Designing Quality into an EMR/CPOE Implementation Kristine Martin Anderson 1
Healthcare Industry Movements Impacting Hospitals Pay for Performance brings employers, purchasers and providers together to discuss performance metrics External Reporting for Patient Safety and Quality - quality is no longer a passive requirement Focus on Health IT specifically in EHR/CPOE with quality as the expected ROI metric Interoperability and Access data portability initiatives to support both access and informed clinical decisions Enterprise Business Intelligence clinical BI is the hardest and the last to be tackled, but critical to respond to the new business environment 2
Rationale for Public Reporting Consumers Right to Know - health care quality and price should be as transparent as other consumer services Incentive for Improvement stimulate competitive drive Drive Consumers to high value providers First step in Payment Reform, which is a reaction to unsustainable growth in expenditures 3
Hospital Quality Reporting Progress To Date Transparency increased Measure development accelerated Efforts for standardization begun Key constituents engaged in productive dialogue about improvement (NQF) Focus on IT emerged, with urgency for value improvements Documentation improved Early evidence suggests that care is improving, but studies are uncontrolled 4
Public Reporting Increases Activity Source: Judy Hibbard, University of Oregon. 5
Public Reporting is the precursor to Pay For Performance Managed Care P4P? 6
Medicare Spending In 2004, Medicare spending was 2.6% of GDP. Personal health expenditures were 13% of GDP during the same period. Social Security Medicare Medicaid Medicare Trustees expect spending to increase at an average annual rate of 7.5% per year from 2004-2013, except 2006 when Part D launches. With new prescription drug benefits, Medicare spending is projected to grow to 3.4% of GDP in 2006 and just under 4% of GDP by 2013. Billions ($) 2500 2000 1500 1000 500 0 2005 2015 Hospital Insurance benefits are expected to grow by 6% per year from 2006 forward. Source: Congressional Budget Office, August 2005 7
Designing for Action = Standardization 8
Organizations Publicly Reporting Hospital Data Centers for Medicare and Medicaid Services JCAHO State Agencies State Hospital Associations Hospitals and Health Systems Private Organizations The LeapFrog Group HealthGrades.com Consumer Organizations (members) (http://www.checkbook.org/hospital/default.cf m) Insurers/Business Coalitions (members) 9
Role of the National Quality Forum CMS and JCAHO have made strong commitments to use the NQF process to endorse measures for public reporting 285 Members Health Plans Professional Societies Hospitals Researchers Government Agencies Hospital Associations 10
Measures for Public Reporting 11 CARDIOPULMONARY (28) AMI (12) CHF (4) Pneumonia (12) SURGICAL CARE (109) Cardiac surgery (24) Cardiac interventional (6) Vascular surgery (17) Neuro/spine surgery (9) Orthopedic Surgery (23) Colon Surgery (11) Other Surgery (16) Hysterectomy (3) OTHER NON-SURGICAL (2) Acute Stroke GI Hemorrhage HCAHPS (Patient satisfaction) (7) PATIENT SAFETY (57) Care Management Events (7) Criminal Events (4) Environmental Events (5) General Patient Safety (7) National Patient Safety Goals (8) Patient Protection Events (3) Patient Safety Structural Measures (6) Patient Safety Surgical Events (14) Product or Device Events (3) PREGNANCY AND PEDIATRICS (27) Pediatric Asthma (8) Childbirth and neonatal conditions (17) Other surgery (2) NURSING (15) Nursing sensitive care measures (15)
Most Common Measure Sets Being Used in Public Reporting and Pay for Performance Hospital Quality Alliance AMI Heart Failure Pneumonia Surgical Infection Prevention JCAHO Pregnancy Patient Safety Goals AHRQ Patient Safety Indicators LeapFrog (verify) CABG mortality Procedure Volumes NQF Safe Practices Patient Satisfaction 12
Should P4P Measures be a subset of Public Reporting Measures? Should the bar be set higher for P4P measures than for Public Reporting measures? Tighter measure definition? More field testing? Verification that the measure incents the intended behavior modification? Audits? NQF does not currently distinguish between measures appropriate for public reporting and measures appropriate for pay for performance. The debate is ongoing are you at the table? 13
How much of the clinical care of the hospital is being measured? Core Included 16% AMI Pneumonia NonCore 63% Core Excluded 21% Heart Failure Pregnancy SIP 14
Are the patients being measured representative of the general population? Mortality Rates - Core vs. Full Sample Pneumonia Mortality Core Sample HF Mortality AMI Mortality 0% 2% 4% 6% 8% 15
16 Do the signals we get in the measured conditions apply when assessing the overall quality of the hospital?
Outlook for the Future Number of Metrics are going up and up 17
Industry Plans for the Future - Hospital Quality Centers for Medicare and Medicaid Services Committed to Hospital Quality Alliance Committed to NQF Approval Process Hospital Quality Alliance expected direction 30-Day Mortality for AMI and HF measures approved by NQF Initial results to be run for HQA Public reporting in 2006 or 2007 Surgical Care Improvement Project Process and Outcome Measures Rollout in 2006 Subset of measures likely to be chosen for public reporting in 2007 NQF Approval Not Yet Sought Children s Asthma (Candidate Core Measure Set JCAHO) Process and Outcome Measures Expected to seek NQF approval JCAHO Expected to remain aligned with HQA 18
Industry Plans for the Future - Efficiency Debate over definition of efficiency least cost for specified level of quality All parties are seeking credible measures Primary focus on physician payment reform Primary care Specialty Use of NCQA HEDIS measures likely Expect alignment of incentives for hospitals and physicians No clear time horizon for adoption of hospital efficiency measures but research is ramping up Avoidable readmissions Avoidable complications of care 19 BOTTOM LINE: payors will not be paying for avoidable errors in the near future
Hospitals are collecting most of this data manually INTERNAL HOSPITAL COSTS INCURRED TO SUBMIT and INITIALLY ANALYZE CORE MEASURES DATA- 3 MEASURE SETS $100,000 $90,000 $80,000 $70,000 $68,100 $60,000 $59,040 $50,000 $40,000 $43,680 $44,280 $30,000 $20,000 $10,000 $0 Care Science NJ Univ Hospital Midwestern Hospital Oregon Hospital System Subscribers System 20
.. We can t succeed in Pay for Performance without technology 21
Business Issues with Impact Please identify the business issues you believe will have the most impact on health care in the next two years? 22 Increasing Patient Safety/Reducing Medical Errors Patient (Customer) Satisfaction Improving Quality of Care Improving Operational Efficiency Cost Pressures Clinical Transformation (Adopting Clinical Best Practices) Medicare Cutbacks HIPAA Compliance Adoption of New Technology Facility Upgrades/Replacement Providing IT in our Ambulatory Environment Obtaining Capital Nursing Shortage Government Regulation Evidence Based Medicine Increased Competition National Health Information Network (NHIN) Managed Care Fee Reductions 57.30% 43.50% 41.90% 40.30% 39.50% 36.40% 34.80% 31.20% 31.20% 21.30% 20.60% 20.20% 18.20% 16.60% 13.00% 12.60% 12.60% 10.70% HIMSS Survey, 2005
IT Priorities Today Please indicate your organizations top IT priorities today. Implement Technology to Reduce Medical Errors/Promote Patient Safety Upgrade Security on IT Systems to Meet HIPAA Requirements Replace/Upgrade Inpatient Clinical Information Systems Implement Wireless Systems (e.g. wireless LANs) Process/Workflow Redesign 53.10% 44.10% 37.60% 35.20% 31.50% Connecting IT at Hospital and Remote Environments (E.g. Physician s Offices) Implement an Electronic Medical Record (EMR) Train Personnel to use Existing and Newly Installed Systems Upgrade Network Infrastructure (LANs, WANs) Design and Implement an IT Strategic Plan 31.50% 29.10% 25.40% 24.90% 23.90% Implement Enterprise-Wide Applications (e.g. MPI, ERP, Clinical Information Sharing) 23.90% Improvement of IS Departmental Services, Cost Effectiveness and Efficiencies Integrate Systems in Multi-Vendor Environment 22.10% 22.10% Implementing Ambulatory Care Systems (Clinical/Financial/Administrative) 18.30% 23 HIMSS CIO Survey, 2005
IT Priorities Next Two Years Please indicate your organizations top IT priorities in the next two years. Implement an Electronic Medical Record (EMR) Implement Technology to Reduce Medical Errors/Promote Patient Safety Replace/Upgrade Inpatient Clinical Information Systems Process/Workflow Redesign Implementing Ambulatory Care Systems (Clinical/Financial/Administrative) Implement Speech Recognition Systems Implement Enterprise-Wide Applications (e.g. MPI, ERP, Clinical Information Sharing) Connecting IT at Hospital and Remote Environments (E.g. Physician s Offices) Improvement of IS Departmental Services, Cost Effectiveness and Efficiencies Replace/Upgrade Inpatient Financial/Administrative Systems Implement Wireless Systems (e.g. wireless LANs) Integrate Systems in Multi-Vendor Environment 54.40% 41.90% 32.80% 32.40% 29.40% 27.50% 25.50% 21.10% 19.60% 19.10% 18.10% 17.20% 24 HIMSS Survey, 2005
IT Applications Importance Please identify the health care applications areas that you consider most important to your organization over the next two years. Electronic Medical Record (EMR) Bar Coded Medication Management Clinical Information Systems Computer-based Practitioner Order Entry (CPOE) Enterprise-Wide Clinical Information Sharing Digital Picture Archiving and Communications System (PACS) Clinical Data Repository Point-of-care Clinical Decision Support Business Intelligence/Decision Support Systems (e.g. data warehouse) Enterprise Master Patient Index Financial/Administrative Information Systems Web-based Applications Ambulatory Systems Supply Chain Management Enterprise Resource Planning Systems (ERP) 61.70% 55.00% 52.20% 50.20% 43.50% 42.10% 41.60% 37.30% 18.70% 17.70% 17.20% 17.20% 16.70% 15.80% 11.50% 25 HIMSS CIO Survey, 2005
Opportunities for ROI with EMR/CPR/CPOE Installations Accepted quality benefits of EMRs Fewer patient safety events Better compliance with evidence-based medicine Advanced opportunities with EMRs Increase access to clinical data for quality reporting Reduce/eliminate chart review for external reporting Enable concurrent care management 26
Gartner Five Generations of CPR Systems Figure 1. The Five Generations of CPR Systems Source: Handler, Thomas J, MD, Enterprise CPR Systems are Nearing the Generation 3 Milestone, April 11, 2005. 27
Hype Cycle for Healthcare Provider Applications Systems, 2005 28 Source: Runyon, Barry, et al, Hype Cycle for Healthcare Provider Applications and Systems, 2005, July 13, 2005.
Are hospitals expecting to realize ROI that they are not enabling? What are you doing to ensure that your EMR will support your quality reporting needs?... All of our clinical teams are working on creating order sets, nursing documentation and physician adoption strategies.. Won t that do it?? 29
Key Questions for Hospitals Implementing EMRs Do you want to feed your quality systems with clinical data? Do you want to eliminate chart review for core measures? Who on your implementation team knows the specific data content requirements for core and other metric algorithms? Do you know what data can be extracted from your EMR? 30
Best Practices: One Health System s Approach to Designing for Quality The project PLANS for the following changes, which are linked to the system s commitment to quality, enabled by IT: 1. Populate CareScience Quality Manager data mart from clinical systems 2. Calculate clinical performance metrics WITHOUT chart review 3. Transmit clinical performance metrics to the Mercy enterprise Business Intelligence application (for dashboard reporting) 4. Manage quality in real time 31
AMI-6 Beta Blocker on Arrival Eighteen (18) data elements required Admission date Admission source Arrival date Arrival time Birthdate Discharge date Discharge status ICD-9 Principal Diagnosis Code Transfer from another Emergency Department Beta blocker allergy Pacemaker Bradycardia Heart failure on arrival or within 24 hours after arrival Shock on arrival or within 24 hours after arrival Other reasons documented by physician or other for not giving beta blocker Beta blocker name Beta blocker time Beta blocker date 32
For each element, gather Current data location (hard copy or electronic) Documentation source Data format Available electronically now? If so, where? Electronic strategy Alternate data location Final data location Data decision hierarchy needed? Element used in other measures? Frequency of measurement Transformations/calculations Reporting issues Comments 33
Gap Analysis AMI & HF Only 3 of the first 12 measures electronically available after planned implementation The most significant gap is due to lack of physician documentation 34
Partial Automation Strategies to Reduce Chart Review o Example 1: Comfort Measures Only Electronic assistance: ICD-9 code for palliative care = Y o Example 2: Moderate or severe aortic stenosis Electronic assistance: ICD-9 secondary diagnosis codes for aortic stenosis = Y Reduces some, but not all chart review 35
Case Study 2 Hospital HIT Goals An integrated medical record for inpatient, ED and Outpatient An integrated inpatient order entry and pharmacy system Platform for departmental systems High availability Be the leader of a regional health record Focus: patient safety 36
Experience implementing CPOE Added benefits: Physician adoption Eliminated legibility issues Better charge capture Improved usage of order sets and protocols New Risks Orders on the wrong patient Downtime and recovery challenges Generic versus brand-name Information overload Poor education and usage of computers 37
Lessons learned from experience with 100% physician usage Training and Education are key to success Implementation is an ongoing process Interfacing with pharmacy system is not acceptable from a patient safety perspective. The interface engine is translating orders into dispensing systems. 38
Lessons Learned: Using CPOE data in Quality Analysis Data analysis: Orders can be helpful in understanding quality data. Example: DNR order and mortality Used DNR order and timing of it to be understand mortality. Included DNR orders to the CS database Areas of weakness:» Unable to benchmark against other institutions» Uncovered issues related to the implementation Lesson learned: Implementation must include PI data requirements, not only the type of data but also the content and structure. 39
Lessons Learned: Using CPOE data in Quality Analysis Implementing quality and safety initiatives: almost always require operational changes often including medical staff and nursing practices. Example: compliance with Pneumovax Determined that using a common admit order set is the best way to resolve this issue Obtained support from Med Exec to mandate common order set for all adult inpatient admissions Admit order set includes an order for Nursing to assess and dispense if appropriate Obtained support from Nursing to adopt assess and dispense orders Challenge: determine whether the patient was admitted under order set versus compilation of orders Lesson learned: some indicators will have to be included as data elements during implementation to identify situations such as using order sets and bundles (IHI) 40