Navigating the Future of Healthcare A New Model of Care Delivery

Navigating the Future of Healthcare: A New Model of Care Delivery Kenneth H. Paulus Diane Gilworth April, 2008 Agenda • Review Allina Structure • Discuss market and key drivers for change • Review the Allina and Harvard Vanguard strategy and patient care model • Key Imperatives What is Allina Health System? • 501c3 non-profit organization • $2.5B health care delivery organization covering the state of Minnesota and western Wisconsin • 23,000 employees • Two tertiary hospitals, three community hospitals, six rural hospitals, a 600 physician clinic with 50 sites What is wrong with health care in the States and what is being done about it? • Costs are „out of control‟ • Quality is not optimal, with no real breakthrough in sight • Value is significantly behind that of other developed countries • There is an opportunity for a significant breakthrough, and it is likely to happen in the next 10 years Costs continue to rise at a steady pace… How are costs changing by diagnosis? Cardiac, Orthopedics, Cancer are the Largest Cost Drivers em e st ss u Sy Ti v y o r nc t m at te ul , C irc u lo s S ys - C k sm ry s us 07 M pl a ri na ion i ng - eo u t rp di on o 13 ue - N enit Con ois P em th,P ans 02 G er r st nd rg th bi 10 O ry a S y il d e O h u ve ns un ,C 17 Inj sti m cy , Se m e ig s an ,Im te 16 - D egn Sy ys oli c 09 Pr ou s ry S tab ue e rv to ss 11 Ne ira ri,M Ti p ut d s es es rio ou ,N 06 Pe ane Cod - R dcr t al 08 En is at bc u /E D in es tic er Su 03 si od & -P C ra s s n ki ie 15 al Pa an al rg du & -S m O si us 12 Re no m tio l A For ec ta 18 d ni nf -I oo ge rs Bl on 01 & de - C od or is 14 lo lD -B ta 04 en n w -M no nk 05 99 -U $30.00 $25.00 $20.00 $15.00 $10.00 $5.00 $0.00 20 03 20 Q4 04 Q 20 1 04 Q 20 2 04 Q 20 3 04 Q 20 4 05 Q 20 1 05 Q 2 20 05 Q 3 How is the industry responding to this challenge? 1. Pass on financial risk and responsibility to the consumer/patient 2. Hold providers accountable for quality and costs via „pay for performance‟ reimbursement Employers are shifting costs to employees $3,552 Dramatic increase in out-of-pocket costs for insured employees $861 $2,276 +56% Co-pay, deductible $580 (wage growth = 12.3%) $2,691 Contribution to premium $1,696 2000 2004 Early results of cost shifting to patients… • Significant reduction of Behavioral Health visits – 30% • Sizeable reduction of physician office visits – 15% • Slight reduction of ED visits – 5% • Virtually no impact on hospital utilization Pay for Performance: Does physician practice variation contribute to outcomes? Distribution of Interventional Cardiologists: Cleveland 120 115 110 Quality Score 105 100 95 1 1 1 3 1 90 85 80 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 Efficiency (Cost Compared to Market Average) Bubble size reflects number of cases seen by MD One view of physician performance tracking… 11 Source: UHC What are the key factors that drive good hospital performance? Continuous Quality Improvement Emergency Cardiac Care 120 Min Street to PCI Operator Experience Inv. Card:125/yr Surg:125/yr EP:50/yr Performance Measurement (STS , ACC, & JCAHO) Facility Experience- Minimums PCI=400/yr Open Heart:450/yr Full Service (Surgical, Interventional Cardiology, EP) UHC Cardiac Services Trend $24.00 $22.00 How much does performance vary? UHC Cardiac Services Center of Excellence Trend Value $24.00 Non-COE Trend COE Trend UHC Blended Trend Cardiac Services PMPM $20.00 $18.00 $16.00 $14.00 $22.00 $12.00 $20.00 Apr-04 Au Cardiac Services PMPM g0 O 3 ct -0 De 3 c0 Fe 3 b0 Ap 4 r- 0 Ju 4 n0 Au 4 g0 O 4 ct -0 De 4 c0 Fe 4 b0 Ap 5 r- 0 Ju 5 nAu 0 5 g0 O 5 ct -0 De 5 c05 18% Trend $18.00 $16.00 $14.00 3-4% Trend $12.00 Aug-03 Aug-04 Apr-05 Aug-05 Oct-03 Oct-04 Dec-03 Dec-04 Oct-05 Source: UHC Dec-05 Jun-04 Feb-04 Feb-05 Jun-05 Other complicating factors… Estimated Physician Supply and Demand, 2020 1,100,000 # of physicians 1,050,000 1,000,000 950,000 900,000 1,060,000 972,000 850,000 800,000 750,000 Number of Physicians Projected 2020 Number of Physicians Needed by 2020 A 15% increase in U.S. medical graduates by 2015 is required to offset the shortfall. Other complicating factors… National Supply and Demand Projections for FTE RNs 2000 - 2020 RN FTE Demand 2,900 2,700 2,500 FTEs (Thousands) Shortage of over 2,300 2,100 1,000,000 nurses in 2020 RN FTE Supply 1,900 1,700 1,500 2000 2005 2010 2015 2020 First, attack the problem from a population perspective… Cost Acuity Second, break down the problem/opportunity into manageable components… The Well The Chronically Ill The Acute End of Life 1st Imperative: Prevent illness efficiently • There is not enough money in the system to manage wellness using physician or nurse time • The Well • Information technology has the potential to drive significant improvements in health very inexpensively – – On-line risk assessment with automated link to provider team Regular „alerts‟ to patients to drive health improvement The assessment of health and genetic risk will start with the patient, and the transfer of knowledge will provide the tools for improvement 2nd Imperative: Aggressively manage chronic illness through the primary care team • The management of chronic illness should be the domain of the primary care team Provide these teams the information needed to assess their panel of chronic illness patients and measure performance The Chronically Ill • • Re-define the roles of the team to push simple chronic illness care and education to the nurse/NP Develop centralized and virtual disease management programs that improve outcomes efficiently • 3rd Imperative: Concentrate acute and tertiary care at Centers of Excellence • Drive tertiary care centrally to reap the benefits of volume and expertise • Establish clear performance metrics for outcomes and cost • Link acute care Emergency and ICU with community providers effectively – Level I program – eICU The Acute 4th Imperative: Provide effective critical and palliative care • A significant percentage of costs reside in care provided in the last six months of life • An effective palliative care program reduces costs and improves perceived quality • The use of data to track the critically ill provides an opportunity for significant savings • A program of targeted home and hospice care can be centralized effectively End of Life 5th Imperative: Integrate care across disease states – Cardiology, Cancer, Neurosciences, Rehab… • Profile provider and team performance by episode and disease treatment effectiveness • Integrate disease management activities across the entire care system – Complex Care (CCC/CHF) program • Drive to high levels of evidence based care by specialty using EMR and data warehouse • Use nurse call center as integrator Patient Care Integration 6th Imperative – Build the patient care model on a solid foundation • Engaged Employees: 65% of employees rate engagement as excellent, 90% rate it as good/excellent Physician Collaboration: Two models of collaboration – independent alliance/or employment. Must meet standards of evidence based medicine and use standard order sets 70% of time Knowledge & Information: Development of a data warehouse and „knowledge utility‟ that is available to all components of the system – providers and patients alike (MyChart) • • • Performance Excellence: 90th percentile of national HCAPS standard for patient satisfaction Engaged Employees Physician Collaboration Knowledge & Information Performance Excellence Complex Care and Congestive Heart Failure Program inadequate pumping increased stress on heart decreased blood flow to end organs increased circulating volume compensatory mechanisms Na+/water retention Etiology of Congestive Heart Failure inadequate pumping increased stress on heart Diabetes increased circulating volume High Blood Pressure Coronary Artery Disease decreased blood flow to end organs compensatory mechanisms Renal Insufficiency Obesity Na+/water retention Hospital Admits/ER visits CHF/CCC Program Patients 2004- 2005 80 70 60 number of a dmiss io ns 50 40 30 20 10 0 2005 All other Admissions 2004 Heart Failure Admissions Jan Feb Mar April May June July aug Sept Oct Nov Dec Changing the Paradigm Disease Management CHF Diabetes CHF CCC PCP Diabetes CRI CAD CRI CAD Care of patient with multiple diseases Traditional Care Patient Centered Care Model Hospital Programs ECF Program Hospital Programs ECF Program Case Management CHF/CCC Program CCC Case Management Primary Care Primary Care Specialty Care Specialty Care Goals of the CCC Program • Improve care experience of patients with multiple co-morbid and/or high-risk conditions. • Improve clinical outcomes and patient quality of life by providing care that is: – coordinated across the practice, – maintains continuity and frequent contact, – excels in patient education and self-management • Reduce overall cost of medical care Intervention • CHF to CCC in 6 sites in 2006 • CHF NP‟s trained -DM, CRI, geriatrics • RN call center-outbound phone surveillance between visits • Social work, nutrition • Collaboration– Cardiology – Nephrology – Endocrinology – Primary Care cards Eligible Patients 2 or more the following co-morbid conditions Primary Disease Hospitalizaiton with 12 months Clinical CHF none systolic or diastolic dysfunction CRI Diabetes CAD none 2 or more for CAD, 2 or more and/or and/or creat Cl < 60 Ha1c > 8.0 SBP > 140 LDL > 100 SBP > 140, Ha1c > 8.0, LDL > 100, PVD Assumption: Improving Clinical Processes and Intermediate Outcome Metrics Will Lead to fewer hospitals and reduced costs Risk reductions occurred early and were maintained Macrovascular end-points 20% overall reduction in cardiovascular events higher than any other single intervention study Microvascular end-point achieved at 4 years were maintained N Eng J. Med 3485.5 Jan 30, 2003 Groups • Control: Patients never enrolled in CHF/CCC program • CCC Only: Patients not enrolled in the CHF program 2003-04, but enrolled in CCC program in 2005 • CHF Only: Patients who were enrolled in CHF program from 2003-05. Not enrolled in the CCC program in 2005 • CHF-CCC: Patients enrolled in CHF program from 2003-05 and then enrolled in the CCC program in 2005 Basic Demographics Control CCC only CHF only CHF to CCC Total Patients 347 138 408 76-89% 302 75-86% % Capitated (Varies by Year) 66-90% 65-76% % Female Mean Age (Median) 53% 76 (77) 42% 71 (73) 48% 72 (74) 52% 74 (76) DxCG Scores Control 2003 13.96 CCC only 7.96 CHF only 12.24 CHF to CCC 11.72 2004 2005 15.24 15.62 10.86 19.06 14.40 16.51 13.17 16.56 2005 DxCG ACC Groups (Partial List) 2005 Cohort DxCG Score Heart Lung Control CCC Only CHF Only CHF-CCC 15.62 96% 50% 19.06 100% 57% 16.51 97% 42% 16.56 100% 44% Diabetes Hematologic 41% 37% 53% 41% 46% 40% 47% 44% Gastrointestinal Neurologic Vascular 39% 13% 31% 48% 22% 36% 44% 17% 33% 41% 22% 35% Cerebro-Vascular Urinary System 14% 53% 18% 53% 11% 49% 11% 46% Infectious and Parasitic Malignant Neoplasm Benign/In Situ/Uncertain Neoplasm 24% 19% 9% 36% 27% 9% 33% 26% 10% 32% 20% 10% Mental Cognitive Disorders Cardio-Respiratory Arrest 22% 13% 11% 30% 8% 12% 23% 8% 10% 19% 9% 8% Process Outcomes % DM/HTN/Lipid Pts with Rx for ACE/ARB 100% 90% 80% 70% 60% 50% 40% Control CCC Only CHF Only CHF-CCC Program % Pts with Rx 2003 2004 2005 Process Outcomes % DM/HTN/Lipid Pts with Rx for Beta Blocker 100% 90% 80% 70% 60% 50% 40% Control CCC Only CHF Only CHF-CCC Program % Pts with Rx 2003 2004 2005 Process Outcomes % DM/HTN/Lipid Pts with Rx for Statin 100% 90% 80% 70% 60% 50% 40% Control CCC Only CHF Only CHF-CCC Program % Pts with Rx 2003 2004 2005 Blood Pressure Screening % HTN Pts with No BP Screening % Pts No Screening 20% 15% 10% 5% 0% Control CCC Only CHF Only CHF - CCC Program 2003 2004 2005 LDL Screening % Lipid Pts with No LDL Screening % Pts No Screening 20% 15% 10% 5% 0% Control CCC Only CHF Only CHF - CCC Program 2003 2004 2005 HgA1c Screening % DM Pts with No HgA1c Screening % Pts no Screening 10% 8% 6% 4% 2% 0% Control CCC Only CHF Only CHF - CCC Program 2003 2004 2005 Blood Pressure Control Median SBP 135 130 SBP Median DBP 78 76 2003 2004 2005 DBP 74 72 70 68 66 2003 2004 2005 125 120 115 Control CCC Only CHF Only CHF - CCC Program Control CCC Only CHF Only CHF - CCC Program LDL Control Median LDL 120 100 80 2003 2004 2005 LDL 60 40 20 0 Control CCC Only CHF Only CHF - CCC Program HgA1c Control Median HgA1c 7.8 7.6 7.4 7.2 7 6.8 6.6 6.4 6.2 7.6 7.2 7 6.7 6.8 7.15 6.9 7.1 2003 2004 2005 HgA1c Control CCC Only CHF Only CHF - CCC Program The Need for Tight Glycemic Control According to UKPDS 35, every 1% drop in HbA1c was associated with: 21% 14% 12% Decrease in risk of stroke 16% Decrease in risk of MI Decrease in any diabetesrelated endpoint P<.0001 Decrease in risk of HF 37% P=.035 P=.016 Decrease in risk of microvascular complications P<.0001 P<.0001 United Kingdom Prospective Diabetes Study (UKPDS) 35 was a prospective observational study of 4,585 DM patients with a mean a ge of 53 years. Median follow-up time (duration of diabetes) was 10 years. HF=heart failure; MI=myocardial infarction. Stratton IM, et al. BMJ. 2000;321:405-412. EPIC-Norfolk Study: Even Small Changes in HbA1c May Impact Mortality Patients with HbA1c 5.0–6.9* 0.1% decrease in HbA1c may result in: 0.2% decrease in HbA1c may result in: 12%† 25%† Reduction in Mortality *Over 80% of population excess mortality associated with HbA1c concentrations above 5% occurred in 70% of the population with HbA1c concentrations of 5.0%–6.9%. †Estimated percentages. EPIC-Norfolk: Norfolk cohort of the European Prospective Investigation into Cancer and Nutrition was a prospective population study of 4,662 men (diabetic and nondiabetic) aged 45-79 followed for 2–4 years. Khaw KT, et al. BMJ. 2001:322:15-18. Disease Progression hospitalization exacerbation CCC enrollment Baseline health status New baseline health status Clinical course- Time Hospital Utilization Adjusted Hospital Utilization Admissions/1000 pts per DxCG Unit 80 70 60 50 40 30 20 10 0 Control CCC Only CHF Only CCC to CHF Groups 2003 2004 2005 ED Utilization Adjusted ED Utilization Visits/1000 pts per DxCG unit 40 35 30 25 20 15 10 5 0 Control CCC Only CHF Only CHF to CCC Groups 2003 2004 2005 CCC Program Enrollment 1997- 2006 2000 1700 1400 1100 800 500 200 26 25 -100 Fe b M ar Ap r M ay Ju n Ju l Au g Ja n 925 total patients Active pts new patients enrolled inactive pts deaths 2005 enrollment 215 patients 36 30 27 29 Se p O ct No v De c 2006 all numbers represent end of month totals Hospital Admits/ER visits CHF/CCC Program Patients 2004- 2006 80 70 60 number of a dmiss io ns 50 40 30 20 10 0 Jan Feb Mar April May June July aug Sept Oct Nov Dec 04 non CHF 05 non CHF 06 non-CHF 04-CHF 05-CHF 06 -CHF Bottom Line • • • Patient Satisfaction – Mr. P Provider Satisfaction – Similar to CHF Clinical Quality – • Greatly improved screening • Greatly improved treatment to target Utilization / Financial • Minimal Change Hosp/SNF Utilization • Decrease in ED Utilization • What will the model look like in the community? The Community Integrated with knowledge, information, and disease management with care provided appropriately, telephonically, electronically and virtually Primary Care Primary Care Primary Care Regional Specialties Community Hospital Regional Specialties/ Community Hospital Primary Care Primary Care Tertiary Center Regional Specialties/ Community Hospital Primary Care Final Thought… “If you think you can run an organization in the next 10 years the way you ran it in the last 10 years you are out of your mind” CEO Coca Cola