The Future is ANALYTICS PAeHI All Committee Meeting Pennsylvania Medical Society Harrisburg, PA November 14, 2012 2012 PAeHI 1
Presented by Ken Coburn, MD, MPH, President, CEO & Medical Director for HQP www.hqp.com William Buddy Gillespie, PAeHI Chair HIE Committee, Director of HealthCare Solutions for Distributed Systems Services (DSS) www.dsscorp.com William Bill Crowell, Managing Partner at Waypoint Consulting 2012 PAeHI www.waypointco.com
The Increased Demand for Healthcare Analytics
The Healthcare IT Landscape Healthcare organizations have traditionally responded to reporting demands (meaningful use, regulatory, quality, productivity) rather than developing comprehensive, coordinated and forward looking analytics capabilities. This presentation will identify the drivers, demand and challenges to establish and maintain comprehensive and predictive healthcare analytics capabilities in organizations.
Healthcare Analytics Healthcare systems in the Mid-Atlantic region face daunting challenges: Aging populations Chronic illness Resistant strains of infectious disease Revolutionary- but expensive- treatments and technologies Fewer than one third of Healthcare organizations effectively use analytical insights effectively. (Healthcare Analytics) Strategic Healthcare organizations are investing solutions: Boost operational efficiency Improve patient care Control rising costs
The Big Problem Healthcare Costs Spending 2024
The Healthcare IT Landscape Drivers Business and Clinical Drivers Homecare Mobile device/wireless access Restructure of medical records department Data warehousing & analytics Meaningful use all stages HIPAA 5010 EDI Clinical documentation & CPOE
The Healthcare IT Landscape Drivers Business and Clinical Drivers Quality measures patient experience HITECH privacy & security Medical home Accountable care organization (ACO) Health information exchange & direct project Mergers & acquisitions Other
The Healthcare IT Landscape - Demand Why the Demand for Analytics? Provides a 360-degree, predictive, view of the patient and provider Improve the quality and reliability of data Establish data standards for Business & Clinical Terminology and understandable Metadata Source of truth Enhance the Continuum of Patient data across HIEs and Direct Continuity of Care Document (CCD)
The Healthcare IT Landscape - Demand Why the Demand for Analytics? Enhance control over hospital finances & assets Real-time business intelligence Encourages knowledge management within an organization Aligns IT with operations and C-suite Supports goals of Lean Six-Sigma Provides dashboard to support the governance of ACOs Consolidation of financial and clinical data across ACO members CMS reporting requirements Full transparency
HITECH Meaningful Use It s a journey 3 Stages over 5 Years
The Healthcare Landscape - Challenges Challenges Implementing Analytics A lot of data and so little time Approaching petabytes of storage Archiving is weak Slow adoption of data standards Analytics not the strong suit of EMR vendors Selecting the right tools for ETL & analytics Availability of knowledge architects Multiple-desperate repositories of data Best of breed applications Multiple data base structures
The Healthcare IT Landscape 2012 HIMSS Survey Drivers
The Healthcare IT Landscape 2012 HIMSS Survey 2012 HIMSS Leadership Survey Analytics
The Healthcare IT Landscape 2012 HIMSS Survey Analytics
HIMSS Analytics EMR Adoption Model Hospitals continue to move up the HIMSS Analytics EMR Adoption Model Analytics
The Healthcare Landscape Another Perspective Berwick on Analytics: Technology Is Ready, but Doctors Need Help But I thought the most promising was predictive analytics, which was take the data and turn loose the ability to go through it looking for weird patterns. The technology was ready. Don Berwick, MD, former administrator at the Centers for Medicare & Medicaid Services, and prior to that, founding CEO of the Institute for Healthcare Improvement Scott Mace, for HealthLeaders Media, May 15,2012 http://www.healthleadersmedia.com/page-4/tec-280126/berwickon-analytics-technology-is-ready-but-doctors-need-help
2012 Top 10 Trends in Healthcare IT 1. Performance Measurement 2. Population Health Management 3. ACOs and Care Coordination Tools 4. Care Management Transitions 5. Decision Support 6. Privacy and Security 7. Private vs. Public HIEs 8. Imaging 9. Mobile Health 10. Personalized Medicine
Ten Trends in BI and Healthcare Analytics 1. Performance Management metrics 2. Mobile BI / Analytics 3. Discovery of data relationships 4. Social Media and Online Information Sources 5. Big Data 6. Data visualization and analysis technologies 7. Self-service data discovery tools 8. New data resources - to compete on analytics 9. Data Governance 10. Information integration and consolidation
Analytics Maturity Model Predictive Analytics Optimization Predictive Modeling Personalized healthcare Readmission Potential Patient behavior Decision support analytics Simulation Forecasting Enterprise analytics Evidence-based medicine Clinical outcomes analytics Data integration; data warehouse Alerts Query/Drill-down Financial/operational dashboards Clinical data repositories Departmental data marts 20 Transaction reporting Ad-hoc Reporting Basic reporting tools Spreadsheets Standard Reporting Applications report modules 2010 2010-2014 2011-2014 2012-2015
How does Data become better Care? Through ANALYTICS enabling: New Programs New Roles New Tasks Ken Coburn, MD, MPH Proud PAeHI Board Member and CEO & Medical Director Health Quality Partners (HQP)
Brief Background on HQP 501c3, non-profit, in Doylestown, PA dedicated to applied R&D of better health care models HQP s community-based care management: A NEW MODEL An Advanced Preventive Service Proven effective in higher-risk chronically ill Medicare population Medicare Coordinated Care Demonstration (since 2002) Aetna Medicare Advantage (since 2010) Fewer admissions and ER visits, lower cost, and fewer deaths for more info on our results see http://hqp.org 2012 Health Quality Partners, Inc. All rights reserved.
Population-level Analytics Identify at-risk subpopulations to target Sophisticated analyses and precision targeting will NOT compensate for an ineffective intervention An effective, robust intervention yields positive results even when imprecisely targeted Trending (sub)population health status Are we moving the BIG DOTs? Are we healthier? Is care less costly? more accessible? higher quality? If not, why not? Supporting financial analysis for populationbased contracting and gain sharing programs 2012 Health Quality Partners, Inc. All rights reserved.
HQP s Replication and Reliability Platform via cellular/wifi Information Technology Cloud/iPad Mash-up of Supporting Capabilities Patient Education Resources Policies, Protocols & Standards Staff Training Support Database Analytics Educational materials, tools, etc. Field-tested and customized. Version control and review system. Local adaptations possible. Training materials with self-test capabilities to support model roles and tasks. Developed by HQP to best support model roles and tasks. In development real-time decision support. Analysis and statistical process control reporting to support model roles and tasks. Humanware Expertise to Operate, Train, Mentor, and Improve Ongoing development of and access to a pool of human talent to operate and improve the platform and train and mentor health systems, nurse care managers, process improvement specialists, and management staff in diverse geographical areas. Expertise is provided to support others in undertaking the core components of implementation; staff selection, preservice training, consultation & coaching, staff performance evaluation, decision support data systems, facilitative administrative supports, systems interventions (1) (1) Fixsen, DL et al, Core Implementation Components, Research on Social Work Practice, 2009. 2012 Health Quality Partners, Inc. All rights reserved.
Data Alone Does NOT Improve Care New (better) teams, roles, and tasks are required to transform our health system Pushing more data to already overloaded programs and outdated roles will have marginal benefit Leaders must follow the evidence as far as it goes then take a Leap of Faith to create new programs, roles, and tasks for better health care 2012 Health Quality Partners, Inc. All rights reserved.
Necessary, but beyond a point, of marginal benefit DATA DATA DATA DATA PUSH TO Inform Traditional Teams, Roles and Tasks - OR - New, but under-designed or ill-defined Teams, Roles and Tasks DATA Analytics to transform the system DATA DATA DATA Analyze to Transform NEW, well designed and defined Teams, Roles, and Tasks DATA related to team/role function, task execution, program reliability 2012 Health Quality Partners, Inc. All rights reserved.
How do analytics help ensure that teams, roles and tasks perform as intended? Reliable program implementation Real-time identification of defects in protocol adherence Task prioritization and adaptability Support for root cause analysis Proactive preventive care Identify emerging/impending risks and gaps in care Risk-factor management (abnormal labs, BP, behavioral, movement & balance, mood, cognition, nutrition, etc.) Learning Analysis of variation Impact on outcomes (effectiveness) 2012 Health Quality Partners, Inc. All rights reserved.
Situation1: Multiple fragmented data elements re: risk factors make it hard to prioritize patient services. Action: New analysis Composite categories (blue, red, yellow, green) per patient forced by risk element most out of range. Distribution within overall nurse panel and drilldown to individual patient roster. Situation2: Figuring out resource needs and deployment strategy for community-based care management. Background: A new health system partner wants HQP to serve its eligible discharged patients. Who are they, where are they located? What size team is needed? Action: New analysis - geo-mapping patient home residence using hospital discharge data allows HQP to determine locations for staff recruiting, training, and touchdown sites. A Few Examples Situation3: Rapid assessment of all patients immediately after Hurricane Sandy. Background: Gaps in team assets because some nurses were themselves taken out of service by the hurricane. The team had limited real-time adaptability due to a lack of analyses to support rapid identification of patients not contacted since an event date to permit fluid reallocation of patients needing contact to other staff. Action: New analysis, deployed via ipad and linked to HQP cloud database now produces roster of patients not seen since event date parameter with easy reassignment to and tracking of activities undertaken by alternative staff. 2012 Health Quality Partners, Inc. All rights reserved.
Advanced Analytics + Leadership = Unleashing Creative Human Potential With good leadership, the pursuit of EFFECTIVENESS drives the system s continued evolution and creativity A way of thinking more aggressively about coupling organizational purpose to analysis Helping ensure that new teams, roles and tasks are Reliably implemented Provided feedback for continuous evolution and adaptation Improve health outcomes (and cost & quality) 2012 Health Quality Partners, Inc. All rights reserved.