospital partners who are planning to implement such designs 1. Identify two similar nursing units with similar features in infection incidence rates, design, and cleaning practices 2) Collect data for 4 months on HAIs rates, and surface cleanliness by random sampling of patient rooms from each of two units. A surface hygiene test called Clean-Trace will be used to collect surface cleanliness data. 3) Implementing a novel casework design (intervention) in one of the units while keeping the other unit as a control unit. 4) Collecting further 4 months of data on HAIs rates, and surface cleanliness, by random sampling of exam rooms both from the intervention unit and the control unit. 5) Comparing the pre-post data and data from intervention and control units. How this is different than related research: Previous studies on HAIs mostly concentrated on changing the behaviors of clinical staff through standardization, compliance, and evidence-based strategies. Moreover, the link between contamination (uncleanliness) and development of HAIs is not very well established or investigated. Many design interventions focus on infection risk reduction through structural changes such as single patient rooms, antibacterial and cleanable surfaces, strategic location of sinks, and hand sanitizer dispensers. There is a need for more studies that investigate the relationship among facility design interventions, cleanliness, and HAIs. The integration of pre-post and intervention-control methods along with extensive data collection on HAIs, and surface cleanliness make this project unique. Milestones & Deliverables: Months 1-2: Coordinate the project with UAB Hospital facilities team, recruit graduate students, obtain IRB approval, and request historical administrative data to analyze infection incidence rates. Months 3-4: Identify two similar nursing units by analyzing historical infection rates and collaborating with hospital facilities team. Months 5-8: Collect patient room surface cleanliness data from both units through random sampling. Months 9-12: Implement the design intervention in one of the unit. Months 12-15: Collect patient room surface cleanliness data from both the control and intervention units through random sampling. Months 16-18: Prepare final report and begin the development of a manuscript for publication to disseminate lessons learned. Short-term deliverable: report for industry member regarding the study findings Long-term deliverable: the improvement in the knowledgebase regarding the potential association between facility design interventions, and cleanliness and HAIs. Potential member benefits: Understanding how implementation of such designs influences HAIs, and surface cleanliness is necessary for decision makers in making evidence-based decision in adopting these novel designs. Obviously, making large capital purchases on patient room caseworks by relying solely upon the assessments of ultimate sellers of these caseworks is not in the best interest of our industry partners. Therefore, healthcare facilities that plan to adopt such casework designs need the results of this project and similar projects to be able to make more informed decisions for their capital projects. Estimated Cost: $50,000 Project Overview and Description Does the Casework Design Affect the Patient Room Cleanliness and Healthcare‐Associated Infection (HAI) Rates? Ferhat D. Zengul, PhD, MBA • Surface cleanliness, a major prevention method for infections. • New casework designs that incorporate wall mounted casework, and faucets. • Elimination of sink desks and having solid surface tops, backsplash and offset drain. • Little evidence on the relationship between these new casework designs, and improvement in cleanliness and reduction in infection rates. • The aim of this project is to evaluate these new casework designs and their impact on patient room cleanliness and HAIs. • Will work with hospital partners who are planning to implement such designs. University of Alabama at Birmingham Approach Project Deliverables / Benefits • Surface cleanliness and HAIs are major issues for hospitals Steps: • Identify two similar nursing (i.e., infection incidence rates, design, and cleaning practices) • Collect data for 4 months on HAIs rates, and surface cleanliness (A surface hygiene test called Clean‐Trace). • Implement casework design in the intervention unit • Collect further 4 months of data • Compare the pre‐post data and data from intervention and control units. • A report to the industry partner regarding the findings of the study. • Due to lack of evidence, the decision makers at hospitals mostly rely on the self‐ assessments of the casework manufacturers. • Be able to make more informed decisions for their capital projects. 3. Enabling HIT & Care Coordination Cluster Chair - Rob Weech-Maldonado, University of Alabama at Birmingham 3.1 Improvements to Root Cause Analysis of Patient Safety Events RFP: Remote Health & Tele-health Ajay Bharathi, Conrad Tucker, & Harriet Nembhard, Pennsylvania State University 3.2 Examining How Lean Six Sigma Processes Reduce Hospital-Acquired Conditions RFP: Remote Health & Tele-health Linlin Ma, Harriet B. Nembhard, & Harleah Buck, Pennsylvania State University 3.3 Does the Casework Design Affect the Patient Room Cleanliness and Healthcare-Associated Infection (HAI) Rates? RFP: Patient Behavior & Self-care Abhinav Singh, Conrad Tucker, & Harriet Nembhard, Pennsylvania State University 3.4 Reducing Readmission after Hip Surgery using Statistical Process Control and Smart Home Care RFP: Transitions of Care Yifeng Yu & Harriet B. Nembhard, Pennsylvania State University NSF IUCRC PROJECT PROPOSAL 3.1 Project Name: An Immersive Virtual Reality Approach for Real-Time, Scalable Learning in Healthcare Primary Investigator(s): Ajay Bharathi, Conrad Tucker, & Harriet Nembhard, Pennsylvania State University Description: The objective of this project is to test the hypothesis that immersive virtual reality (VR) environments provide comparable learning outcomes in healthcare, compared to brick and mortar environments. The outcome of this project has broad impacts in healthcare. From a patient’s perspective, learning how to utilize a specific machine prior to a laboratory examination could potentially reduce anxiety and safety risks. For example, if a patient could simulate the actions of preparing for an examination involving a Magnetic resonance imaging (MRI) machine within an immersive virtual reality machine, they may be more prepared for the real physical examination. For healthcare practitioners, immersive virtual reality platforms have the potential to minimize risk associated with learning new procedures or training for existing ones, as the consequences of mistakes are shielded from real world impact. Recent technologies such as the oculus rift have opened the door to low cost, scalable methods of achieving 360 immersion in virtual reality environments. A recent study by the PIs have determined statistically significant differences in learning outcomes of engineering students using immersive environments and engineering students using traditional virtual reality environments. This project will expand on these initial findings by exploring the impact of immersive environments in healthcare learning environments. Experimental Plan: Step 1: Identify a healthcare population sample where the hypothesis will be tested Step 2: Given the healthcare population, partition into two samples: i) brick and mortar sample and ii) Immersive VR sample Step 3: Outline a healthcare task and test the hypothesis to determine whether a there exists a statistically significant learning outcome between subjects performing a task in an immersive VR environment, compared to the brick and mortar environment Step 4: Report results through presentations and conference and journal publications How this is different than related research: Until recently, a major limitation of traditional virtual reality platforms has been the lack of an immersive experience that not only provide content to individuals, but also enables them to interact and learn in a completely 360 degree immersive environment. There exists a knowledge gap in terms of how these immersive virtual reality platforms impact learning in a healthcare environment. Milestones & Deliverables:  Discovery of the impact of immersive VR environments in enhancing learning in healthcare  Developed a strategy with Industry partners for hardware acquisition and data storage and transfer requirements for the upcoming trials with patients from Hershey Medical Center. Potential member benefits: Potential member benefits 1. For patients, insurance companies and hospitals, immersive VR systems will transform the manner in which learning occurs in healthcare, towards personalized wellness outcomes 2. IT industries can benefit largely from the software platforms developed under this project and a better understanding of the data acquisition, transfer and management needs. 3. For the NSF apart from the practical research these two areas provide they also provide intriguing algorithmic questions for us to solve from the confluence of various fields. Estimated Cost: $20,000 Project Overview and Description An Immersive Virtual Reality Approach for Real‐Time, Scalable Learning in Healthcare Ajay Bharathi, Conrad Tucker, PhD Harriet Nembhard, PhD Hypothesis: Immersive virtual reality (VR) environments provide comparable learning outcomes in healthcare, compared to brick and mortar environments Brick and Mortar Immersive Virtual Reality The Pennsylvania State University Approach Step 1: Identify Healthcare Use Case Step 4: Data Mining Knowledge Discovery Project Deliverables / Benefits Step 2: Create Immersive Virtual Reality Health Care Platform Step 3: Test Hypothesis Brick and Mortar Immersive Virtual Reality • Investigate the impact of immersive VR in healthcare • Individually Customized Healthcare • Enable the remote management and interaction between patients and healthcare providers Immersive VR Environments 360 Degree Hardware Learning NSF IUCRC PROJECT PROPOSAL 3.2 Project Name: Assessment of a Telehealth Device in Promoting Heart Failure Patient Engagement and Self-Care in Rural Areas Primary Investigator(s): Linlin Ma, Harriet B. Nembhard, & Harleah Buck, Pennsylvania State University Description: Chronic Heart Failure (HF) involving approximately six million people is the most common cause of hospitalization in the U.S. with subsequent costs estimated at nearly 41 billion dollars per year. It is currently unknown how many of these people reside in rural areas. What is known is that HF patients in rural, primarily medically underserved areas are more likely to be older and in poorer overall health than their suburban and urban counterparts. Rural HF patients also frequently lack easy access to community based support, like outpatient clinics, that are taken for granted in more populous areas. Telehealth technology, by addressing geographic distance, is one potential solution to improve HF self-management while providing added professional support as needed in rural areas. However, HF management for rural patients is challenging and in need of innovative, interprofessional management strategies and technologies. Experimental Plan: Three phases: 1) Analyze and visualize health disparity data and prevalence, using Geographic Information System (GIS) tools. Then target communities or HF patient groups in most need. 2) Develop/modify and test a telehealth technology system in a series of waves to assess the feasibility of each component of the telehealth technology system and the acceptability of the overall program using project adherence and retention as measurable outcomes. 3) Design interview schedules and surveys for the targeted patients. Surveys will assess self-care, HRQOL, and rehospitalization rates of HF patients. Four categories for characterizing potential HF patient groups: 1) Availability of HF related resources 2) Access of the resources, 3) Vulnerability of the population, 4) Affordability of the healthcare services. A financial projection and cost-effective analysis in a broader market setting will be made. Data of HF prevalence, healthcare resources distribution across the state can be searched through databases such as National Center for Health Statistics. How this is different than related research: 1) through comprehensive analysis of HF care disparities, specifically in rural area, we are able to visualize and target patient groups in the most need and 2) Its relation to this innovative telehealth technology for HF self-care. In an integrative research review (IRR) in 2013, only four clinical trial studies using telehealth in the management of heart failure in rural settings were found, and studies reviewed were limited to the strategies of telephone follow-up calls and internet-based virtual visits. Milestones & Deliverables: Literature review and preliminary market analysis with public available data/mapping have been carried out. The next steps are:  Understand raw datasets and extract the metrics that matches this study’s need, and then visualizing health disparity data and prevalence, using GIS tools.  Build models, define measurements for the patient groups in the most need.  Define test procedures, sample size, and carry out developmental clinical trials, measure adherence and retention. Potential member benefits:  Awareness of target market with the strongest potential.  Better understanding of patient behavior and patients’ expectations.  With real clinical trials, reliable financial projections of this device can be carried out.  From interview/survey records from both patients and care providers  Help healthcare providers to improve the quality of care of HF patients, which will bring a positive financial influence. Estimated Cost: $20,000 Project Overview and Description • The gap between rural health disparities and the Chronic Heart Failure (HF) patients needs in rural areas. Assessment of Telehealth in Promoting Heart Failure Patient Engagement and Self‐ Care in Rural Areas Linlin Ma, MS Harriet B. Nembhard, PhD Harleah G. Buck, PhD The Pennsylvania State University – Elderly population – Low availability of health care resources – Relatively high Coronary Heart Disease hospitalization and obesity population – Difficulty in transportation • Better describe rural population with HF care related metrics. – Social economic indicators – Health risk factors – HF care resource locations • Addressing HF management in rural areas by telehealth technology. Approach • Statistical analysis & regression model ‐ study the correlation among the metrics and HF prevalence. • Predictive model ‐ flag communities with high HF risk. • Geographic Information System (GIS) – visualization of HF care related health disparity issues. • Different from previous research – Address HF related characteristics not in the same level, but in 5 categorized groups. – The potential relation between social economic indicators and HF risk/prevalence – Improve HF‐management from patient interview/follow‐up feedback Project Deliverables / Benefits • Milestones & Deliverables: – Visualization of HF care disparity data and HF prevalence, using GIS tools. – Building models, define measurements for the patient groups in the most need. – Define test procedures, sample size, and carry out developmental clinical trials, measure adherence and retention. – Cost‐effectiveness analysis • Benefits – Awareness of target market with the strongest potential. – Better understanding of patient behavior and patient expectation. – With real clinical trials, reliable financial projections of this device can be carried out. – From interview/survey records from both patients and care providers, preference for this engagement tools can be addressed. – Help healthcare providers to improve the quality of care of HF patients, and narrow down the gap between current product and patient expectation, which will bring a positive financial influence. NSF IUCRC PROJECT PROPOSAL 3.3 Project Name: Investigating the Impacts of a Patient’s Social Network in Achieving Gamification Solutions in Personalized Wellness Management Primary Investigator(s): Abhinav Singh, Conrad Tucker, & Harriet Nembhard, Pennsylvania State University Description: The term “gamification” is an emerging paradigm that aims to employ game mechanics and game thinking to change behavior. In order to successfully employ gamification principles to change behavior in personalized self-care, researchers must understand the concepts of game design, where:  mechanics represents the basic processes that drive the action forward and generate player engagement. These include game elements such as challenges, competition and cooperation  dynamics represents the big-picture aspects of how a gaming system works. These include game elements such as achievements, avatars, badges, levels and points  aesthetics represents the visual stimuli that engage an individual. These include game elements such as game expression, game narrative and game discovery The authors hypothesize that statistically significant differences exist in the social network structure of patients with successful gamification outcomes versus patients with unsuccessful gamification outcomes. The above three game design concepts of mechanics, dynamics and aesthetics have a social network component, connecting individuals with one another. For example, the competition component of the mechanics concept would require that patients have someone to compete with in a meaningful way. Furthermore, the quality of competition may be just as significant as the competitive task itself. While there exist models that outline gamification features, there exists a knowledge gap in how individuals’ social networks impact their motivation to completing tasks. This research aims to fill this knowledge gap. Experimental Plan: Step 1: Identify a patient population sample where the hypothesis will be tested Step 2: Given the patient population, mathematically model their social network Step 3: Outline a gamification task and test the hypothesis to determine whether a patient’s social network has a statistically significant impact on a patient’s success on a gamification task Step 4: Report results through presentations and conference and journal publications How this is different than related research: The main limitations of existing techniques are that patients do not engage with such systems for a prolonged period of time. Furthermore, the individuals that typically utilize such self-improvement platforms are they themselves already committed to the success of their wellness management. There exists a knowledge gap in how a patient’s social network influences their ability to adhere to wellness management protocols. Milestones & Deliverables:  A mathematical model of a patient’s social network, as it relates to their personalized wellness  Developed a strategy with Industry partners for hardware acquisition and data storage and transfer requirements for the upcoming trials with patients from Hershey Medical Center. Potential member benefits: Potential member benefits 1. For patients, insurance companies and hospitals, gamification will transform the manner in which wellness management is designed and advanced 2. IT industries can benefit largely from the software platforms developed under this project and a better understanding of the data acquisition, transfer and management needs. 3. For the NSF apart from the practical research these two areas provide they also provide intriguing algorithmic questions for us to solve from the confluence of various fields. Estimated Cost: $20,000 Project Overview and Description Investigating the Impacts of a Patient’s Social Network in Achieving Gamification Solutions in Personalized Wellness Management Abhinav Singh Conrad Tucker, PhD Harriet Nembhard, PhD • “gamification” employs game mechanics and game thinking to change behavior • How does a patient’s Social Network impact their success in behavior change? The Pennsylvania State University Approach Step 1: Creation of Gaming System Steps 4 : Correlation: Patients’ Social Network and Gamification Outcomes Project Deliverables / Benefits Steps 2 : Identify Patient Population Step 3: Model Patients’ Social Network • Game Dynamics to motivate patients • Adaptive learning from patient response • Enable physicians to remotely track patient adherence to their prescribed treatment in a quantitative manner Incentives in the gaming system Recording patient response Customization NSF IUCRC PROJECT PROPOSAL 3.4 Project Name: Reducing Readmission after Hip Surgery using Statistical Process Control and Smart Home Care Primary Investigator(s): Yifeng Yu & Harriet B. Nembhard, Pennsylvania State University Description: The unplanned readmission after hip surgery has become an increasingly serious problem. In fiscal year 2015, Centers for Medicare & Medicaid Services (CMS) has started to penalize hospitals for high readmissions after elective hip replacement. In addition, cause of readmission after hip surgery varies from patient to patient, involving both hospital-care and home-care problems. Thus, high-quality and more coordinated care should be provided to the hip-replacement patients. To achieve this goal, it is necessary to improve the hip-surgery process in hospitals by applying evidencebased practice. Moreover, monitoring and estimating patient recovery during the first six weeks after discharge is also indispensable for improving patient adherence to physicians’ instructions and detecting potential problems in recovery. By an attempt to seamlessly integrating preoperative, intraoperative, and postoperative care, it is probable that the readmission rate of hip replacement can be effectively reduced. Experimental Plan: Efforts on reducing readmission should be made on improving both hospital care and home care. As a result, the objectives of this project are to 1) set up logistic regression model to determine the impact of surgical factors and nonsurgical factors on readmission, and 2) utilize riskadjusted statistical process control (SPC) to adjust for the different pre-surgery risk of patients, and conduct real-time monitoring of a hospital’s hip-surgery readmissions, and 3) identify the high-risk patients, and develop smart home-care device to monitor and predict their movement recovery after discharge, 4) perform a cost-effectiveness analysis of the smart home-care intervention to justify its future applications. How this is different than related research: 1) risk-adjusted control statistics and control limits are constructed for real-time monitoring of the readmission rate of a hospital (i.e., it alarms immediately if the control statistics exceed the control limits, thus root-cause diagnosis can be launched right away), and 2) the risk-adjusted control statistics enable the detection of both deterioration and improvement in the hip-surgery quality, and 3) the smart home-care device can monitor and predict the recovery status of patients, and this information can be transferred immediately to the physicians for further feedback and instruction. Milestones & Deliverables: The expected deliverables are a method for conducting real-time SPC of the hip-surgery process, and a design for monitoring and estimating patient movement recovery during the home care stage. The next steps are 1) request data on hip surgery with readmission information from hospitals and medical centers (1-3 months), and 2) set up logistic regression model to determine the impact of surgical factors and nonsurgical factors on hip-surgery readmission (4-5 months), and 3) construct control statistics and control limits for risk-adjusted SPC of hip-surgery quality (6-7 months), and 4) develop smart home-care device for postoperative monitoring and prediction (8-10 months), and 5) disseminate the results (11-12 months). Potential member benefits: Gaining insights into providing patient-centered care for diverse patient clusters, depending on patient characteristics and risk factors; monitoring and improving the quality of hip-surgery process at hospitals; benefit industries by defining clinical processes and data management; designing and developing more effective and advanced home-care technologies for preor postoperative care; enhancing patient adherence at the postoperative stage; reducing hospital readmission rate of hip replacement as well as penalty from CMS; providing more coordinated and integrated care to increase patient satisfaction; developing effective interventions and the best practice of perioperative care for hip replacement. Estimated Cost: $20,000. Project Overview and Description • High hospital readmissions after hip replacement began to be penalized Reducing Readmission after Hip Surgery using Statistical Process Control and Smart Home Care by CMS in fiscal year 2015 • Cause of readmission after hip surgery varies from patient to patient, involving both hospital‐care Yifeng Yu, MS PhD student Harriet Black Nembhard, PhD and home‐care problems • Surgical site infection, dislocation, hematoma, deep vein thrombosis, non‐ infected draining wound, … The Pennsylvania State University Improved perioperative care for hip‐ surgery patients is necessary to effectively reduce hospital readmissions Project Deliverables / Benefits Approach • A method for real‐time monitoring of hip‐surgery quality, and a design for seamless home care for discharged patients Surgical and nonsurgical factors: Age? Co‐morbidity? LOS? Integrate preoperative, intraoperative, and postoperative care for hip‐ replacement patients Cost‐effectiveness analysis: High‐risk patients? $/QALY? Provide patient‐ centered care for diverse patient clusters Improve the quality of hip‐surgery procedures Develop the best practice of perioperative care for hip‐surgery patients Benefit industries by defining clinical processes and data management Design effective home‐care technologies for pre‐ or postoperative care 4. Access & Efficiency Cluster Chair - Deirdre McCaughey, Pennsylvania State University 4.1 Ebola epidemic regional and facility response models RFP: Organization & System Design Hande Musdal & James Benneyan, Northeastern University 4.2 Identifying and Utilizing Inexpensive Technologies to Manage Patient Populations RFP: Remote Health & Tele-health Amy Y. Landry, University of Alabama at Birmingham 4.3 Robust healthcare staff scheduling RFP: Organization & System Design Sibel Sonuc & James Benneyan, Northeastern University 4.4 Challenges in telemedicine – a systematic review and engagement with rural communities RFP: Remote Health & Tele-health Eva K Lee, Jean Kang, & Isabella Carbonell, Georgia Institute of Technology NSF IUCRC PROJECT PROPOSAL 4.1 Project Name: Ebola epidemic regional and facility response models Primary Investigator(s): Hande Musdal & James Benneyan, Northeastern University Description: The purpose of this project is to develop and test two simulation models to help evaluate and improve hospital and regional response plans for a potential Ebola epidemic. Many hospitals in the US have been developing response plans and running practice drills in the event of another epidemic, but little rigorous analysis has been conducted as to how well they will work under a range of various scenario conditions (volume of patients, rate of spread, extent of regional infection, resource availability, and so on). This project therefore will develop two models, one of a region’s response plan and one of an individual hospital’s plan, working closely with ICU and ED leaders who have developed their plans and/or are responsible for their execution. These models then will be used to “stress test” these plans under a wide variety of conditions, identify the greatest opportunities for improvement of greatest potential failures, and provide decision support for their refinement. If useful, the developed models will be made publically available, and also may be adaptable for analysis and improvement of response plans for other types of epidemics and natural disasters. Experimental Plan: This project aims to 1) develop understanding of response plan specifics through available documentation, interviews, process mapping, and cross-validation, 2) develop computer simulation models at the facility and regional levels, and validate them through face-validity methods, and 3) identify and conduct analyses with CHOT partners of their current response plans and potential improvements. How this is different than related research: Hospitals and communities across the US have been developing epidemic response plans and conducting simulation/practice drills as to how they will respond to an epidemic, in terms of the process of care for patients, how to room them, capacities, protocols, and the like. Milestones & Deliverables: 1) Analysis of current Ebola response plans of multiple CHOT members, 2) Development and validation of two computer simulation models, first one at the facility level and second one at the regional level, 3) Comparison of the current response plans with the alternative ones through the use of simulation models developed, 4) Development of CHOT report, member webinar, and journal-ready publication(s) to disseminate findings. Potential member benefits: Analysis and potentially improvement of Ebola response plans, identification of failure conditions beyond which a facility can adequately accommodate care needs, and development of a tool kit with potential to adapt to similar concerns. Estimated Cost: $42,000 Project Overview and Description Rationale Hande Musdal, PhD • Develop and test simulation models to help evaluate and improve hospital and regional response plans for potential Ebola epidemic • Little rigorous analysis conducted as to how well response plans will work under various scenario conditions (e.g., volume of patients, rate of spread, extent of regional infection, resource availability, etc.) James Benneyan, PhD Relevance Ebola Epidemic Regional and Facility Response Models Northeastern University Approach 1. Develop understanding of response plan specifics through available documentation, interviews, process mapping, and cross‐validation 2. Develop two computer simulation models, one of an individual hospital’s plan and one of a region’s response plan 3. Validate the developed models through face‐validity methods 4. Identify and conduct analyses with CHOT partners of their current response plans and potential improvements • Many US hospitals have been developing response plans and running practice drills in the event of another epidemic • Will help identify opportunities for improvement of potential failures, care for patients, how to room them, capacities, protocols, etc. Project Deliverables / Benefits Milestones / Deliverables • Analysis of current Ebola response plans of multiple CHOT members • Development and validation of two computer simulation models, at the facility and regional level • Comparison of current response plans with alternatives • CHOT report, member webinar, and journal‐ready publication(s) to disseminate findings Potential Member Benefits • Analysis and potential improvement of Ebola response plans • Identification of conditions beyond which a facility can adequately accommodate care needs • Potential to adapt to similar concerns NSF IUCRC PROJECT PROPOSAL 4.2 Project Name: Identifying and Utilizing Inexpensive Technologies to Manage Patient Populations Primary Investigator(s): Amy Y. Landry, University of Alabama at Birmingham Description: The ACA is producing a shift in focus for many large health systems from a sickness model to a wellness model. Caring for patients outside the confines of a hospital and beyond acute episodes of care is proving to be challenging for organizations built upon incentive systems that traditionally reward more sickness and more interventions. In learning to communicate and better manage patients beyond the hospital walls, the utilization of inexpensive technologies (e.g. apps) to manage patient populations shows some promise. The main objectives of the project are to 1) identify mobile apps that can be utilized in managing specific patient populations; 2) educate clinical staff and physicians on the appropriate indications for each app; 3) “prescribe” one of the selected apps to a sample of patients, and 4) evaluate the effects of the technology on patient engagement and health management. Understanding how inexpensive technology can be used to manage chronically ill patients is the goal of this research, so the patient population we are targeting includes diabetic and pre-diabetic patients. The incidence of diabetes is higher in states located in the Deep South region, where our industry partners are located, compared to the rest of the country. Engaging patients to participate in the management of their own health is very important with the diagnosis of diabetes. After gaining approval from UAB’s IRB, we intend on partnering with groups of primary care physicians to identify patients for participation in our study. Experimental Plan: We will use a three-pronged approach to achieve our research objectives. Our first objective involves the identification of mobile apps that are effective in engaging diabetic and prediabetic patients in managing their health. In the second phase of our study, we will work with the appropriate clinical staff at the primary care offices engaged in our study. In the final phase of our study, we will evaluate the effects of the technology on patient engagement and self-management. How this is different than related research: A 2012 report by the Pew Research Center suggests that 85% of U.S. adults own a cell phone, and 53% of those own smartphones. Smartphone owners routinely gather health related information on their phones, and this type of health information seeking behavior is increased in individuals with some sort of medical crisis or condition. Almost 20% of smart phone owners have at least one health app on their phone; however, no research exists to assess the way that using such widespread technology influences health behaviors or outcomes. Simple, inexpensive technology has great potential to improve disease management of chronically ill patient populations. Milestones & Deliverables: In the first quarter of the study year, apps will be identified through qualitative assessment and clinicians/physicians will be educated on the selected apps. In the 2 nd and 3rd quarters of the study, clinicians/physicians will “prescribe” these apps to appropriate patients. In the final quarter, we will survey participating clinicians and patients on their perceptions of patient engagement and improvements in clinical indicators. We will produce a report outlining the availability and efficacy of the selected apps. Findings will include ease of implementation of these apps as population health management tools, and the influence their utilization had on clinician and patient perceptions of engagement and self-health management. Potential member benefits: Learning to effectively use inexpensive technologies to manage chronically ill patients has the potential to make a large impact on population medicine strategies employed by larger systems. Empowering patients to use mobile health apps is a cost-effective way to manage the care of chronically ill patients that can be easily and rapidly disseminated. As reimbursement mechanisms continue to shift away from a volume-driven to a value-driven system, finding inexpensive ways to keep patients healthy will be critical. Estimated Cost: $50,000 Project Overview and Description • Overview: Identifying and Utilizing Inexpensive Technologies to Manage Patient Populations Amy Yarbrough Landry, PhD University of Alabama in Birmingham Approach • We will identify three mobile apps that can be used to engage diabetic/pre‐diabetic patients in managing their health • We will work with clinical staff in selected primary care offices to identify patients and “prescribe” apps • We will evaluate the effects of the technology on patient engagement and self‐management – Analysis of how inexpensive, easily accessible technology (e.g. apps) can be used to promote patient self‐management of chronic conditions • Description: – Identify mobile apps that can be utilized in managing specific patient populations – Educate clinicians/physicians on appropriate indications for apps – “Prescribe” selected apps to a sample of patients – Evaluate the effects of the technology on patient engagement and health management Project Deliverables / Benefits • Project Deliverables – Q 1: IRB approval; Qualitative assessment of apps for diabetics/pre‐diabetics; Clinician/physician education on apps – Q 2 and 3: Apps are “prescribed” for use to diabetic/pre‐diabetic patients – Q 4: Survey of patient perceptions of engagement and self‐ health management; report preparation; manuscript preparation • Benefits – Understanding how to effectively use simple, easily accessible technology to improve the self‐management of chronic disease patients will benefit health care providers by offering them an inexpensive disease management tool NSF IUCRC PROJECT REPORT 4.3 Project Name: Robust and adaptive optimal healthcare staff scheduling Primary Investigator(s): Sibel Sonuc & James Benneyan, Northeastern University Description: This project will develop and test several robust optimization approaches to staff scheduling, using OR nurse staffing as an initial test-bed. While staff scheduling and other optimization models are seemingly useful in healthcare, a ubiquitous problem is either accounting for uncertainties in the optimization algorithms and/or developing staff schedules that are robust to uncontrollable exogenous events. In the OR, examples include uncertainties in the number of surgeries that ultimately will be scheduled, their times and days of week, their durations, and changes in staff schedules such as due to sickness or other reasons. Since staff schedules are set several weeks or months in advance, the result typically is schedules with excessive overtime, case delays, last minute call-ins at greater costs and less than ideal skill matches, and safety concerns. Since operating rooms and staffing both represent significant portion of hospital costs, we focus our initial work on OR nurse scheduling. Results and the developed general methodologic approach also may be generalizable to many other contexts in healthcare, such as outpatient scheduling, capacity planning, inventory management, and others. Experimental Plan: In this study, we will 1) develop understanding of OR nurse staffing logic at multiple CHOT members and develop classic optimization models of these, 2) extend these models to probabilistic and robust frameworks as appropriate (this is the general manner by which robust models are developed, i.e., extending deterministic models), 3) apply the developed models to at least one test bed (ideally 2 or more) and compare/contrast results and benefits off-line, numerically and/or via computer simulation, 4) adapt optimal results into an actual pilot application and compare beforeafter benefits in actual practice, as well as to those suggested by the models, and 5) disseminate our findings. How this is different than related research: Almost all staff scheduling (and optimization models more generally) in healthcare are classic deterministic models, with fewer venturing into stochastic programming and recourse types of formulations. These models assume for the most part that events are known and deterministic, whereas over the past several years (mostly outside of healthcare) significant advances have been made in robust optimization methods and applications. We therefore believe there is an opportunity to contribute both to staff scheduling problems of CHOT members as well as to more general methods research in the use of operations research in healthcare. Milestones & Deliverables: 1) Documentation and analysis of OR nurse staffing logic at multiple CHOT members, 2) Development of deterministic and probabilistic robust optimization models of these, 3) Application of the developed models to ≥ 2 test beds and evaluation of results off-line, 4) Adaption of optimal results into an actual pilot application and comparison of before-after benefits, 5) Completion of report and journal-ready paper on findings. Potential member benefits: Cost savings from more robust OR nurse staff schedules, with less overtime, fewer last minute scrambles, and better skill matches, understanding of potential for similar approaches to other optimization problems, and better flow and fewer case delays. Estimated Cost: $42,500 Project Overview and Description Rationale Robust and Adaptive Optimal Healthcare Staff Scheduling Sibel Sonuc, PhD, James Benneyan, PhD, Northeastern University • OR nurse schedules are difficult to create due to surgery uncertainties and complexity • Develop a robust system design to further advance theoretical field of robust optimization and apply to healthcare • Almost all staff scheduling in healthcare is based on deterministic models with few exceptions venturing into stochastic programming Relevance • ORs are expensive and schedules are planned weeks in advance, so accounting for surgery variability and unexpected staff absences is difficult • Better skillset match, increased safety and shift preference Approach • Develop deterministic, probabilistic, and robust operations research models to design a robust staffing schedule • Provide optimal staff schedules even when the resulting scenario deviates from the predicted best case • Timeline: – Phase 1: Develop understanding of OR nurse staffing logic at multiple CHOT members and develop classic optimization models – Phase 2: Show feasibility of robust models and test solution in 2 or more test beds with an analysis of before‐and‐after benefits • Applied and theoretical contribution to scheduling for CHOT members Project Deliverables / Benefits Milestones / Deliverables Analysis of OR nurse staffing logic at multiple CHOT members Development of optimization models Apply developed models to ≥2 test beds and evaluate results off‐line Adapt optimal results into an actual pilot application and compare before‐ after benefits • CHOT report, member webinar, and journal‐ready publication(s) on findings • • • • Potential Member Benefits • Cost savings from more robust OR nurse staff schedules, with less overtime, fewer last minute scrambles, and better skill matches • Potential to adapt similar approaches to other optimization problems • Better flow and fewer case delays NSF IUCRC PROJECT PROPOSAL 4.4 Project Name: Challenges in telemedicine – a systematic review and engagement with rural communities Primary Investigator: Eva K Lee, Jean Kang, & Isabella Carbonell, Georgia Institute of Technology Description: Tele-health is the use of electronic information and telecommunications technologies to support long-distance clinical health care, patient and professional health-related education, public health and health administration. Telehealth could be as simple as two health professionals discussing a case over the telephone or as sophisticated as doing robotic surgery between facilities at different ends of the globe. It encompasses preventive, promotive and curative aspects. Within the clinical usage, it has been widely used in diagnosing via medical images, conferencing between patient and healthcare provider for assessments and history taking; exchanging health services or education live; diagnosing and disease managing via medical data; advice on prevention of diseases and promotion of good health by patient monitoring and followup; and health advice by telephone in emergent cases (tele-triage). In multiple UK clinical trials, it has been reported that its usage has led to a 45% reduction in mortality rates, 20% reduction in emergency admissions, 15% reduction in A&E visits, 14% reduction in elective admissions, 14% reduction in bed days, 8% reduction in tariff costs, and 95% cost reduction for patients suffering from infertility. Although these studies have demonstrated a positive impact from the use of telehealth and remote patient monitoring, there are dissenting studies. A 2012 US study of 205 elderly patients with a high risk of hospitalization showed a significant increase in the mortality rate over 12 months, with rates over 12 months for the telemonitoring group at 14.7%, compared with 3.9% for the usual care group. Compounding the challenges on evidence of positive outcome are the federal requirements of efficiency, economy and quality of care for reimbursement. Since its introduction almost 20 years ago, the adoption of telemedicine and the level of engagement and services provided across healthcare facilities remain uneven and far from optimal. There is enormous opportunity to expand the service so as to provide more timely communication and consultation to patients, reduce the face-to-face demand, and the cost of delivery. Experimental Plan: This is a collaborative study (Grady, Morehouse, Northside, CHOA, and the rural Georgia community). We will conduct hospital visits and interviews to analyze the scope of telehealth services across the region. We will interview patient groups to learn their preference and experience in tele-health. Next, we will work with the Georgia rural community to better understand their healthcare service needs and how they are being met. The team will attempt to align and optimize potential demands and resource and provide a report on the findings. Initial needs and capacity assessment for potential change/expansion opportunities will be reported to healthcare organizations for closer review regarding investment. Outcome measures will be documented to understand impacts on change. How this is different than related research: This study attempts to combine social-economic and demographics demands, hospital resources, and evidence of tele-health to assess the value and implementation challenges of tele-health for quality and effective healthcare delivery. Milestones & Deliverables: This study will produce 1) a summary of types of tele-health services, the associated hospital resource usage, and the type of patients served; 2) patient preference on telehealth services and existing gap in meeting the demand; 3) resource usage and capacity assessment, demand gaps, and reimbursement logistics; and 4) opportunities for expansion of tele-health service, implementation assessment, and demand alignment. Potential Member Benefits: improve efficiency of care; improve timeliness of care, reduce waste, serve more needed patients; improve demand-resource alignment, reduce prolonged LOS, and improve surge capability (in the event of pandemic or disaster response). From the patient standpoint, it offers access to care, timeliness of care, reduces unnecessary face-to-face visits, and reduces costs. This work also has the potential to reduce healthcare delivery disparities. Estimated Cost: $40,000 Project Overview and Description • • • Challenges in Telemedicine – A Systematic Review and Engagement with Rural Communities Georgia: Strength and Weaknesses • • Eva K Lee, Isabella Carbonell, Jean Kang, Jihwan Oh, Sang Wook Park • Georgia Institute of Technology • Approach Collaborative study: Grady, Morehouse, Northside, CHOA, and the rural Georgia community 1. Interview/visit to analyze the scope of telehealth services 2. interview patient groups to learn their preference and experience 3. Work with Georgia rural community to explore their healthcare service needs, potential gaps, and opportunities 4. Align and optimize potential demands and resource 5. Establish initial needs and capacity assessment for potential change/expansion opportunities Georgia Tech Copyright Material Tele‐health: cost‐effective means to provide access/timeliness of care Adoption and engagement across healthcare facilities remain uneven Enormous opportunity to expand the service to provide more timely communication and consultation to patients, reduce the face‐to‐face demand, and the cost of delivery. • 350+ locations with 200+ specialists, 600 healthcare partners/providers All 224 public health sites, and over half of GA’s hospitals are equipped for telehealth, Ranked 16th in the nation. 52% GA physicians are within 5 areas that serve just 38 percent state’s population. Ranks 40th in the nation w.r.t. adequate distribution of doctors by specialty and geographic location. Project Deliverables / Benefits Deliverables • Summary of tele‐health services, users, and associated hospital resource usage • Patient Preference and existing service gap, resource usage and capacity assessment, and reimbursement logistics • Rural health: Opportunities for expansion of tele‐health service, implementation assessment, and demand alignment. Potential Benefits • Improve access to care, timeliness of care • Reduces unnecessary face‐to‐face visits, reduces costs. • Reduce healthcare delivery disparities, serve more needed patients • improve demand‐resource alignment • Improve surge capability (in the event of pandemic or disaster response) • Generalizable model for other states/regions 5. Macro/Policy Cluster Chair - Jim Benneyan, Northeastern University 5.1 Understanding Group Practice Trends in 2015 and into the Future RFP: Organization & System Design Bita A. Kash & Sean Gregory, Texas A&M University 5.2 Surgical Care Trends and the Future Role of Hospitals RFP: Organization & System Design Bita A. Kash & Michael A. Morrisey, Texas A&M University 5.3 Modeling ACOs as macro systems of care RFP: Transitions of Care Tannaz Mahootchi & James Benneyan, Northeastern University 5.4 Patient Flow in Children’s Hospitals: Research-Informed Strategies to Influence Discharge Time and Capacity RFP: Organization & System Design Bita A. Kash, Texas A&M University 5.5 Translating UBRICA’s Vision for Kenya to Evidence-based Strategy and Funding Jill Zarestky, Lesley Tomaszewski & Patience Appiah, Texas A&M University 5.6 Hospital acquired conditions - systematic and adaptive approach RFP: Transitions of Care Eva K Lee & Prashant Tailor, Georgia Institute of Technology NSF IUCRC PROJECT PROPOSAL 5.1 Project Name: Understanding Group Practice Trends, Physician Burnout, and Engagement Primary Investigator(s): Bita A. Kash & Sean Gregory, Texas A&M Description: The proportion of self-employed physicians in group practices fell from 35% to 28% between 1983 and 1994 while the proportion of physicians practicing as employees rose from 24% to 42%. Today’s trends might be similar, driven by slightly different market dynamics and enablers, but are not well documented and understood yet. These changing employment structures may affect physician engagement and burnout. Thus, the overall objective of the project is to acquire a better understanding of group practice trends and the resulting effect on engagement and burnout, divided into the following three aims: 1) To profile the current physician group market in terms of group practice development and physician engagement for anesthesia professionals and two to three other specialties selected by Studer Group using secondary data, 2) To interpret these trends using qualitative content analysis and information on upcoming job openings in these specialties, 3) To recommend a general plan for the ASA and Studer Group to gather and maintain, as an ongoing activity, key information regarding anesthesia- related and other select physician group practice development and trends. Experimental Plan: 1. Compilation and interpretation of information currently available in peer-reviewed and gray literature (this includes targeted phone calls and e-mails directed to relevant associations); 2. Identification of available secondary data and secondary data analysis; and a comparative analysis of ownership/practice transactions for anesthesia and two to three specialties as defined by Studer Group; 3. Primary data collection related to job openings (over a period of 4 months in spring 2016) to complement the structural analysis and better predict future trends. How this is different than related research: Currently there is no consensus about physician practice trends and the future outlook of the various specialties, and very little data on how these changes affect physician engagement. Study results and data interpretations are often conflicting and changing constantly. This study will take a segmented (by specialty), mixed methods approach relying on multiple data sources to provide a better understanding of the complexities of group practice trends today. Milestones & Deliverables: June to August: Comprehensive review of recent studies and physician surveys conducted by organizations and physician associations such as the AMA, the Physician Foundation, and the Medical Group Management Association (MGMA). Identification of multiple sources of data for the last 5 to 8 years. September to October: Compilation of existing study findings. Agreement reached with the ASA and Studer Group contacts on final set of multiple datasets, including publicly available data and data supplied by the ASA. Data cleaning and merging. November to January: Secondary data analysis using trend analysis methodology January to April: Primary data collection on physician job listings for the following specialties: anesthesiology, radiology, and pathology, and two more specialties as defined by Studer Group. Continue refining secondary data analysis. April to May: Report writing and refining of data analysis to produce relevant results. Report will include 1) results from analysis of recent studies and physician surveys (a compilation), 2) result from secondary data analysis identifying trends, and 3) results from primary data collection on new job openings and recommendations for ASA and Student Group. Potential member benefits: Information about current physician practice trends are difficult to interpret due to the extremely dynamic nature of the physician group practice marketplace and its effect on physician engagement. This mixed methods approach provides a more comprehensive understanding of current trends and uses multiple sources of data to enhance trend analysis and prediction modeling. The project includes an ongoing plan to maintain and update this important information set for the ASA and Studer Group. Estimated Cost: $100,000 Project Overview and Description Understanding Group Practice Trends, Physician Burnout and Engagement Bita A. Kash, PhD, MBA, FACHE Sean Gregory, PhD, MBA, MS Kayla Cline, MS, CPA Texas A&M University Project Deliverables / Benefits Approach • We will use a mixed methods approach understand group practice trends and compare physician burnout by employment type: – Compile information currently available in the academic and non‐academic literature and physician associations – Analysis of publicly available data by physician specialty and employment type (employed vs. self‐employed) – Primary data collection on job openings in the spring of 2016 to complement secondary data analysis and predict future trends • Timeline: Months 1‐3 • Review existing knowledge Months 4‐5 • Compile findings & select data Months 6‐7 • Secondary data analysis Rationale • In the late 1980s and early 1990s, physicians moved from group practices to hospital employment • Some assert that the same trend is happening today, but this has not yet been documented and understood • This project will attempt to understand Group practice trends, physician burnout, engagement and satisfaction by employment type. Relevance • Identifying trends in physician employment across multiple specialties is valuable for both large hospital systems and provider organizations for planning and decision making • Understanding the drivers of changing physician employment practices and effects of physician satisfaction can help organization leaders assess and address physician needs Months 8‐10 • Collection of job listing data Months 11‐12 • Report compilation • Deliverables will include – Information about current group practice trends and physician burnout in multiple specialties using secondary data – Interpretation of causes of current trends using qualitative data – Prediction of future trends using primary data on upcoming job openings • These findings will provide unique, up‐to‐date information on where physicians are working and variations in burnout and satisfaction based on employment type, which will be helpful to hospitals who employ or work with self‐employed physicians to improve resource allocation and staff planning NSF IUCRC PROJECT PROPOSAL 5.2 Project Name: Surgical Care Trends and the Future Role of Hospitals Primary Investigators: Bita A. Kash & Michael A. Morrisey, Texas A&M University Description: The emergence of ambulatory surgical centers (ASCs) since the 1970s has generally been driven by physician groups who saw an opportunity in the surgical care market. While physicians have taken the lead in self organizing and moving away from the hospital setting and establishing competing ASCs, over the last 10 years, there is now a parallel trend of physician employment by hospitals and health systems. The ASC trend is often viewed as a positive development that aligns with reducing the cost of surgical care while providing today’s price-sensitive consumer a better value. However, this positive market trend does require hospitals to rethink surgical care services and their positioning in their specific markets. A 2012 study by the American Medical Association (AMA) found that 53% of physicians still remain full or part owners of their practice. Practice ownership stays even higher among specialty surgeons, anesthesiologists and radiologists according to the 2012 AMA report. Experimental Plan: 1) compilation and interpretation of information currently available in peerreviewed and gray literature (this includes targeted phone calls and e-mails directed to relevant associations); 2) identification of publicly available secondary data on surgical care volume, outcomes, cost, and settings, followed by secondary data analysis to develop relevant predictions models; 3) primary data collection through interviews with key informants identified in step 1 to help with hospital strategy formulation. How this is different than related research: Currently there is no consensus about ASC and physician ownership trends and the future outlook of the various surgical specialties. Study results and data interpretations are often conflicting and changing constantly. In contrast, this study will take a segmented (i.e., by specialty and by region), mixed methods approach relying on multiple data sources to provide a more comprehensive understanding of the complexities of ASC ownership and pricing trends today. Study results will result in research-informed strategy development for the hospital sector. Milestones & Deliverables: June to August: Comprehensive review of recent studies and physician surveys conducted by organizations and associations such as the AMA and the ASCA. Identification of multiple sources of data, including the Physician Compare National Data File through the CMS and AHA’s Annual Survey database, for information on the last five years. IRB approval through Texas A&M University. September to October: Compilation of existing study findings. Agreement reached with INTEGRIS Health contacts on final set of multiple datasets, including publicly available data and data supplied by INTEGRIS. Data cleaning and merging. November to December: Secondary data analysis using trend analysis and econometric modeling methodology (dependent on data availability). January to March: Primary data collection on relevant surgical specialties from key informants identified as part of the comprehensive literature and information searches: Continue refining secondary data analysis. April to May: Report writing and refining of data analysis to produce prediction models and relevant results. Report will include 1) results from analysis of recent studies and surveys published in peerreviewed and gray literature (a compilation), 2) result from secondary data analysis identifying trends and prediction models, and 3) results from primary data collection informing strategy for the hospital sector. Potential member benefits: Information about ASC trends are difficult to interpret and predict due to the extremely dynamic nature of the physician (surgical) group practice market place today. Results from this analysis will inform future strategy for the hospital sector by regional market characteristics and surgical specialty. Estimated Cost: $50,000 Project Overview and Description Rationale Surgical Care Trends and the Future Role of Hospitals Bita A. Kash, PhD, MBA, FACHE Michael A. Morrisey, PhD • The emergence of ambulatory surgical centers (ASCs) since the 1970s has generally been driven by physician groups • The ASC trend is often viewed as a positive development that aligns with reducing the cost of surgical care, while providing today’s price‐ sensitive consumer a better value • 53% of physicians still remain full or part owners of their practice • Practice ownership is high among specialty surgeons, anesthesiologists and radiologists, according to a 2012 study by the American Medical Association (AMA) Relevance Texas A&M University • It is important for hospitals and health systems to acquire a better understanding of today’s trends in preparation for future models of surgical practice and payment Project Deliverables / Benefits Approach The research approach will: 1) Use publicly available data to profile and describe trends in surgical care organization and physician (i.e., surgical specialties) group practices 2) Include a comprehensive market trend analysis, a review of the literature and relevant websites, as well as the analysis of various datasets and information sources, including the Physician Compare National Data File through CMS and AHA’s Annual Survey Database 3) Utilize a mixed methods approach – Compilation and interpretation of information currently available in peer‐reviewed and gray literature – Identification of publicly available secondary data on surgical care volume, outcomes, cost, and settings, followed by secondary data analysis to develop relevant predictive models – Primary data collection through interviews with key informants to help with hospital strategy formulation Timeline/Deliverables • • • • • Months 1 to 3: Comprehensive literature review of recent studies, identification of multiple sources of data, including the Physician Compare National Data File through CMS and AHA’s Annual Survey database for information during the last five years, and IRB approval. Months 4 to 6: Compilation of existing study findings, final set of multiple datasets, including publicly available data and data supplied by INTEGRIS Health. Months 7 to 9: Secondary data analysis using trend analysis and econometric modeling methodology. Months 10 to 12: Primary data collection on relevant surgical specialties from key informants identified, in part with the comprehensive literature review. Refinement of secondary data analysis. Months 13 to 15: Final report of data analysis, prediction models and results. Potential Member Benefits • • A more comprehensive understanding of current trends and uses of multiple sources of data to enhance trend analysis and prediction modeling. Results from this analysis will inform future strategy for the hospital sector by regional market characteristics and surgical specialty. NSF IUCRC PROJECT PROPOSAL 5.3 Project Name: Modeling ACOs as macro integrated systems of care Primary Investigator(s): Tannaz Mahootchi & James Benneyan, Northeastern University Description: This project will develop and test a system-wide analytic model of patient, information, and personnel flow across all aspects of accountable care organization and other loosely coupled healthcare affiliations. As part of current healthcare reform trends, financial and care considerations are leading to numerous health system mergers, business relationships, and couplings between healthcare organizations “across the continuum” of care and health – e.g., inpatient, specialty care, primary care, skilled nursing facilities, home – resulting in a resurgence of interest to improve the care and health maintenance of patients as they flow across these larger health/healthcare ecosystems. Very commonly, quality improvement projects are trying to improve these systems at the boundaries between the components (e.g., care transitions, care continuity, integrated primary and specialty care, etc.). This project aims to model these longitudinal and inter-organizational processes at macro level and illustrate the use of these models to improve the overall system. Experimental Plan: There are two specific phases and objectives to this project. The first objective (phase 1, months 1-6) is to develop and validate a macro model or set of models of the longitudinal and inter-organizational processes (most likely Markov, Erlang-R reentrant queuing, and simulation models). The second objective (phase 2, months 7-12) is to demonstrate and validate the use of these models as a decision support platform to help improve key issues facing most ACOs today, using system-wide capacity and care integration as initial test-bed problems. How this is different than related research: As part of healthcare reform and as ACOs become more commonplace, more interest and attention is starting to focus on transitions between various aspects of the overall system of care, as a system, whereas in the past they often have been viewed (and optimized) as fairly disjoint entities. While an increasing topic of ACO administration, little work has been done to optimize (that is, using engineering tools) ACOs as a system. Milestones & Deliverables: 1) Development and iteratively validation of each model using 1-3 CHOT members (Maine Health, Partners Health, others) (phase-1, months 1-6), 2) Use of the developed models to inform capacity and care integration analysis and improvements, with an additional objective of demonstrating how such models can be used for analysis, decision support, and optimization, 3) Development of CHOT report, member webinar, and journal-ready publication(s) to disseminate results. Potential member benefits: Improved care at lower cost of patients, especially those under risksharing agreements and reduced utilization outside of the ACO due to sub-optimal access. Estimated Cost: $42,500 Project Overview and Description Rationale Modeling ACOs as Macro Integrated Systems of Care Tannaz Mahootchi, PhD James Benneyan, PhD Northeastern University Approach • Develop and test a system‐wide analytic model of patient, information, and personnel flow across all aspects of an integrated system • Demonstrate how such models can be used for analysis, decision support, and optimization • Little work has been done using engineering tools to optimize ACOs as a system Relevance • Financial and care considerations are leading to health system mergers, resulting in interest to improve patient care across these larger systems • Impact on cost, access, and care coordination Hospital Front Door PACU OR and Procedure Rooms CCU ED DC’ed, Transferred, Expired IP Project Deliverables / Benefits Phase 1 (months 1‐6) Milestones / Deliverables • Develop models of the longitudinal and inter‐organizational processes, including but not limited to • Development and iteratively validation of model(s) using 1‐3 CHOT members • Use developed models to inform capacity and care integration analysis and improvements • Demonstrate how such models can be used for analysis, decision support, and optimization • CHOT report, member webinar, and journal‐ready publication(s) to disseminate findings – Markov models – Erlang‐R queuing networks – Simulation models • Validate developed models Phase 2 (months 7‐12) • Demonstrate and validate use of these models as a decision support platform to help improve key issues facing most ACOs today • Use system‐wide capacity and care integration as initial test‐bed problems Potential Member Benefits • Improved care at lower cost of patients, such as those under risk‐ sharing agreements • Reduced utilization outside of the ACO due to sub‐optimal access NSF IUCRC PROJECT PROPOSAL 5.4 Project Name: Patient Flow in Children’s Hospitals: Research-Informed Strategies to Influence Discharge Time and Capacity Primary Investigator(s): Bita A. Kash, Texas A&M University Description: American hospitals and health systems are pursuing strategies to improve and optimize patient flow through modeling, redesign and influencing the arrival and discharge of patients. Various operational models and case studies to address effective and efficient hospital resource management have been published in peer-reviewed literature. Despite this, hospital administrators struggle with capacity challenges. Innovative and evidence-based models of practice are needed for specific hospital types and service categories to improve hospital bed capacity and early discharge in light of recent changes in payment methodology and focus on reducing length of Stay (LOS). This study will focus mainly on Acute Care in both community hospitals and academic centers and aim to identify innovative models to manage operating rooms (ORs) and the lower acuity cases. Experimental Plan: The research approach will address the following three aims: 1. Conduct an iterative scientific and systematic literature review within the six months of the project, while supplying Texas Children’s Hospital and other interested CHOT members with monthly updates of the findings resulting in refinement of search criteria; 2. Identify and study 5 to 8 innovative operational models for OR and Acute Care capacity (Models of Practice) relevant to the children’s hospital setting; 3. Conduct 2 to 3 physician leader focus groups at the CHOT industry member site to examine operational model fit and recommend targeted implementation strategies. These professional key informants will be asked to serve as advisers to the CHOT research team. How This Is Different from Related Research: This research is focused on assisting industry members with research-informed decision making regarding patient flow. This study will provide targeted evidence-based strategies to reduce LOS and accelerate discharge for Acute Care and find innovative models for OR scheduling to ensure expedited turnaround specific to children’s hospitals. Milestones & Deliverables: Months 1 to 6: Literature review, interviews with key informants, and identification of Models of Practice. An initial first-order model (FOM) of the innovative operational model for ICU and OR concept will be developed in the first two month and discussed with 5 to 8 professional advisors during the third and fourth months. Months 6 to 8: Refinement of Models of Practice studied and identification and description of additional operational characteristics. Complete interviews with up to 8 key informants in models of interest using the snowball sampling approach. A report of the qualitative analysis of these results will be presented to the sponsor by March 2016. Months 9 to 12: Focus groups with physicians, final report development, and review article manuscript draft for publication in peer-reviewed journal. Texas Children’s Hospital will have a chance to test implementation opportunities based on physician feedback. Final report will include recommended implementation strategies based on evidence-based Models of Practice, ease of implementation (barriers and challenges) based on key informant interviews, and site-specific considerations based on physician focus group results. Potential Member Benefits: A focused strategy targeting reductions in discharge time for OR and ICU patients by hospital type and nature of intervention. Evidence-based models of practice will be presented such that they are easily translated into implementation strategies and action plans. Estimated Cost: $50,000 Project Overview and Description Rationale Patient Flow In Children’s Hospitals: Research‐Informed Strategies to Influence Discharge Time and Capacity Bita A. Kash, PhD, MBA, FACHE Texas A&M University • American hospitals and health systems are pursuing strategies to improve and optimize patient flow through modeling, redesign and influencing the arrival and discharge of patients • Innovative and evidence‐based models of practice are needed for specific hospital types and service categories to improve hospital bed capacity and early discharge in light of recent changes in payment methodology • Focus on reducing Length of Stay (LOS) and acute care in both community hospitals and academic centers • Aim to identify innovative models to manage operating rooms (ORs) and the lower acuity cases Relevance • Focus on assisting industry members with research‐informed decision making regarding patient flow • Provide targeted evidence‐based strategies to reduce LOS and accelerate discharge for acute care • Find innovative models for OR scheduling to ensure expedited turnaround specific to children’s hospitals Approach Project Deliverables / Benefits The research approach will address the following three aims: Deliverables 1) • Months 1 to 6: An initial first‐order model (FOM) of the innovative operational model for ICU and OR concept will be developed in the first two months and discussed with 5 to 8 professional advisors during the third and fourth months. • Months 6 to 8: Complete interviews with up to 8 key informants in models of interest using the snowball sampling approach. A report of the qualitative analysis of these results will be presented to the sponsor by March 2016. • Months 9 to 12: Texas Children’s Hospital will have a chance to test implementation opportunities based on physician feedback. Final report will include recommended implementation strategies based on evidence‐based Models of Practice. 2) 3) Conduct an iterative scientific and systematic literature review within six months of the project, while supplying Texas Children’s Hospital and other interested CHOT members with monthly updates of the findings resulting in refinement of search criteria. Identify and study 5 to 8 innovative operational models for OR and acute care capacity (Models of Practice) relevant to the children’s hospital setting. Conduct 2 to 3 physician leader focus groups at the CHOT industry member site to examine operational model fit and recommend targeted implementation strategies. Timeline • • • Months 1 to 6: Literature review, interviews with key informants, and identification of Models of Practice. Months 6 to 8: Refinement of Models of Practice studied and identification and description of additional operational characteristics. Months 9 to 12: Focus groups with physicians, final report development, and review article manuscript draft for publication in peer‐reviewed journal. Potential Member Benefits • A focused strategy targeting reductions in discharge time for OR and ICU patients by hospital type. • Evidence‐based models of practice will be presented, such that they are easily translated into implementation strategies and action plans. NSF IUCRC PROJECT PROPOSAL 5.5 Project Name: Translating UBRICA’s Vision for Kenya to Evidence-based Strategy and Funding Primary Investigator(s): Jill Zarestky, Lesley Tomaszewski & Patience Appiah, Texas A&M University Description: Ustawi Biomedical Research Innovation and Industrial Centers of Africa (UBRICA) is planning to develop a socio-economic development and human health project in the Great Rift Valley of Kenya. This planned project is referred to as UBRICA ONE and involves the use of 4,000 acres of land for development as a health sciences center. The ultimate goal is to transform UBRICA ONE into a home for world-class medical facilities, state of the science research, an industrial park, and residential and recreational facilities. The vision of UBRICA is to become a leading company at creating sustained conversion of knowledge for promoting health and human development in the frontier markets. This study will be focused on critically evaluating various socio-economic development and human health improvement theories and frameworks that relate to UBRICA’s vision and UBRICA ONE’s goals to develop a final grant proposal such as the NSF Grant Opportunities for Academic Liaison with Industry (GOALI) or other NSF grants or UN grant opportunities Experimental Plan: 1. A comprehensive, critical literature review on theories, frameworks and practices on international development models on health, social advancement specific to Kenya, and international development 2. Needs Assessment (secondary data): epidemiology and needs assessment of the Great Rift Valley region of Kenya 3. Needs Assessment (primary data collection and mixed methods): this will include stakeholder interviews with key community leaders, Kenyan content experts, and other key contacts identified by steps one and two. 4. Strategy and Proposal development will be inter-disciplinary as it might include human health and animal health factors. Concepts that will be considered and developed in detail for final proposal include: a) environments that produce health, b) inclusion of local people and cultures, c) corporate and social responsibility, d) industry impact on community and health How this is different than related research: Evaluation and strategy development for environments that produce health specific to Kenya and the region of the Great Rift Valley. The goal of the framework for action will be development that creates social advancement for local Kenyans. The study will be informed by knowledge embedded on the country and region. Develop a framework for action for health production that is part of the social and ecological system. Milestones & Deliverables: May 2015: Kick-off meeting with key UBRICA leaders and project goal refinement August 2015: IRB approval through TAMU September 2015: Initial literature review results, review with key UBRICA leaders and refinement of literature search December 2015: Literature review is completed; Needs Assessment (secondary data) is completed; key informants are identified March 2016: Key informant and stakeholder interviews completed May 2016: Comprehensive data analysis complete, grant proposal draft is the final deliverable Potential member benefits: NSF Grant Opportunities for Academic Liaison with Industry (GOALI) or other proposal, such as through other NSF grants or UN grant opportunities, developed by May 2016 Estimated Cost: $50,000 Project Overview and Description Translating UBRICA’s Vision For Kenya To Evidence‐Based Strategy and Funding Jill Zarestky, PhD Lesley Tomaszewski, PhD Patience Appiah Texas A&M University Rationale • Ustawi Biomedical Research Innovation and Industrial Centers of Kenya (UBRICA) are planning to develop a socio‐ economic development and human health project in Kenya’s Great Rift Valley • The ultimate goal of UBRICA ONE is to become the home to world‐class medical facilities, state of the art research, industrial parks, and residential and recreational facilities • Create a sustained conversion of knowledge for promoting health and human development in the frontier markets Relevance • To provide intellectual support and experience in order to promote and enhance human and animal health Approach Project Deliverables / Benefits The research approach will consist of: • A comprehensive literature review on theories, modules, frameworks, practices, social improvements, and international developments specific to Kenya • Identification of needs assessment, involving key stakeholders and the use of data collection (primary, secondary, and mixed methods) specific to Kenya’s Great Rift Valley • Strategy and proposal development will be inter‐ disciplinary, as it might include human health and animal health factors Project Milestones/Deliverables • May 2015: Kick‐off meeting with key UBRICA leaders and project goal refinement • August 2015: IRB approval through TAMU • September 2015: Initial literature review results, review with key UBRICA leaders and refinement of literature search • December 2015: Completion of literature review; needs assessment (secondary data); and key informants are identified • March 2016: Key informant and stakeholder interviews completed • May 2016: Comprehensive data analysis completed, and deliverable of the final grant proposal draft Potential Member Benefits • NSF Grant Opportunities for Academic Liaison with Industry (GOALI) or additional proposals through other NSF grants or UN grant opportunities, developed by May 2016 NSF IUCRC PROJECT PROPOSAL 5.6 Project Name: Hospital acquired conditions - systematic and adaptive approach Primary Investigator: Eva K Lee & Prashant Tailor, Georgia Institute of Technology Description: According to a 2014 CDC study, about 1 in 25 U.S. patients has at least one infection contracted during the course of hospital care, resulting in about 75,000 deaths during hospitalizations. The most common types of infections are pneumonia (22%), surgical site infections (22%), gastrointestinal infections (17%), urinary tract infections (13%), and bloodstream infections (10%). Among the pediatric population, the highest rates of HACs occur in the Neonatal ICU, Infant neurosurgery, hematology/oncology, neonatal surgery, cardiology/cardiovascular surgery, Pediatric ICU and infant total medicine areas. HAC compromises outcome of patients and ties up unnecessary resources. Challenges include: suboptimal adherence to current prevention recommendations; limitations in surveillance strategies; lack of efficient mechanism for reporting adverse events; inconsistent metrics of measurement; and at times, lack of system-wide research. Most studies are site-specific. The interdependencies and multi-faceted potential personnel and process contribution to HACs make it difficult to pinpoint sources for early detection and intervention. Experimental Plan: This is a collaborative project (CHOA, Grady, Morehouse, Northside, Restore) It will focus on SSI, CLABSI, CAUTI, and HAP in both adult and pediatric populations. We will 1) Collect epidemiology data and EMR data to perform risk and outcome analysis for patients from 2013 and 2014; 2) review, identify, and consolidate national gold standards and best practices; 3) identify HAC sources across various hospital units; 4) establish interdependencies and process and system maps; 5) conduct procedural and compliance benchmark comparison. The design will cover a wide range of stakeholders, including patients, caretakers, healthcare providers, and facility workers. Compliance of process guidelines will be recorded. The team will identify critical infections risk factors from retrospective study of treated patients. We will evaluate current processes to identify bottlenecks and perform system optimization over workflow processes to minimize potential infection and susceptible. How this is different than related research: This study involves multiple hospitals, units, and services, and environmental service, and multiple stakeholders (care givers and providers, patients, and facility/cleaning workers). Terminal cleaning tools and processes will also be observed. Further, pediatric and adult population will both be analyzed and findings will be contrasted. The study’s designed to uncover susceptible areas, process, procedures and behavior over the entire hospital stay period where infection/conditions are acquired with the objective to cultivate a pro-active surveillance system of awareness of infection-prone situations. The team will immense in the day-to-day processes and will map out the multi-faceted inter-dependencies across processes and systems. Multi-site comparison will be performed. Milestones & Deliverables: Deliverables: Complete benchmark of current practice against national standards, best practice, compliance, deficiencies; complete a year-long retrospective risk and outcome analysis; perform time motion study, develop interdependency process maps for multi-unit system, and identify risk factors; identify environmental factors in relationship to HAC; analyze cleaning data and processes to correlate risk factors; promote HAC surveillance aware; develop system simulation optimization models; prioritize root causes for HACs; recommend and develop improved processes, guidelines, and surveillance awareness and checkpoints for HAC reduction; implement recommendations in hospital; and evaluate changes and compare results for further refinement. Potential Member Benefits: Improve quality of care and treatment outcome for patients; reduce unnecessary length of stay and extra medical care; improve provider and patient compliance, improve hospital surveillance; improve hospital resource utilization; improve providers’ morale and confidence; and establish a conducive atmosphere for sustainable process and change transformation where HAC awareness is integral and second nature to service process. Estimated Cost: $70,000 Project Overview and Description Hospital Acquired Conditions – A Systematic Collaborative Approach • One in 25 U.S. patients get/develop • HAC • 1.7 million infections, 75,000 deaths in 2011 (CDC) • $28–$33 billion in excess costs Inconsistent metrics of measurement Limitations in surveillance strategies Non‐adherence to guidelines Approach ‐ Major Tasks 16% 22% 17% 22% SSI Gastrointestinal UTI Bloodstream HAC Inefficient mechanisms for reporting Lack of system‐ wide research • CABG: Reduced LOS & resource usage, reduced SSI incidents. • CLABSI: Predicted high‐risk expired patients for target intervention (prevent death) • OVERALL: High morale, good buy‐ in, improve compliance, training and documentation Project Deliverables / Benefits Systems & Collaborative ‐ 10% 13% Some successes Challenges Georgia Institute of Technology CHOA, Grady, Morehouse, Northside, Restore Working along multiple stakeholders Active learning Pneumonia Others Eva K Lee, Prashant Tailor ‐ Compromises outcomes, ties up resources, increases readmissions Hygiene Cleaning Processes, medication Providers Environment Resources Patients Key Techniques • Mixed model: prospective + retrospective studies • Machine learning, predictive analytics • Systems modeling, optimization, decision analysis • Measure of outcome metrics Deliverables • • • • • • • Benchmark practice against national standards and best practice Highlight compliance and process deficiencies Establish HAC surveillance awareness protocol Identify risk factors, system interdependencies and environmental factors Recommend improved processes, guidelines, and surveillance awareness and checkpoints for HAC reduction Prioritize recommendations for implementation Evaluate improvement, refine analysis Potential Benefits • • • • • • • • Reduce HAC incidents, thus reduce unnecessary LOS and extra medical care Improve quality of care and treatment outcome Improve provider and patient compliance Improve providers’ morale and confidence improve hospital surveillance Improve hospital resource utilization Establish a sustainable process and change transformation Reduce unnecessary penalties 6. Collaborative Research Proposals Chair - Harriet B. Nembhard, Pennsylvania State University 6.1 Replicating a Study of the Efficacy of Quality Improvement Processes in Reducing Hospital Acquired Conditions RFP: Organization & System Design Deirdre McCaughey, Pennsylvania State University & Scott Buchalter, University of Alabama at Birmingham 6.2 Technology Trends and Smart Interventions to Mitigating Patient Risk at Critical Transitions for Total Joint Arthroplasty (TJA) RFP: Transitions of Care Eric R. Swenson, Pennsylvania State University, Kayla M. Cline, Texas A&M University, Harriet B. Nembhard, Pennsylvania State University, & Bita A. Kash, Texas A&M University 6.3 Social Network Analysis: Examining Interactions among Providers at the Network Level RFP: Organization & System Design Nancy M. Borkowski, University of Alabama at Birmingham, Ravi Behar, Florida Atlantic University, & Gulcin Gumus, Florida Atlantic University 6.4 Choosing wisely and reducing practice variation RFP: Human Technology, Organization & System Design James Benneyan, Northeastern University, Susan Haas, Northeastern University, Eva K. Lee, Georgia Institute of Technology, Raghav Srinath, Georgia Institute of Technology, & Haozheng Tian, Georgia Institute of Technology NSF IUCRC PROJECT PROPOSAL 6.1 Project Name: Replicating a Study of the Efficacy of Quality Improvement Processes in Reducing Hospital Acquired Conditions Primary Investigator Deirdre McCaughey & Scott Buchalter, Pennsylvania State University/University of Alabama in Birmingham Description: The Hospital-Acquired Condition (HAC) Reduction Program, implemented by the Centers for Medicare & Medicaid Services (CMS), serves the purpose to achieve better patient outcomes while slowing health care cost growth. Hospital performance under the HAC Reduction Program is determined based on a hospital’s total HAC Score and all hospitals that rank in the worst quartile of HAC scores will receive a payment reduction of one percent for all CMS services. The project seeks to extend the previous work conducted by the principal investigator (McCaughey) examining the efficacy of Lean Six Sigma processes to examine sources of system breakdowns that result in HACs occurring. In conjunction with the University of Alabama at Birmingham (UAB), this project will replicate the previous study conducted in 2014 at Hershey Medical Center (HMC). This replication will aid in validating the first study by examining the Lean Six Sigma process at a new peer academic medical center, UAB. Given that both hospitals are members of the University Hospital Consortium (UHC), the data for this study is uniquely well aligned with the data used in the previous study. Experimental Plan: The project has 3 phases. In the first phase the research team will we will conduct a retrospective review of the HAC events (Patient Safety Indicators) from 2012 to 2014. Through data analysis, we will identify the relevant antecedents to HAC occurrences and the effect of HAC events on incremental hospital costs and length of stay indicators. In the second phase, Lean Six Sigma methodology will be utilized to identify root cause factors contributing to HAC events. We will conduct a rapid improvement event with all stakeholders to document current process, confirm the identified root causes, and develop action plans. In the third phase, we will compare findings from this replication study with our first study (HMC, 2014) to identify the efficacy of the Lean Six Sigma in reducing HAC events. How this is different than related research: Limited research exists that examines the efficacy of Lean Six Sigma processes (e.g. rapid improvement events) on quality improvement in healthcare organizations. Research evidence is needed that explores and identifies how using process improvement methodologies positively impacts HAC performance. Replicating our first study will serve as a unique validation of our initial finding and provide an evidence-based foundation from which this methodology can be utilized to improve hospital HAC performance. Milestones & Deliverables: Acquisition, coding, and cleaning data will occur in months 1-3. Analysis of data and identifying emerging process issues will take place in months 4-5. Rapid improvement event will be held in month 6 and process improvement strategies will be developed in months 6. Months 7-9 will serve to monitor the effect of the rapid improvement event and serve to prepare final reports for presentation to both practitioner and academic audiences as well disseminate finding to participating hospital stakeholders. Potential Member Benefits: The results of this research will assist all hospitals in better utilization of Lean Six Sigma methodologies to examine deficient hospitals processes that result in HACs. Further, replicating our previous research (2014) at a peer hospital will offer a critical evidence-based “next - step” in utilizing the study results to improve patient care thereby fostering greater utilization of this research. Estimated Cost: $20,000 Project Overview and Description Replicating a Study of the Efficacy of Quality Improvement Processes in Reducing Hospital Acquired Conditions Deirdre McCaughey, PhD,MBA Scott Buchalter, MD The Pennsylvania State University & The University of Alabama at Birmingham • In fiscal year 2015, approximately 724 will have their CMS payments reduced by 1% under the Hospital Acquired Condition (HAC) Reduction Program1. • Hospital performance is determined by the total HAC score, and reduction in payment is applicable for hospitals in the worst quartile of scores. • Healthcare systems must address HACs in order to optimize revenues and margin. • Lean Six Sigma (LSS) methodology has the potential to identify root causes of HACs and is a mechanism that involves key stakeholders in the improvement efforts. 1 Approach • Research question: Can Lean Six Sigma methodology be utilized to identify system breakdowns contributing to HACs , as previously found at a peer academic medical center? • Using HAC occurrence data at the University of Alabama at Birmingham (UAB) for 2012‐2014, the project will: – Identify HAC frequency & relevant antecedents to HACs and the impact on hospital costs and length of stay – Use Lean Six Sigma processes will be used to identify root causes of HAC events – Conduct a RIE with all stakeholders of the selected root cause to document the current process, identify pain points, waste, and rework, and develop action plans – Compare findings of this replication study with our first study (CHOT 2014‐2015) to determine the efficacy of using Lean Six Sigma in reducing HACs CMS, 2015 Project Deliverables / Benefits • Data analysis: – Acquisition, cleaning & coding of data – Identification of emerging process issues – Process improvement strategies developed • Action step: – Incorporate LSS methodology to validate & extend findings • Dissemination: – Stakeholder presentations & feedback • Implementation: – Incorporate LSS methodology action items & monitor results • Benefit: – Replication of Lean Six Sigma methodology as a means to reduce HACs NSF IUCRC PROJECT PROPOSAL 6.2 Project Name: Technology Trends and Smart Interventions to Mitigating Patient Risk at Critical Transitions for Total Joint Arthroplasty (TJA) Primary Investigator(s): Nancy M. Borkowski, Ravi Behar, & Gulcin Gumus, University of Alabama in Birmingham/Florida Atlantic University Description: As part of a larger program to incentivize hospitals to shift from a pay-for-service to a payfor-health-outcome model, the Center for Medicare and Medicaid Services (CMS) is penalizing hospitals with above average risk-adjusted readmission rates for TJA. Not all readmissions are preventable, but they all occur after a patient is discharged and outside the hospitals direct control. By identifying patient readmission risk prior to discharge, hospitals can tailor effective intervention strategies to improve patient health outcomes and decrease financial risk. By incorporating past readmissions and EHR data, we can build a predictive model to risk stratify patients. Patient readmission risk will inform care provider decisions on appropriate techniques, technologies and intervention strategies to apply to achieve positive health outcomes. By gaining a better understanding of current trends in perioperative technology development, we can equip perioperative physician leaders with the knowledge and understanding of the complexities of these technological trends, experiences, and future demands and needs. This riskstratification and technology trend information will allow providers to make cost-effective decisions for resource allocation, predict future readmission rates and penalties, and ultimately improve coordination of care. Experimental Plan: This overall objective is twofold: (1) to better understand the causes and key care coordination transitions that lead to readmissions and (2) to identify key trends in technologies that can facilitate care coordination to improve patient outcomes. Penn State University researchers will address the first and Texas A&M researchers will address the second. The first objective will be accomplished as a series of sub-objectives: 1) detailed literature review to determine the factors that lead to readmission, 2) identify the critical transition points, gaps, and barriers in the care delivery process, 3) identify best practices to reduce readmissions and apply them to total joint replacement surgery patients, and 4) use data analytics to develop a model that predicts patient likelihood of readmission from TJA. The second objective will be accomplished via a literature review of current perioperative technology trends in TJA followed by qualitative content analysis of key informant interviews with external experts, system and medical technology vendors, targeted orthopedic surgeons, and additional targeted care team members. How this is different than related research: This project will employ a multifaceted approach employing both engineering and health services research experts; a focus on understanding readmission risk and technology trends from the perspective of the end user (provider); and a general process-improvement emphasis rather than a focus on one particular risk adjustment technique or technology. Milestones & Deliverables: Months 1-3: Literature reviews, IRB, meeting with key informants, and patient shadowing. Months 4-5: Obtain EMR/readmission data; develop key informant interview script and schedule interviews Months 6-9: Build patient profiling model and conduct key informant interviews with 30-40 interviewees Months 10-12: Develop best practices and predictive model; perform content analysis; disseminate results. Potential member benefits: A better understanding of readmissions, the benefit of using care coordination personnel and technology to facilitate improved outcomes, and tools to monitor the outcomes of care coordination. Estimated Cost: $70,000 Project Overview and Description Objective: Improve Patient Health Outcomes from TJA Overview: Technology Trends and Smart Interventions to Mitigate Patient Risk at Critical Transitions in Total Joint Arthroplasty(TJA) Eric R. Swenson, MS Kayla M. Cline, CPA, MS Harriet B. Nembhard, PhD Bita A. Kash, PhD, MBA, FACHE The Pennsylvania State University & Texas A&M University  Traditional hospital care model focused on pre and peri‐op.  Bundled payments and readmission penalties: forcing hospitals to seek better health outcomes.  Opportunity: Annual demand for TJA expected to reach 4M by 2030. Description: Process improvements (barrier reductions/smart interventions) reduce readmissions, decrease financial penalties, and improve health outcomes. Motivation: Reduce the GAP by increasing hospital influence over health outcome. Method: Assessing process and readmission risk, with technology and interventions at deliberate transitions. • Compare cost effectiveness of interventions/best practices • Conduct interviews with key informants (surgeon, vendors, other expert) Process Patient Health Outcome Gap Deliberate Transitions Technology Discharge Low Peri‐op Approach – EMR+ current assessment = risk profile = tailored strategy – Estimate readmission cost by profile Risk High Pre‐op Phase 1: Discovery • Survey TJA best practices in patient transition through D+90days • Form patient advisory group • Conduct patient journey to map TJA process • Review grey/peer reviewed literature to ID peri‐op technologies • Develop key informant interview script • ID key informants using snowball sampling approach Phase 2: Analysis • Data: Hospital’s Financial Responsibility Hospital Control Over Patient Health Outcome Post Op D+30 Interventions Project Deliverables / Benefits • Deliverables: – – – – White paper on best practices in care transitions TJA process map identifying key transitions/gaps and recommended interventions Assessment of resource and technology applications that extend hospital care into patient home Cost benefit analysis of care interventions (cost vs. cost savings) • Benefit to NSF‐CHOT Industry partners: – – – – Improved understanding of causes of readmissions and technologies to improve patient care Increase TJA process effectiveness in era of expanding bundled payment options and penalties Fosters collaboration across colleges: CoE, CoM, CoN, CoHSP Merger of medicine, engineering, and technology to improve patient health outcome • Timeline: – Months 1‐3: Literature reviews, IRB, meeting with key informants, and patient shadowing. – Months 4‐5: Obtain EMR/readmission data; develop key informant interview script and schedule interviews – Months 6‐9: Build patient profiling model and conduct key informant interviews with 30‐40 interviewees – Months 10‐12: Develop best practices and predictive model; perform content analysis; disseminate results NSF IUCRC PROJECT PROPOSAL 6.3 Project Name: Social Network Analysis: Examining Interactions among Providers at the Network Level Primary Investigator(s): Eric R. Swenson, Pennsylvania State University, Kayla M. Cline, Texas A&M University, Harriet B. Nembhard, Pennsylvania State University, & Bita A. Kash, Texas A&M University Description: Providers work predominately alone; however to accomplish their work, physicians and other providers create social networks - formed by the sharing of patients. It is within these social networks that patient care is delivered. Therefore by analyzing provider social networks (not individual providers), the effectiveness of these networks can be examined for varying population groups by disease conditions. Using social networking analysis (SNA), this project will advance our understanding of the complexity of providers’ interactions, the resulting network for delivering care to patients, and the effectiveness of the networks’ outcomes regarding quality and cost. Previous studies that examined interactions among providers utilizing SNA focused on patient sharing and referral patterns, hypothesizing that the structure of such relationships can influence costs and clinical outcomes of healthcare services. This project will allow us to move to the next level and examine not only the developed network but the results obtained from these networks regarding outcomes (efficiency and effectiveness) for specific patient populations and disease conditions. Experimental Plan: 1) Construct provider networks based on observed patient sharing using claims data (from managed care organizations, insurer/payers, third-party claims administrators), 2) consider all providers who were involved in a patient’s care (for specific diagnoses). In other words, rather than studying an individual provider or individual group of providers (e.g., physicians) in isolation, we will consider combinations of various kinds of providers who all contributed efforts toward treating a patient. How this is different than related research: With the creation of Accountable Care Organizations (ACOs) and other models of managed care delivery, advancing our understanding of medical provider networks has become a priority to foster effective and efficient care coordination for specific patient populations among multiple of healthcare providers. Using SNA, ACOs and other managed care entities’ selection criteria for provider contracting can be based on providers’ participation in effective social networks. Specific patient population groups could then be directed to the most effective provider network for their specific disease condition. Milestones & Deliverables: Anticipated duration will be two years. Year 1 milestones: Months 1-4: Recruit graduate students, obtain IRB approval, then acquire, code and clean claims data. Month 5: Determine the disease condition(s) to be studied within a pre-selected geographic area. Months 6-9: Using the claims data, we will construct the medical provider networks and perform analysis. Months 10-12: Prepare final report for presentation to partner as well as begin the development of a manuscript for publication to disseminate lessons learned. The long-term deliverable for this project will be the ability to apply social network analysis to examine the effectiveness of provider networks regarding quality outcomes and costs. Patients can then be directed to those networks providing optimal care by disease conditions. Potential member benefits: Given the unsustainable growth of healthcare spending over the past four decades, U.S. healthcare reforms call for various cost containment measures, while encouraging more coordinated service delivery by medical providers who collectively care for groups of patients. This goal becomes quite challenging when a patient’s care is delivered by a multiplicity of healthcare providers. However, with increased technology capabilities, SNA can be applied to examine the effectiveness of provider networks regarding quality outcomes and costs. Patients can then be directed to those networks providing optimal care by disease conditions. This greater emphasis on effective care coordination and direction of patients can create a process under which clinical integration is improved. Estimated Cost: $50,000 Project Overview and Description Social Network Analysis: Examining Interactions Among Providers at the Network Level Nancy M. Borkowski, DBA, CPA, FACHE, FHFMA Ravi Behar, PhD Gulcin Gumus, PhD University of Alabama & Florida Atlantic University Approach • Advance our understanding of medical provider networks • Consider all providers involved in a patient’s care (for specific diagnosis) • Determine cost and quality effectiveness of networks versus individual providers or selected group of providers • Overview – Analysis of care effectiveness of medical provider social networks for various patient populations by disease conditions. • Description – Construct provider networks based on observed patient sharing using claims data (from managed care organizations, insurer/payers, third‐party claims administrators). – Include all providers who were involved in a patient’s care (for specific diagnosis in pre‐selected geographic area). – Examine not only the developed network but the results obtained from these networks regarding outcomes (efficiency and effectiveness) for specific patient populations and disease conditions. Project Deliverables / Benefits • Deliverables – Months 1‐4: Recruit graduate students, obtain IRB approval, then acquire, code and clean claims data. – Month 5: Determine the disease condition(s) to be studied within a pre‐selected geographic area. – Months 6‐9: Using the claims data, we will construct the medical provider networks and perform analysis. – Months 10‐12: Prepare final report for presentation to partner as well as begin the development of a manuscript for publication to disseminate lessons learned. • Benefits: – Short‐term • Selection criteria for medical provider contracting • Patient population groups directed to the most effective medical provider network for their disease. – Long‐term • Determining and sharing characteristics of effective medical provider social networks for further improvement of care coordination and delivery NSF IUCRC Project Proposal Report NSF IUCRC PROJECT PROPOSAL 6.4 Project Name: Analysis and reduction of practice variance (Collaborative project, GT/NU) Primary Investigator(s): Eva K. Lee, Raghav Srinath, Haozheng Tian, Jinha Lee, Georgia Institute of Technology; James Benneyan, Dr. Susan Haas, Northeastern University Description: This is a collaborative research project between Northeastern University and Georgia Institute of Technology to use system engineering methods to study, predict, and reduce practice and outcome variation. This is a significant and ubiquitous problem across almost all healthcare sectors, and many clinical societies have released consensus recommendations aligned with the national “Choosing Wisely” campaign to reduce practice variation and over/under use of unnecessary diagnostics and procedures (e.g., overuse of imaging, Doppler testing for DVT, standing daily labs, and others). This project will be conducted in multiple healthcare organizations in Georgia and Massachusetts to apply and combine workflow analysis, statistical analysis, predictive modeling, reliability science, and other systems engineering methods to develop a better understanding of causality, identify best practices, target interventions, increase compliance (reduce guideline variation), and reduce variation in both practices and outcomes. Anticipated focus areas are unnecessary referrals and imaging, obstetric practices and harm, pediatric services and anesthetic services. Anticipated partnering health systems include five IAB sites from the greater Atlanta area and at least two from the greater Boston area. We note that working closely with clinical investigators, Georgia Tech has have successes in reducing practice variance, and establishing best practice and new clinical practice guidelines in OR and ICU areas. Experimental Plan: This study will occur in multiple phases. Each site will work with its member health systems to map process logic (process observations, documentation, data analysis), use analytic methods to investigate potential causality (data mining, machine learning, statistical methods), identify and test best practices, optimize local implementation, and test generalizability (via working across multiple systems). This will include monthly cross-site conference calls including the PIs, graduate students, and health system partners to cross-walk, compare/contrast, and align methods, learnings, and next steps. The work initially will start locally (i.e., within-state), but after the first 6 months we also will explore opportunities to cross-replicate any aspects of each university’s work in each other’s health systems, possibly including a breakout session at the fall IAB to launch this phase of the project. How this is different than related research: While practice and outcome variation is focus of individual quality improvement efforts, less effort has gone into more engineering-oriented methods to help understand and impact the problem, nor via a more standardized, scientific, and generalizable process. There also is growing consensus on many choosing wisely guidelines, but work is only just beginning to start to work on implementation and new workflows, with numerous opportunities for systems engineering models to support this. Finally, this project will explore a more robust and generalizable approach to such problems that can be applied and replicated in other healthcare systems. Milestones & Deliverables: 1) Documentation and analysis of practice and outcome variation in each application, 2) Analysis of potential causality and root causes via graphical and statistical analytics, 3) Testing of process changes in multiple sites and analysis of impact, 4) Development of a unified approach to studying and improving similar problems in other contexts. Potential member benefits: Reduction in practice variation, implementation of best practices, and improvements in associated cost, access/flow, and care coordination. Assistance getting started on internal choosing wisely work. Estimated Cost: $70K GT, $42.5K NU Project Overview and Description Rationale Analysis and Reduction of Practice Variance Collaborative Project (GIT/NU) Eva K. Lee, PhD HaozhengTian, Raghav Srinath, Jinha Lee James Benneyan, PhD Susan Haas, MD Georgia Institute ofTechnology & Northeastern University Approach Phase 1 (months 1‐6) • Each site will work with its member health systems to – map process logic (process observations, documentation, data analysis) use – analytic methods to investigate potential causality (data mining, machine learning, statistical methods) – identify and test best practices, optimize local implementation, and test generalizability (via working across multiple systems) • Monthly cross‐site conference calls to cross‐walk, compare/contrast, and align methods, learnings, and next steps Phase 2 (months 7‐12) • Explore opportunities to cross‐replicate any aspects of each university’s work in each other’s health systems • Breakout session at the fall IAB to focus on cross fertilization • Use system engineering methods to study, predict, and reduce practice and outcome variation • Anticipated focus areas are unnecessary referrals and imaging, obstetric practices and harm, pediatric services, and anesthetic services. • Consensus recommendations aligned with “Choosing Wisely” to reduce over/under use of unnecessary diagnostics and procedures Georgia Tech has have successes in OR and ICU areas • Relevance • Significant common problem in healthcare • Impact on cost, access/flow, and care coordination Curbside Consultation Incoming Referral Requests to Comprehensive Neurology Decision Tree Support Tool Redirected treatment plan Leave System Appointment still required Scheduled Appointment Occurs Project Deliverables / Benefits Milestones / Deliverables • Documentation and analysis of practice and outcome variation in each setting • Analysis of potential causality and causes via graphical/statistical analytics • Development of a unified approach to studying and improving similar problems in other contexts • Testing of process changes in multiple sites and analysis of impact Potential Member Benefits • Reduction in practice variation • Implementation of best practices • Improvements in associated cost, access/flow, and care coordination 2014-2015 Research Project Updates NO. TITLE 1 Characterizing and Reducing Avoidable Outside Utilization 2 Identifying Emergency Department Efficiency Frontiers and the Factors Associated with their Efficiency Performance Predictive Models for System Utilization, Capacity, and Flow Optimization Shared Commons Game Theory Models to Improve Antibiotic Stewardship Understanding the Dual Effect of Hospital Safety Culture on Patients & Care Providers; Optimizing Hospital Safety Culture & Reducing Safety Events Bundle Science Statistical Models and Analysis Economic and Potential financial Model of the Perioperative Surgical Home (PSH): Developing a Framework for PSH Design and Action Healthcare Improvement Spread Models Healthcare System Redesign: Advancing Delivery Quality and Effectiveness An Integrated Data Mining and Data Visualization Methodology for Managing Patient Adherence Burnout Among Primary Care Physicians: A Test of the Areas of Worklife Model Evaluating a Medical Screening and Referral Program for Rural Emergency Departments Chronic disease management - clinical, community, and patient-centered approaches Practice Variance: Outcome-Driven Process Redesign & Systems Optimization Hospital Acquired Conditions - Systematic Analysis & Adaptive Approach Quantifying the Impact of Pay-for-Performance Financial Incentives to Reduce Healthcare-Associated Infections The Pediatric Medical Home: Results From A Systematic Literature Review Using Lean Six Sigma to Reduce Hospital Acquired Conditions (HACs) A Combined Human-Factors and Quality Improvement Approach to Assess Electronic Health Records Technology Usability Automatic language translation for improving care management Designing Health Information Technologies to Help Patient Care Teams Identify and Manage Information Problems Gamification and Self-Monitoring of Patients for Enhanced Wellness Outcomes 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 NSF CHOT IUCRC PROGRESS REPORT - PROJECT #22 Gamification and Self-Monitoring of Patients for Enhanced Wellness Outcomes Research team Conrad Tucker, Harriet Nembhard, Abhinav Singh Description The objective of this project is to investigate the fundamental aspects of gaming (both traditional hardcore gaming and casual mobile gaming) that make them engaging, rewarding and stimulating and apply those research findings towards a more immersive healthcare wellness management solution that can be adopted by patients. The video game industry has grown to become a ~$100 billion industry, with the average age of gamers being 30. The success of mobile games such as angry birds, candy crush, etc. has extended the definition of a “gamer” to include a broad range of individuals of all ages and demographics. The term “gamification” is an emerging paradigm that aims to employ game mechanics and game thinking to change behavior. The current physician-patient relationship is top down in nature; a physician provides a patient with a specific set of instructions that they must comply with and a patient goes home and is left to manage their wellness until the next hospital visit. In the context of healthcare, gamification aims to transform the patientphysician relationship into a more collaborative experience, where patients themselves are motivated to succeed in their wellness management goals. considered wellness management systems in the past). This project will focus on maintaining engagement in the wellness management apps through a theoretical understanding of how/why the gaming industry is often successful in maintaining user engagement for extended periods of time. Milestones achieved to date 1. Identification of Gamification features that motivate users and make them successful 2. Draft of the manuscript prepared and being improved regularly 3. Visited Hershey Medical Center to plan the setup of equipment and identify participants to be included in the study 4. Identified potential plans for execution of study with the partner (HMS) 5. IRB in final review stage with the Hershey Medical Center IRB Next Steps 1. Completion of the paper – “A Customized Gamification Model for enhancing patient adherence to Physical Therapy Protocols” – The paper will serve as a theoretical framework for the research to be conducted. 2. Conduct the study with subjects. 3. Submit for journal publication. How is this different than related research? The goal of our project is to create the “angry birds/candy crush” of wellness systems, based on the gamification paradigm that appeals to a broad range of individuals (that may not have Potential member benefits Our industry partnership with Verizon has led to an understanding that for patients, insurance companies and hospitals, gamification will transform the manner in which wellness management is designed and advanced. IT industries can benefit largely the software platforms developed under this project and a better understanding of the data acquisition, transfer and management needs. NSF CHOT IUCRC PROGRESS REPORT - PROJECT # 21 Designing Health Information Technologies to Help Patient Care Teams Identify and Manage Information Problems Research team Madhu Reddy, Jennifer Kraschnewski, Alison Murphy Description Patient-care teams frequently encounter information problems during their clinical decision-making process. These information problems include wrong, outdated, incomplete, missing, or segregated information. Information problems can negatively impact the patient-care workflow, lead to misunderstandings about patient information, and potentially lead to medical errors. How is this different than related research? Although these information problems have existed for some time in paper records, there is an increasing need to focus on them in electronic records due to the tremendous growth in the use of health information technologies (HIT). Consequently, we will investigate the role that HIT plays in supporting or hindering patient care team members’ ability to identify and manage information problems in an inpatient unit of Hershey Medical Center (HMC). Milestones achieved to date Mar-Aug 2014: Conducted 155 hours of patient care team observations Dec 2014: Presented a CHOT Webinar discussing our preliminary analysis Aug 2014-Apr 2015: Currently analyzing the observational data and creating the interview protocol Mar 2015: Submitted a journal paper based on the analysis of the study Apr 2015: Presenting a CHOT Webinar discussing the analysis related to the journal paper submission Next Steps  Continue analyzing data  Finalize the interview protocol  Conduct interviews with participants  Identify HIT features based on data analysis  Develop low-fidelity prototypes  Gather user feedback on prototypes  Finalize report on project findings and contributions Potential member benefits This research can benefit the CHOT members by identifying the types of information problems that patient care teams encounter and describing health information technology (HIT) features that can help reduce the occurrence of these information problems. The project contributions can help to improve the quality of healthcare delivery in hospitals, decrease the chances of medical errors occurring, and lead to the better design of HITs. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #20 Automatic language translation for improving care management Research team Eva K Lee, Karan Uppal, Cody Wang, Georgia Tech; Dr. Prabhu Shankar, Dr. Harold Simon, Emory/CHOA Description Language barriers pose problems for communication and interaction among patients and healthcare providers. Yet, proper communication is critical for optimal health management and outcomes. To improve patient-provider communication for patients with limited English proficiency (LEP), it is necessary to interpret spoken language and translate written clinical documents to the patient’s primary language of communication. This study addresses the translation services and tests computer-assisted translation and machine translation (MT). We utilize freely available open source tools such as Google/Bing Translate, along with our advanced computing machine translation services to improve the accuracy of clinical documents translations. As a pilot we plan to translate elements of discharge summaries of Emergency Department visits, such as discharge advice, medications and other treatments prescribed, explanation about the ailment and actions to be taken in case of emergency, to various other languages, commonly encountered at Children’s Healthcare of Atlanta. How is this different than related research? Although hospitals strive to provide 24/7 interpretation/ translation services, the full implementation of professional language services is less than optimal due to the scarcity of bilingual healthcare professionals, prohibitive costs for professional language interpreter/ translation services and time constraints. In acute care settings, the limited language services could be too slow, not enough and not easy to access, delaying patient management, discharge workflows and frustrating everyone involved. Currently, there is a gap in the care delivered to the LEP population, especially non-Spanish speaking population, where the discharge summaries are given in English. The overall objective of this project is to study the language interpreter/ translation services workflow and find opportunities where advanced informatics solutions could provide robust solutions to the problems associated with language barriers. Our system is the first attempt to create automation customized to the healthcare needs, where the resulting machine translator will continue to learn and improve through multi-level usage. Milestones achieved to date The team has: • Developed, analyzed, and annotated initial ED workflow diagrams with language interpreters and translator service. • Collected summary of ED language statistics for the last five years. • Collected samples of discharge notes and developed templates for streamlining and structuring discharge summaries.. • Benchmarked state-of-the-art computer translation capability of Google Translate, and commercial Canopy system to identify limitations and challenges. • Designed automated content discovery algorithm (using machine learning and natural language processing) for phrase extraction and key content construction from de-identified unstructured clinical discharge notes. • Incorporated SNOMED-CT and controlled vocabularies and dictionaries to ensure quality of content extracted. • Performed preliminary translation on a small set of discharged note: translated extracted content via Google Translate or Bing Translate into 3 top used languages in the hospital: Spanish, Vietnamese, and Burmese. • Evaluated quality of translation via 1) bilingual Evaluation Understudy (BLEU); and 2) medical language expert. • Designed a preliminary web portal to check validity of translation system. Potential member benefits The work will reduce time to translate documents and improve the quality of the discharge process by providing the documents in the language the patient understands. It will also enhance the discharge for patients speaking languages for which there are no translators. It will facilitate standard of care and reduce disparity of care: closing the gap on the missing written discharge information. This work will facilitate the hospital to set up a community language bank. The tools could be applied to various situations across the entire healthcare system where language barriers pose problems and to materials such as health education and disease related documents, brochures, health guides and research briefs. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #19 A Combined Human-Factors and Quality Improvement Approach to Assess Electronic Health Records Technology Usability Research team David Munoz, Hyojung Kang, Chris Deflitch, Harriet Nembhard Description Electronic Health Records (EHR) play a major role in the safety, quality, and efficiency of clinical operations. Although the main objective of EHR is to provide support to clinical activities, studies have reported that these systems are still underused, especially, due to usability challenges. Some of these challenges arise from the disconnection of the EHR's designer and the final user. This causes a tremendous gap between the potential benefits and the actual benefits of the EHR systems. How is this different than related research? The American Recovery and Reinvestment Act (ARRA) of 2009 claimed for a meaningful use of EHR technologies, however, only a few studies have investigated their usability from a detailed user’s perspective. Therefore, the impact of EHR on adoption, satisfaction, and efficiency has not been fully answered. Our combined HF-QI framework provides a detailed mapping of usability issues that inform clinical and designer stakeholders about potential areas of usability improvement. In this research, we propose a combined human-factors (HF) and Quality Improvement (QI) approach to investigate EHR usability issues and their impact on clinical workflow. The framework proposed includes an assessment of key tasks involving humancomputer interaction (HCI) and the measurement of key metrics related to usability, satisfaction, and mental workload. The tools used include a “think aloud protocol” in which users are asked to perform different tasks while several usability metrics are being recorded. Based on the results obtained, QI tools are used to identify key system issues affecting usability, satisfaction, and efficiency. Milestones achieved to date A Questionnaire for User Interface Satisfaction (QUIS) was initially implemented to diagnose the EHR systems of the Emergency Department (ED) at the Penn State Hershey Medical Center (PSHMC). According to this questionnaire, the most pressing issues were related to the speed, reliability, and difficulty of correcting mistakes. From the knowledge gained in this unit, a better connection and coordination between the ED and palliative care services was needed. In response, the HF-QI framework was applied to investigate the palliative care screening tool (PCST) at the PSHMC. A QUIS and a Think Aloud Protocol (TAP) were conducted to investigate the usability of the PCST. From this analysis, various issues were found related to the scoring method for referring patients to palliative care, the mental burden associated with the display and questions, interface design, and guidance for user experience. In order to facilitate the interpretation of the results, a fishbone diagram was developed and distributed to the entire unit. The results served to support and convince the board in charge of this tool that a revision was needed. Additionally, they agreed on that the end-user (nurses) should have a part in this revision. A team of nurses already started to propose improvements to the interface and wording of the tool. Next Steps Next steps of this project include the calibration of the scoring method of the PCST, a QUIS and TAP for the revised tool and a comparison with the current one. For the TAP, a verbalization analysis procedure will be generated to investigate in more detail the areas that can be improved to impact clinicians’ satisfaction and the efficiency of clinical operations for the palliative care referrals and services. Potential member benefits Identifying and quantifying EHR usability issues at the task level, certainly, represent a huge opportunity to inform EHR designers and provide a more user-centered interface. This is expected to have a positive impact on user’s satisfaction, and therefore, an impact on the adoption of technology and efficiency of clinical operations. The concepts investigated in this study are highly aligned with the requirements of the ARRA 2009 that put the “meaningful use” of EHR as a central priority. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #18 Using Lean Six Sigma to Reduce Hospital Acquired Conditions (HACs) Research team Deirdre McCaughey, Wronowski, Lauren Nembhard, Maria Hamilton, Jade McManemin, Harriet Description In fiscal year 2015, CMS will implement the Hospital-Acquired Condition (HAC) Reduction Program. This program mandates that hospitals in the lowest quartile for hospitalacquired infections (conditions that patients did not have when they were admitted to the hospital) or the lowest quartile for medical errors, will receive a 1% penalty on reimbursement, meaning they will only be paid 99% of what otherwise would be paid under inpatient prospective payment system (IPPS). With the average American hospital earning approximately 5 % margin on, a loss of 1% revenue has the potential to be a significantly negative effect on the financial viability of some hospitals. Further, hospital-acquired conditions are largely preventable and thus programs that serve to reduce HACs are an important facet of optimal patient care. How is this different than related research? Limited research has examined HACs using Lean Six Sigma process improvement methodologies. Utilizing a rapid improvement event (RIE) to identify process improvement opportunities will serve as a unique examination of the efficacy of Lean Six Sigma in reducing HAC event frequency. By using both clinical (clinical documentation) and nonclinical (coding process) workflow processes to examine the data and identify process breakdown, this project will serve to identify the surgical and coding process mechanisms with respect to HAC events. Identifying process breakdowns will thereby aid in preventing HACs and improving the medical center’s over HAC score. The RIE methodology serves as a mechanism to purpose improvement action items and develop metrics to track progress over time. HAC event frequencies are expected to decline post RIE. Milestones achieved to date • Project started September: – Preliminary research team meetings August, HMC & PSU – Team established. – Literature search started for HACs frequency and sources. • October progress: – HAC data access permission granted. – Literature search started on utilization of rapid improvement events in healthcare industry. • January progress: – Research team finalized. – RIE planning month. – Identified sample HAC: accidental punctures and lacerations (PSI 15). • February progress: – Continued RIE planning; clinical documentation and coding process determined – Multiple analyses of HMC’s HAC data including frequencies, grouping by service lines, etc. – Literature review of best practices to improve clinical documentation and coding – RIE occurred Feb. 23 – 27 – System/process breakdowns identified & eliminated. • March progress: – Weekly calls with RIE team to monitor action items progress and improvement efforts, – Draft report for stakeholders; disseminated for feedback Next Steps • Month 10-12: – HMC monitoring HAC event frequencies. – Final reports & dissemination for participating hospital stakeholders, industry partners, & academic peers Potential member benefits: The results of this research will assist all hospitals in better utilization of Lean Six Sigma methodologies to examine deficient hospitals processes that result in HACs. Further, by incorporating a rapid improvement event, the project will offer hospitals an important “next - step” in utilizing the study results to improve patient care thereby fostering greater utilization of this research. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #17 The Pediatric Medical Home: Results From A Systematic Literature Review Research Team Bita Kash, Debra Tan Description With the recent act of the Patient Protection and Affordable Care Act of 2010, medical homes have been utilized and implemented as a method to improve health outcomes, as well as reduce rising health care expenditures. Interest in the medical home model has grown exponentially across many spectrums of the health care system. The pediatric medical home can serve as a way to provide high quality preventive care to the patient, as well as enhance team-based care and revitalize the field of primary care. This study applied a patient-segmentation approach, which organizes health care based on value to patients. How is this different than related research? A patient-segmentation framework addresses the issue that not “one-size-fits-all” as every child is extremely different in terms of health care needs. Pediatric primary care should be organized and tailored differently around specific subgroups of patients with similar needs. Milestones Achieved To Date Months 1 to 5: Identification of Models of Practice  Comprehensive, systematic literature review  IRB approval  Identification of “Evidence-Based Models of Practice”  Discussion with professional advisor regarding “Evidence-Based Models of Practice” Months 5 to 10: Refinement of Models Studied and Focus Groups  Report to industry member regarding “Evidence-Based Models of Practice”  Focus groups with key informants regarding “Evidence-Based Models of Practice” Next Steps    A qualitative analysis report regarding findings Prioritization and evaluation of recommendations for sustainability and implementation A final report will be presented to the industry member regarding operational and financial details of models for future implementation Potential Member Benefits Our industry member, as well as other hospitals and policy makers will benefit from a clear understanding of operational, staffing, and financial details for improvement and sustainable change regarding the pediatric medical home model. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #16 Quantifying the Impact of Pay-for-Performance Financial Incentives to Reduce Healthcare-Associated Infections Research team Nathaniel D. Bastian, Hyojung Kang, Paul M. Griffin, Harriet B. Nembhard Description Healthcare-associated infections (HAIs) are infections that patients contract while receiving treatment for medical or surgical conditions, which impose a considerable strain on the US healthcare system. According to the Centers for Disease Control and Prevention, roughly 1 out of every 25 hospitalized patients contract some form of HAI, 25.6% of which are deviceassociated infections such as central line– associated bloodstream infections (CLABSI). In addition to the resulting morbidity and mortality, HAIs have significantly contributed to the rising cost of hospital care. Overall annual direct medical costs of all HAIs in the US have been estimated to fall in the range of $35.7 to $45 billion. On a per-case basis CLABSI are the most expensive HAIs in the US, costing an average of $45,814 per case. Since approximately 70% of HAIs are preventable, there has been substantial attention to the benefits and approaches for prevention. Public reporting of HAIs and valuebased purchasing approaches, such as pay-forperformance (P4P), are two advocated strategies for hospital quality improvement. P4P programs provide financial incentives to providers that achieve specified quality metrics with the goal of improving health outcomes at a lower cost. As insurance providers deploy P4P programs, evaluating the resulting improvement in quality of care is important for assessing a broader implementation. In this project, we evaluate the impact of Highmark’s Quality Blue (QB) Hospital P4P Program on the improvement of quality outcomes by focusing on CLABSI. We investigate two research questions. First, we determine whether hospitals that participate in Highmark’s QB program have a lower expected number of CLABSI compared to hospitals that do not participate. Results indicating a positive association between QB and realized improvements in quality outcomes adds to the understanding of the impact of P4P programs, especially since CLABSI rates are counted nationally in both HAI reduction programs and value-based purchasing programs. Second, we assess whether the number of years a hospital participates in QB affects the expected number of CLABSI. Findings indicating that continued hospital participation in P4P programs results in even further, sustained improvement adds to the evidence-base of the effectiveness of healthcare financial incentive models. How is this different than related research? The literature on whether P4P programs achieve their intended goals is mixed. In particular, the contribution of the financial incentive to the improvement of quality remains unclear. Although some studies have found a link between the incentive and improvement, several studies have found no effect. Milestones achieved to date 1. Manuscript under review for publication in Medical Care Research and Review. 2. Key findings: a. On average, those hospitals that participated in the QB program had 0.727 times the CLABSI as those hospitals that did not participate in the program. b. Hospitals that participated in QB for 4 or more years had on average 3.13 few CLABSI per year compared to those hospitals participating for less than 4 years. Next Steps Determine the cost-effectiveness of Highmark’s QB program for participating hospitals in Pennsylvania. Evaluate the economic benefit of the P4P program in terms of return-oninvestment. Potential member benefits Highmark, as a NSF-CHOT partner, has identified the strategic priority around a better understanding of financial incentives for HAI. This project is potentially significant for all NSF-CHOT hospital partners, and we expect to leverage their participation in the effort as appropriate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“$6\VWHPV$SSURDFKWR 5HGXFLQJ&HQWUDO/LQH$VVRFLDWHG%ORRG6WUHDP ,QIHFWLRQV”WR$PHULFDQ0HGLFDO,QIRUPDWLFV $VVRFLDWLRQ 3RWHQWLDOPHPEHU EHQHILWV x 5HGXFH+$&LQFLGHQWVWKXVUHGXFHXQQHFHVVDU\OHQJWKRIVWD\DQGH[WUDPHGLFDOFDUH x ,PSURYHTXDOLW\RIFDUHDQGWUHDWPHQWRXWFRPHIRUSDWLHQWV x ,PSURYHSURYLGHUDQGSDWLHQWFRPSOLDQFHLPSURYHSURYLGHUV¶ PRUDOHDQGFRQILGHQFH x ,PSURYHKRVSLWDOVXUYHLOODQFH x ,PSURYHKRVSLWDOUHVRXUFHXWLOL]DWLRQ x (VWDEOLVKDFRQGXFLYHDWPRVSKHUHIRUVXVWDLQDEOHSURFHVVDQGFKDQJHWUDQVIRUPDWLRQZKHUH +$&DZDUHQHVVLVLQWHJUDODQGVHFRQGQDWXUHWRVHUYLFHSURFHVV NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 14 Practice Variance: Outcome-Driven Process Redesign & Systems Optimization: II. Practice Variance at Epidural Processes Research team Eva K Lee, Haozheng Tian, David Little, Jr., Jinha Lee, Raghav Srinath, Will Lewis, Cody Wang, Georgia Tech Description Practice variance is an important issue to analyze as a means to optimize care delivery (quality and efficiency) and to encourage collaborative learning for broad quality improvement. The project focuses on epidural procedure for labor delivery. Epidural anesthesia is used to improve pain relief in a variety of lower abdominal and lower extremity surgeries, including cesarean and labor deliveries. While epidural analgesia is used in hospitals across the country to assist in deliveries, there is significant inter and intrafacility practice variance. Our objective is to quantify needle-based epidural strategies: focusing on effectiveness vs efficiency vs outcome vs patient satisfaction. In particular, we seek to understand and optimize medication resource usage (dose, duration vs patient response and outcome); develop metrics to measure outcome (quality vs patient satisfaction, initial process vs final outcome); identify best practice, and potentially develop clinical practice guidelines. We aim to capture the variance and establish evidence-based outcome documentation. How is this different than related research? General practice for injecting anesthesia applies medication primarily through the catheter. Little is known or published regarding needle-based approach. Dosage, efficacy and safety have not been documented. This study aims to provide evidence that needle-based epidural practice is safe; and that patients can achieve proper sensory level within the same duration as in the catheterbased approach. We will analyze the proficiency of physician practice, and provide insights on their preference in terms of medication and dosage. Milestones achieved to date We have thus far:  Performed on-site observation and documentation of practice and variance among 41 providers. A total of 412 patient cases were collected.  Established preliminary process maps and annotated practice variance among providers.  Analyzed patient sensory response versus medication, dosage, providers, and varying care practice.  Analyzed providers’ preference and outcome response.  Identified practice variance among providers.  Identified potential error-causing factors.  Performed patient-dose sensory response analysis  Completed process maps for epidural procedures  Designed decision-simulation model to analyze practice reliability to determine best practice.  Reported needle-based findings to practitioners and prioritized recommendations. Next Steps  Prepare first journal article for public dissemination.  Capture and document practice and variance in C-section and vaginal birth.  Contrast outcome against hospital resources and patient satisfactory  Collaboratively design clinical practice guideline for optimal needle-based epidural practice. Potential member benefits  Establish dose response factors and practice characteristics  Improve quality of care, and reduce potential complications  Provide documentation and evidence of needle-based epidural practice  Facilitate development and standardization of best practice clinical guidelines  Facilitate evidence-based practice NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 14 Practice Variance: Outcome-Driven Process Redesign & Systems Optimization: I. Pediatric Heart Network Practice Variance: Collaborative Learning Research team Eva K Lee, Jinha Lee, Ankit Agarwal, Georgia Tech Description It is recognized significant practice variation exist in early post-operative management among pediatric cardiac centers. This variation may impact important outcome measures. A large factor is the variation in patient characteristics and risk factors. Non-patient factors include experience, resources, and experimentation. Some centers may commit greater resources to certain procedures. Other centers may encourage experimentation, resulting in adoption of changes in surgical and medical care that appear promising and divergence in management practices from those at other institutions. Practice variance is an important issue to analyze as a means to optimize care delivery (quality and efficiency) and to encourage collaborative learning for broad quality improvement. This study will focus on the entire process of congenital heart surgery from surgery to end of post-operative care. We aim to identify potential improvement.. How is this different than related research? This is a nationwide collaborative study that involves multiple pediatric heart centers. Site visits and observation may be particularly valuable in quality improvement for congenital heart surgery given critical role of communication among various clinical teams (anesthesiologists, surgeons, cardiologists, nurses and others) involved in the care of an individual patient. Collaboration with both inter-facility and intra-facility has the added potential of stimulating new ideas for investigation or new management techniques, and increases our ability to conduct prospective research in a highly specialized clinical setting. Experimentation and discussion among colleagues can lead to the rapid adoption of innovations and avoid the replication of disadvantageous techniques. Collaborative learning in pediatric cardiac surgery requires a multi-institutional approach due to relatively low volumes. A national structure for collaborative site visits has never been tried, to our knowledge, in any field. Milestones achieved to date We report herein work completed for this study so far. In particular, we have:  Performed on-site observation, documentation and contrast of practice variance across five sites: CHOA, CHOP, C.S. Mott Children’s, Primary Children’s and Texas Children’s  Established process maps, workflow and procedures occurring in OR, ICU, step-down and discharge unit. Duration, decision making process, patient care steps and coordination were documented.  Summarized and contrast causes of practice variances and potential impacts;  Consensus development to collaboratively establish best practice and new clinical practice guideline (CPG).  Implemented procedural change for care improvement following CPG.  Evaluate awareness of new CPG in key staffs of all five sites.  Completed and submitted a paper “Practice Variance Analysis for Process Improvement in Post-Operative Care of Congenital Heart Surgery” to the American Medical Informatics Association. Next Steps  Evaluate outcome findings and compare before and after CPG implementation results.  Improve CPG through collaborative feedbacks.  Establish CPG for broad national dissemination. Potential member benefits  Improve quality and efficiency of care, and treatment outcome for patients  Facilitate successful dissemination of best practice  Reduce length-of-stay through early extubation  Improve care coordination and management  Establish new CPG for broad national dissemination NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 13 Chronic disease management - clinical, community, and patient-centered approaches Research team Eva K Lee, Xin Wei, Cory Girard, Georgia Tech Description 68% Medicare spending goes to people with five or more chronic diseases. Reports found that between 44% - 57% of older patients take more than one unnecessary drug. The management of multiple diseases is complicated and offers daunting challenges to healthcare providers. More drugs are prescribed for treatment, which causes reduced adherence of patient to drug therapy, higher possibility of drug-drug interactions, more side effects observed on patients, less effective treatment, and more frequent changes in drug therapies. This results in more hospital visits, heavier burden on the use of health resources and higher medical expenses. The objective of this study is to optimize the medical interventions, treatment plan, and drug therapy decisions to reduce risks of adverse/side effects, increase efficacy of treatment, minimize mortality risk, and improve quality of life. How is this different than related research? First, the project focuses on coexisting multiple conditions, rather than a single disease. Thus, it is more challenging, interesting, and clinically relevant. So far, there is no mathematical model developed for long-term, dynamic, and all-around treatment of multiple diseases. A quantitative model based on clinical desirable outcome will reduce the negative effect of individual provider’s subjectivity on decision making process on managing treatments and drug therapy. The project helps to identify guidelines of multiple disease treatment. It will reduce the time pressure of doctors on unnecessary patient visits, and assists doctors to manage complex treatments. This project considers multiple stakeholder perspectives (patients, doctors, caretakers). Chronic disease also requires pro-active patient participation as well as fostering a community and culture for healthy living. Active home and community engagement provides a supporting environment. Remote sensors can be fun and offers unique opportunity for health engagement and communication between providers and patients for sustained health improvement. Milestones achieved to date  Completed retrospective review of 2011 and 2012 patients with multiple chronic conditions. Data relevant to treatment of patient, drugs, disease patterns were analyzed.  Examined decision making process of doctors on treatment planning (e.g., treatment, drug prescription). This was done via interviews, observations, and analyzing of clinical notes.  Focusing on diabetes with other chronic conditions, mathematical models were derived and implemented for optimal drug therapy decision and intervention plan.  Established polypharmacy relationship on drugs used in multiple-diseased patients.  Prepared a clinical paper for submission (still in progress). Next Steps  Compare our treatment plan against actual treatment.  Design new treatment practice guidelines  Design remote patient sensors and monitoring devices.  Incorporate community outreach and activities to promote healthy living environment. Potential member benefits      Produce quality personalized treatment plans for patients with multiple conditions. Return optimal outcome-driven treatment for multiple conditions with lower cost and better control of disease symptoms. The resulting treatment will also use minimum amount of drugs, thus reducing the risk of adverse/side effects and increasing the efficacy of the treatment (more drugs mean high risk of non-compliance). This all will translate to improve the quality of care and quality of life of patients. From hospital care coordination viewpoint, it will allow clinicians to optimize patients’ hospital visits and focus on personalized outcome-driven treatment. Positive and healthy home and community environment facilitate pro-active patient health engagement, and promote healthy eating. Remote sensors offer care continuation (outside clinic), promote active engagement to sustain broader health improvement. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #12 Evaluating a Medical Screening and Referral Program for Rural Emergency Departments Research team Murray J. Côté, Tiffany A. Radcliff, Dylan G. Dacy, Description Emergency Department (ED) overcrowding due to nonemergent use is an ongoing concern facing most health systems. In 2011, a health system that primarily serves rural communities in Texas instituted a new program to medically screen and refer nonemergent patients to nearby affiliated rural health clinics (RHCs). Program evaluation was conducted in two parts: 1) description of the program goals, process, and early implementation experiences at two sites that adopted the program before wider implementation within the health system, and 2) detailed analysis of patientlevel data for four sites that included ED visits and RHC visits to determine quantitative effects of the program. How is this different than related research? This research represented a comprehensive, longitudinal evaluation of a medical screening policy to better match demand (i.e., arriving patients) with capacity (i.e., the “most appropriate” service component of the health system). We were able to study the health system’s program from its conception, to phased implementation, to postimplementation evaluation. The participation of the health system was essential in providing unfettered access to all key decision makers and relevant operational and financial data. Such level of detail is rarely available when health system operating policies are proposed and adopted. Milestones achieved to date 1. Qualitative analysis of program has been completed with results published: Menser, TL, TA Radcliff, and KA Schuller, “Implementing a Medical Screening and Referral Program for Rural Emergency Departments,” Journal of Rural Health, 2015, 31 (2), 126-34. 2. Patient level data obtained for four sites for ED visits and RHC visits covering November 1, 2010 through December 31, 2013. 3. Descriptive analysis of patient level data indicated approximately 55,000 unique ED patients and approximately 54,000 unique RHC patients across the four sites. 4. After program implementation, general themes indicated that overall ED volume decreased and distribution of arrivals by triage changed with a greater proportion of arrivals in “severe” triage levels. Both results allude to program success in matching patient demand with appropriate health system capacity. Next steps 1. Complete quantitative analysis of patient level data to determine change in arrival rate that indicates program effectiveness. 2. Develop appropriate quantitative models to illustrate system capacity effect of program implementation and adoption. 3. Prepare and submit manuscripts to appropriate peer-reviewed journals to share methodologies, results, and implications. Potential member benefits The health system was able to leverage excess capacity of affiliated RHCs to accommodate lowacuity patients referred from the ED and may lead to improvements in Triple Aim goals of increased patient satisfaction, better population health and outcomes, and lower per capita costs. Lessons learned from this program may inform similar processes aimed to reduce nonemergency ED utilization by other rural health systems. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #11 Burnout Among Primary Care Physicians: A Test of the Areas of Worklife Model Research team Sean Gregory, Terri Menser Description Examinations of the current state of the physical workplace in the United States and globally indicate a declining overall well-being and specifically increasing levels of burnout. The consequences of these effects include early retirements or exits from the medical profession, difficulties improving the patient experience, and low levels of provider engagement with clinical-level and systemlevel initiatives. Such consequences affect physicians, healthcare organizations, and patients. While most research has focused on identifying burnout, cataloging its effects, and creating a case for attending to its impact, relatively few studies have focused on exploring the antecedents of burnout for physicians. How is this different than related research? The goal of this study was to test an etiological model, the Areas of Worklife Scale (AWS), for practicing primary care physicians. Using the AWS and the Maslach Burnout Inventory, the study used a longitudinal survey research design method to query primary care physicians employed at a large integrated delivery system in the United States. Data collected successfully fit the AWS model for burnout among primary care physicians, supporting the theory that workplace drivers are responsible for burnout. Workload, control, and values congurence are the largest drivers of burnout for practicing primary care physicians. Milestones achieved to date  Final sample size for the study was 153 unique physicians: o 97 at baseline, o 91 at the 3-month follow-up, and o 56 at the final 6-month follow-up assessment.  All six dimensions of the AWS and all three dimensions of the MBI were consistent with the established literature on the validity of both instruments  There were no significant modification indices, indicating no paths that would offer an improvement to the specified model for both the AWS and MBI measures. Next Steps The next part of this study will be to replicate the study to test this model on physician populations that experience differing levels of control (e.g., independent physicians, specialists, surgeons). Given that the results of this study show the lack of a significant path between rewards and EE among practicing primary care physicians, gaining further insight into the generalizability of this finding would make both a theoretical and a practical contribution. Whatever the current reasons for burnout, or its historical rationale, its endemic situation among physicians is unsustainable and calls for further examination for the good of the profession, the professionals, and the population. Potential member benefits The AWS model provides key insights into the domains of work that cause stress and ultimately burnout for physicians, and these domains can guide physicians and managers to develop interventions to fight the rising incidence of burnout. NSF CHOT IUCRC PROGRESS REPORT - PROJECT # 10 An Integrated Data Mining and Data Visualization Methodology for Managing Patient Adherence Research team Conrad Tucker, Harriet Nembhard, Ishan Behoora Description Patient non-adherence to physician-prescribed disease and wellness management protocols is a major challenge in the healthcare industry and has led to an increase in hospital visits, health risks and medical costs. For example, the non-adherence to prescribed medication results in over 125,000 deaths per year and a financial burden to the healthcare system exceeding $100 billion in direct costs. This project will explore patient adherence for those who adopt a proposed sensor and visualization system for remote wellness management and feedback. How is this different than related research? Systems such as AutoCITE reveal that remote patient supervision has tangible impact on patient health outcomes. The main limitations of existing techniques are that they are physically invasive, often requiring patients to wear some digitally connected device for an extended period of time. Furthermore, these systems do not provide an integrated healthcare delivery strategy that connects the sensing results to the patients and healthcare officials in a seamless, visually straightforward manner. The proposed project aims to not only predict patient adherence, but also provide feedback to both patients and physicians, which can then help physicians prescribe alternative solutions if a patient is nonadherent. Milestones achieved to date 1. Developed a machine learning driven methodology for remote monitoring of adherence of patient's with Parkinson's disease. 2. Our methodology utilizes non-invasive sensors and is able to differentiate between patients on and off their medication with high precision ranging from 95-99% 3. A journal article on the research is under review by computers in Biology and Medicine 4. Additionally developed an interactive graphical user interface which could help monitor adherence of patients undergoing physical therapy and provide feedback in real time. Next Steps 1. Incorporation of gamification elements into the graphical user interface to improve patient engagement 2. Further development of machine learning techniques for real time accurate detection of adherence in patients undergoing physical therapy. Potential member benefits Our industry partnership with Verizon has led to an understanding that for patients, insurance companies and hospitals, a convenient and automated technique to monitor treatment progress can lead to large time and money savings. In particular, industries can benefit largely from the research into sensor placement and data management and transfer. This will be an increasingly important field, as low cost sensors we use in our homes become more prevalent. NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 9 Healthcare System Redesign: Advancing Delivery Quality and Effectiveness Research team Eva K Lee, Cody Wang, Matthew Hagen, Kevin Liu, Georgia Tech Description Individual health systems provide various services and allocate different resources for patient care. Healthcare resources including professional and staff time are constrained. Patient lifestyle patterns are mostly suboptimal with adherence with pharmacotherapy is often limited. This study aims to 1) identify critical variables that impact outcomes (e.g. control of risk factors and prevention of hospital/ED admission) and inform allocation of limited time and resources for greater effect; 2) address realistically modifiable social determinants of health that will improve community health; and 3) seek greater use of treatment evidence (e.g. secondary EMR usage, “OMICs” data) to advance quality and effective of care delivery. We aim to increase quality and timeliness of care, maximize financial performance, and decrease practice variability across the organization. How is this different than related research? This study attempts to combine socialeconomic and demographics demands, hospital resources, and evidence of treatment (including EMR, Omics, and other laboratory data) to redesign the delivery process for quality and effectiveness of healthcare delivery. While efficiency is often performed via process improvement, patient risk factors, disease patterns and treatment characteristics may shed lights on resource needs and care requirement, and provide holistic health systems redesign opportunities for improving care quality and effectiveness. Milestones achieved to date  Collected 2.7 million patient data from 400 clinical providers  Created a secured relational database containing health records pertaining to procedures, demographics, diagnosis codes, vitals, laboratory measurements and medications.  Designed a user interface for retrieving, filtering, summarizing, and visualization of health records, statistics, and health trends.  Designed a real-time an-demand visualization of patient diseases and geo-spatial distribution.  Uncovered health/disease trend against different treatment modalities, providers, and demographics  Identified providers practice variance Next Steps  Uncover treatment outcome evidence of various types of diseases.  Investigate polypharmacy and design optimal individualized treatment plans  Establish demand and hospital resource usage patterns  Optimize providers’ resource against patient demands and needs.  Implement results for a chosen set of hospital units.  Document and evaluate improvement. Potential member benefits • Improve quality and efficiency of care • Reduce waste; serve more needed patients • Improve demand-resource alignment • Reduce prolonged LOS (and thus reduce hospital acquired conditions), • Improve capability in the event of pandemic or disaster response • For patients: timeliness and personalized evidence-based care; reduce unnecessary hospital stay, and associated risks and costs NSF CHOT IUCRC PROGRESS REPORT - PROJECT # 8 Healthcare Improvement Spread Models Research team James Benneyan, Dayna Martinez, Cory Stasko, Ram Prashanth Radha Krishnan Model of Network Spread and Analysis of Idea Adoption Description This project focuses on improving our understanding of how improvement ideas spread across healthcare and how to accelerate adoption The overall approach aims to:  Understand the topologies of existing Quality Improvement Networks (QINs)  Develop mathematical models to simulate the spread of ideas  Use these models to evaluate the impact various topological and other factors have on spread  Find optimal interventions that could be used to maximize the read of spread and adoption of ideas on QINs Model Analysis How is this different than related research? This research builds upon previous work of Rogers and the evolving literature on the diffusion of innovation, with the focus on healthcare QINs. We are studying node-level and network-level properties. Two modeling philosophies are developed and compared: (1) Ideas are treated as a continuous quantity that flows; (2) The spread of an idea is considered a binary, but stochastic event In addition, we have separated awareness and adoption of an idea, so that, for example, passing trends may be distinguished from sustained changes in practice. This work also combines the study of several actual QINs with artificial networks exhibiting topologies otherwise not represented. Where and how ideas get stuck, when examined for a large scale, is an important are that has not seen significant progress Milestones achieved to date  Developed continuous and binary simulation models for the spread of ideas  Mapped and studied thirteen total QINs  Use network generation algorithms to produce artificial networks for study  Investigated the effect of 4 different network-level parameters and 3 nodelevel parameters on spread performance  Completed paper on the simulation approach and network properties studied  Developed both exhaustive and genetic optimization models for finding optimal network structure modifications and strategic idea seeding to maximize spread  Drafting paper on network and spread strategy optimization Network Structure Optimization Optimal λ Duration Initial Node Next steps  Publish academic paper on the empirical network study and spread simulation model  Evaluate the sensitivities of optimal network design and seeding strategy  Complete and publish academic paper on the network and seeding optimization Potential member benefits 1. Enhanced understanding of Quality Improvement Network structure, properties, and evolution 2. Identification of network structures that most effectively facilitate rapid and sustainable spread 3. Clarify and compare the impact of potential interventions on promoting the spread of innovation NSF CHOT IUCRC PROGRESS REPORT - PROJECT #7 Economics and Potential Financial Model of the Perioperative Surgical Home (PSH): Developing a Framework for PSH Design and Action Research team Bita Kash, Kayla Cline Description The “perioperative surgical home” is a relatively new concept that is based, at least in part, upon the patient-centric characteristics of the medical home combined with foci on team science, micro-systems, service line management, care-coordination, and bundled payment. The purpose of this study is to continue to define the “surgical home” conceptually and to identify and describe the economics and detailed financial model of one selected PSH model in the U.S. To better understand the financial model, we obtained data on both the costs and benefits of eight activities central to the PSH that were identified in the first year of this project: 1. Coordinated preoperative testing, 2. Prehabilitation 3. Early patient education 4. Blood utilization programs 5. Operating room scheduling initiatives 6. Nausea and vomiting initiatives 7. Early patient mobilization 8. Coordinated discharge planning Benefit data were culled from comprehensive literature reviews conducted for each of these activities. A time-dependent activity-basedcosting approach (TD-ABC) was employed to assess individual costs for each of these activities. Under this approach, researchers developed a comprehensive TD-ABC survey to be pilot tested and disseminated to PSH programs operating around the country. How is this different than related research? Significant prior research has been devoted to each of the eight activities listed above. Most of this research describes the effect (generally, the benefit) of a specific intervention at a specific hospital site or physician practice. This project incorporates the evidence across interventions and sites to develop a more comprehensive picture of the benefits accrued by each activity. In addition, data on the cost of implementing such an activity is very limited in the literature because cost measurement for such activities is problematic in practice. The development of a survey tool using timedependent activity-based costing is a novel approach to solving this problem, as it requires data on staff time rather than staff costs. Milestones achieved to date - Eight individual comprehensive literature reviews on benefits of each of the eight PSH activities - Compilation of literature review findings on these benefits into easily readable one-pagers - Site visit to assess the feasibility of obtaining hospital-specific cost data - Development of TD-ABC survey tool to be piloted and then disseminated to PSH programs around the United States - Comprehensive literature review of cost assessment methods (in progress) Next Steps The completion of the comprehensive literature review on cost assessment methodology will be completed by the end of April. A report on benefits reported in the literature and methodologies that could be used to measure costs, with an emphasis on further development of the TD-ABC survey, will be completed and disseminated to the industry member by the end of May. This survey will be made available to the industry member in PDF and Qualtrics format to facilitate pilot studies and further dissemination as needed. Potential member benefits 1. Identification of the primary benefits associated with each PSH activity 2. Examples of specific interventions employed for each of these activities 3. Ability to highlight which activity is most beneficial to achieve a given outcome 4. Access to a time-dependent activity-based-costing survey that can be used to assess the time and other resources required to initiate and operate each of these activities NSF CHOT IUCRC PROGRESS REPORT - PROJECT # 6 Bundle Science Statistical Models and Analysis Research team James Benneyan, Eralp Dogu, Aven Samareh Description The objective of this project is to investigate statistical methods for patient safety “bundles” and risk-adjusted binary data. It would be beneficial to monitor bundle compliance over time, and analyze relative importance and interaction of bundle elements. A particular focus here is on investigating statistical quality control charts under ‘real world’ conditions of messy data with an assumption that process parameters are not known to us. Bundle Control Chart Example: Total Joint Replacement SSI Bundle p Chart 1.0 0.9 0.8 provided an analysis concerning the required number of samples, sample sizes and number of elements in a bundle. This includes:  Developed a simulation model by which we generated a phase I data sets for different samples, sample sizes and number of elements of bundle in Matlab.  Developed a Markov chain code in Matlab as an accurate approximation for average run length (ARL), to compare performance under ideal and above cases. It could be seen that ARL performance is sensitive to the choice of the samples and sample size. This study could be very useful for researchers for designing np charts in order to detect minor process variations in evidencedbased events and improving quality of care. Furthermore, ARL values are close to the design value of 500 for lower number of bundle elements and sample sizes as well as for low compliance rates. Correct Incorrect 0.7 1 2 3 4 5 6 7 8 9 10 11 12 How is this different than related research? Despite becoming part of routine improvement projects, the evidence based bundles is limited at best. In this work we develop a general bundle science framework and tools to compare and monitor bundle compliance over time. Milestones achieved to date An indispensable assumption for construction of control charts is that the process parameters are assumed to be known. In practice, the process parameters are rarely known, and are usually estimated from an in-control historical data set (phase I). When the parameters are estimated, the performance of the control charts differs from the known parameters case due to the variability of the estimators used during the Phase I. Hence, we developed and extended statistical methods for bundle monitoring by deriving the run length properties of the investigated np charts, and Next steps  Extend these results to other risk-adjusted data and estimation error contexts  Begin analysis of bundle compliance data, relative effect sizes, aggregate impact, and inclusion criteria Potential member benefits 1. Validated statistical methods for comparing and monitoring bundle compliance over time 2. Understanding of the relative importance of different bundle elements 3. Development of a general bundle science framework NSF CHOT IUCRC PROGRESS REPORT - PROJECT #5 Understanding the Dual Effect of Hospital Safety Culture on Patients and Care Providers; Optimizing Hospital Safety Culture and Reducing Safety Events Research team Deirdre McCaughey, Maria Hamilton, Lauren McManemin, Jade Wronowski, Harriet Nembhard Description The healthcare industry in the USA continues to report among the highest rates of workplace injury and illness of all industries. Many studies examine care provider personal safety perceptions and have found these perceptions influence care provider health & wellness. With respect to patient safety, hospitals continue to struggle with effective tools and processes to reduce patient safety events. Retrospective data shows that many of the facets that promote a safe environment for care providers are the same facets as those that promote a safe environment for patient care. This project will identify and assess the facets safety culture that influence patient safety events by utilizing care providers’ perceptions of safety. This information will direct hospitals in the most efficacious manners to optimize their patient safety culture. How is this different than related research? While there are substantive literature bases in patient safety, there is a dearth of studies that examine the variations in patient safety perception by care provider type (e.g. nurses, physicians, aides) or by unit type (e.g. surgery, medicine, emergency department). Given the variations in care provider duties and patient interactions, it is highly probable that patient safety perceptions will vary across units and positions. In addition, there remains a lack of exploratory analysis into the influence of specific facets of safety culture on overall provider safety ratings. Are the core components of positive patient safety culture perceptions derived from teamwork, hospital managements’ support individual training, etc. for patient safety, Milestones achieved to date • Project started September, 2014: - Preliminary research team meetings August, HMC & PSU. - Team established. - Safety data access permission granted. • October progress: - Safety culture data file created. - Data coding and preliminary started. analysis • November progress: - Literature review of patient safety culture studies and AHRQ patient safety data. - Data coding and preliminary analysis started. • February progress: - Safety culture data analyzed. - Literature review on subject complete. • March progress: -Final data models examined. - Draft report completed. -Weekly calls with HMC to review data findings and adjust models as applicable. Next Steps • Month 10-12: - Finishing up data analysis. - Evidence-based recommendations for optimizing hospital safety culture. - Final reports & dissemination for participating hospital stakeholders, industry partners, & academic peers. - Examine opportunity to re-evaluate data longitudinally. Potential member benefits The results of this research will assist all hospitals in developing a better understanding of the relationship between patient safety culture and care provider safety perceptions. By identifying the variations in safety perceptions, hospitals with identify critical areas of focus to improve the hospital’s safety culture and reduce patient safety event frequencies. NSF CHOT IUCRC PROGRESS REPORT - PROJECT #4 Shared Commons Game Theory Models to Improve Antibiotic Stewardship Description The objective of this project is to help understand antibiotic resistance dynamics and how to best limit its growth by using game theory and system dynamics models. Antibiotic resistance remains a growing problem of broad health and cost concern, with significant focus on antibiotic stewardship as one important intervention. We have begun developing three types of operations research models to describe this problem and inform policy, including game theory, agent-based population dynamics, and system dynamics models. In behavioral economics, stewardship can be viewed as a “tragedy of the commons”, Hardin's analogy of a shared town pasture for which each herder’s incentive to graze their sheep without concern for others thereby reduces the long-term value to everyone. For antibiotic stewardship this equates to their over-use, with short-term incentives to use antibiotics for individual care episodes but at the consequence of reducing their long-term effectiveness across a community. System Dynamics Model # of deaths due to CDI Public concern Susceptible inpatients Repeated CDI patients Milestones achieved to date  Developed a system dynamics stock and flow model to simulate the interaction of factors that determine the prevalence of a particular resistant infection (Clostridium difficile).  Demonstrated with economic incentive analysis why the rational individual equilibrium motivates poor provider stewardship even though providers achieve greater value with a coalition approach to stewardship in a cooperative equilibrium.  Developed regional resistance temporal and spatial simulation model to project the spread of antibiotic-resistant bacteria in Massachusetts based on the stewardship strategies hospitals and nursing homes Economic Incentive Analysis 0.48 0.47 0.46 0.45 0.44 0.43 0.42 0.41 0.4 Rational strategy Poor stewardship Provider utility Research team James Benneyan, Awatef Ergai, Brendan Bettinger, Cory Stasko, Anne-Marie Chouinard Cooperation always adds value Individual Action 0.0 CDI First-time CDI First-time outpatients inpatients Inpatient CDI spread rate ABX Inpatients prescription CDI exposure with CDI rate outside CDI exposure in health facilities Susceptible Outpatients outpatients with CDI Outpatient CDI spread rate Collective Action 0.2 0.4 0.6 0.8 1.0 Stewardship commitment level Next steps • Revise and publish paper currently under review on the system dynamics model • Recommend how to change incentive structures to promote ideal stewardship • Project prevalence growth of resistant bacteria for scenarios of interest to inform regional policy How is this different than related research? While many efforts have been pursued to increase antibiotic stewardship, to our knowledge, limited work has been conducted using these types of models. Potential member benefits 1. Improved understanding of how stewardship policies, participation rates, and consistency impact resistance and specifically costs associated with C. diff infections. 2. Methodology to identify most effective interventions to reduce the extent and spread of resistance. NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 3 Predictive Models for System Utilization, Capacity, and Flow Optimization Research team James Benneyan, Samuel Davis, Kendall Sanderson Description This was a phase-1 project to scope and initiate a portfolio of work in the general area of predictive modelling to help manage patient flow and care. Applications included predicting (1) bed demand in intensive care and inpatient units one-throughseven days in advance on a rolling basis, (2) number of daily ED patient admissions, and (3) unnecessary specialty referrals. In each case preliminary results were generated and evaluated in order to assess decision making utility and specifics of future projects. How is this different than related research? The value of predictive information in healthcare is increasingly appreciated, such as for patient risk identification, but less explored in other potentially useful logistics contexts. Milestones achieved to date  Developed bed demand forecasting simulation tool that is being tested in multiple health systems  Identified applications in 3-4 health systems  Developing adaptive algorithms to optimally respond to bed and nurse demand prediction results  Began literature review of predictive and response models in healthcare and other industries Critical Care Patient Flow Example Bed Demand Output with Validation Our approaches for this work included:    Probability model involving a large-scale convolution Monte-Carlo simulation model embedded in an Excel visual basic model Excel-based simulation and dynamic programming response model Next steps Next steps are to spread these tools in specific applications:    MICU, SICU, CCU, ICU, NICU Emergency departments System-wide flow Potential member benefits 1. Understanding of how to use predictive modeling for bed demand, nurse demand, system utilization, and patient management. 2. Identification of challenges and opportunities. 3. Improved system utilization, costs, flow, and outcomes. NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 2 Identifying Emergency Department Efficiency Frontiers and the Factors Associated with their Efficiency Performance Research team Hyojung Kang, Chris Deflitch, Harriet Nembhard Description To improve the efficiency of care, hospitals have collected performance measures of emergency department (ED) processes and developed initiatives that focus on reducing waiting times. However, using disaggregated measures independently imposes several limitations. In particular, a simple comparison of the performance metrics between different systems can lead to biased conclusions. Data Envelopment Analysis (DEA) can be an effective tool for overcoming the limitations of using a single outcome measure to evaluate ED efficiency among a set of peer groups. Also, the analysis allows hospitals to identify the frontier EDs with an efficient system and benchmark against them. How is this different than related research? Many studies have used time intervals (e.g., door to doctor, door to bed, and length of stay) to measure efficiency of EDs. However, the set of information reflects limited parts of an entire system. Also, a simple comparison of the numbers can lead to inaccurate conclusions when the definitions of the metrics are not the same and when other significant factors affecting the efficiency are not considered. By using a DEA and statistical methods, this study developed an aggregated ED performance measure that incorporates multifaceted aspects of the care system. Milestones achieved to date This study developed DEA models that include three inputs and four outputs. Using the models and a large dataset including over 300 EDs across the nation, we analyzed scale and technical efficiencies of the EDs. The results showed that many EDs operated at less the optimal level. The decomposition of between scale efficiency and technical efficiency indicated that many EDs may need to focus their efforts on re-engineering their processes to utilize key inputs more efficiently rather than modifying the size of their operations to improve overall efficiency. Also, the DEA results pointed out that patient volume was closely associated with scale and technical efficiencies. In the second stage of the study, we investigated the significant exogenous factors associated with EDs’ technical efficiency. Using a multivariate logit model, we identified that several variables, associated with hospital and ED characteristics, had a significant influence on the performance of ED technical efficiency. This analysis provided insights into effective ED benchmarking. *Manuscript under review for publication in European Journal of Operational Research. Next Steps A future study will link the findings from the current DEA models to quality measures in order to investigate the relationship between efficiency and quality. Potential member benefits Siemens, as a NSF-CHOT partner, has identified the strategic priority around proliferating best practices in emergency departments. This project will contribute to increasing knowledge of various factors contributing to efficiency levels. It will also provide effective strategies for ED managers and external healthcare organizations to find comparable ED benchmarking and to design EDs with respect to crucial resources. NSF CHOT IUCRC PROGRESS REPORT – PROJECT # 1 Characterizing and Reducing Avoidable Outside Utilization Research team James Benneyan, Hande Musdal; Parth Vadera, Cory Stasko, Anne-Marie Chouinard Description The objectives of this project are to: 1) to explore the utility of a variety of analytic methods to help understand, characterize, and describe referrals and leakage patterns and 2) to help reduce, disrupt, or prevent leakage. Outside referrals, or “leakage”, is a ubiquitous problem for many health systems, especially accountable care organizations and other health systems with risk-sharing contracts. Leakage occurs when patients within a health system’s population are referred to or otherwise receive care outside that system, with both cost and continuity implications. For various reasons an index referral leads to a chain of additional referrals with unclear patterns and visibility as to how these referrals are occurring. In characterizing leakage, this work develops a flexible multi-phase Bayesian methodology capable of inferring a network from time series patient visit data, with additional phase(s) based on the type and specificity of data available. → ∗P → → | Comparison of Improvement Approaches In reducing leakage, four approaches are compared ranging from a naïve greedy algorithm that would be easily implemented to more difficult to implement genetic algorithms. Project Framework Obj. 1: Characterizing Leakage • Network structure analysis • Data mining to identify signals of costly referrals • Predictive modeling of patient referral pathways Obj. 2: Preventing Leakage • • • • System dynamics model Simulation of flows Network interdiction Comparison of algorithm accuracy and feasibility How is this different than related research? This is the first work of this type in characterizing and preventing outside utilization, using analytical methods from industrial engineering and operations research. Most approaches to managing outside utilization focus on methods to identify inappropriate referrals without considering the complex network flows involved. Other previous work has studied ways of educating providers or effectively introducing new contractual mechanisms. Our project complements this domain of work by applying operations research methods to achieve a network-based understanding of how to characterize, prevent, and minimize leakage. Bayesian Update for Timing and Frequency Data Milestones achieved to date  Developed a system dynamics model for the system of factors that cause leakage  Illustrated network analysis approach to better understand referral patterns  Created Monte Carlo model that simulates a given network-scenario to estimate total costs of the scenario  Developed models for all four network interdiction optimization methods  Compared performance of the four models for various levels of data specificity in terms of leakage reduction, model run time, and model complexity Next steps  Test algorithms on a wider range of possible input data to identify networks for which more advanced algorithms would be most valuable  Identify and partner with health systems to validate and apply both the leakage characterization and reduction models Potential member benefits 1. Better understanding of how and why leakage occurs 2. Identification of potential sources and patterns of avoidable leakage 3. Approaches to detect, prevent, and mitigate avoidable out-of-network referrals NSF’s Online Level of Interest & Feedback Evaluation (LIFE) Forms (For all Industry attendees) NOTE: Access works via either laptop or smart phone STEPS: 1. Go to: IUCRC.COM 2. Select: “CHOT” Meeting 3. Enter Password (ALL CAPS): Scott2015 4. Click on: “IAB” 5. Click on project(s) 6. At a minimum, make a selection to indicate your estimate of your firm’s level of interest in each project. 7. Then, IF you have a comment, question and/or suggestion enter them into the appropriate box(es). To Whom it May Concern: We are pleased to have you as a guest at our Semi-Annual Meeting of the Center for Health Organizational Transformation. Please take a moment to review and sign the attached Confidentiality Statement. Please leave the signed statement at the registration table prior to the opening of the meeting. The statement is designed to help us ensure that our member companies retain the privilege of exclusive access to technical information from presentations and written reports, and to protect the integrity of intellectual property. Thank you for your interest in the Center for Health Organization Transformation and for your respect for our confidentiality policy. Thank you, Bita A. Kash, PhD Texas A&M University Eva K. Lee, PhD Georgia Institute of Technology James Benneyan, PhD Northeastern University Harriet Black Nembhard, PhD Pennsylvania State University Robert Weech-Maldonado, PhD University of Alabama at Birmingham In consideration of my acceptance of the invitation to attend the Bi-Annual Meeting of NSF Center for Health Organizational Transformation (CHOT) on April 16-17, 2015, I,_________________________________________________________, as representative of ____________________________________________, agree that I will keep the information I gather at this meeting confidential. Here “information” shall mean all data and material presented, disclosed, or discussed with the intent that it is for the privilege of the current members of the Center for Health Organization Transformation That is, I will not, without prior consent of the Center for Health Organization Transformation Director, disclose any details, in whole or in part, other than to help company administrators decide if my company should become a center member. Further, I agree that at the conclusion of my participation in this meeting, I will return all printed materials representing unpublished research proposals or results, as furnished by Center for Health Organization Transformation staff at registration to enhance my understanding of meeting presentations. I understand that as my company gives serious consideration to joining the Center for Health Organization Transformation, we may request executive summaries of relevant presentations from the Director. Name: ___________________________________________________________ Signature: ___________________________________________________________ Date: ________________________________________________________________ Title: ________________________________________________________________ Organization: __________________________________________________________

Project Overview and Description

Related documents

Products

Support

Project Overview and Description

Related documents

Add this document to collection(s)

Add this document to saved

Suggest us how to improve StudyLib