1. The Challenge of Practice Variations And the Future of Primary Care 2009 Blanchard Memorial Lecture John E. Wennberg May 1, 2009

2. Dartmouth Atlas of Health Care Hospital Referral Regions In The United States www.dartmouthatlas.org/

3. Unwarranted Variation in Health Care Delivery: Variation that can’t be explained by Illness, medical evidence or patient preferences Variation that can’t be explained by Illness, medical evidence or patient preferences

4. The Three Categories of Unwarranted Variation in Health Care Delivery Effective Care: Effective Care: Evidence-based Care That All with Need Should Evidence-based Care That All with Need Should Receive Receive Preference-Sensitive Care Preference-Sensitive Care Supply-Sensitive Care Supply-Sensitive Care

5. Preference Sensitive Care Effective Care Supply Sensitive Care Proportion of Medicare Spending Attributed to Each Category of Unwarranted Variation

6. Preference-Sensitive Care Involves tradeoffs -- more than one treatment exists and the outcomes are differentInvolves tradeoffs -- more than one treatment exists and the outcomes are different Decisions should be based on the patient’s own preferencesDecisions should be based on the patient’s own preferences But Provider Opinion Often Determines Which Treatment is UsedBut Provider Opinion Often Determines Which Treatment is Used

8. Knee Replacement: An Example of Preference-sensitive Care Ratio of knee replacement rates to the U.S. average (2002-03 ) 1.30 to to1.78 (40) (40)1.10 to < to <1.30 (75) (75) 0.90 to < to <1.10 (120) (120) 0.75 to < to <0.90 (46) (46) 0.36 to < to <0.75 (25) (25) Not Populated

9. Knee replacement per 1,000 Medicare enrollees 1.0 3.0 5.0 7.0 9.011.0 1992-932000-01 Red dot = U.S. average: 4.03 5.64 40% increase

10. Relationship Between Knee Replacement Rates among hospital referral regions in 1992-93 and 2000-01

12. Determining the Need for Hip and Knee Arthroplasty: The Role of Clinical Severity and Patients’ Preferences... Among those with severe arthritis, no more than 15%... Among those with severe arthritis, no more than 15% were definitely willing to undergo (joint replacement), emphasizing the importance of considering both patients’ preference and surgical indications in evaluating need and appropriateness of rates of surgery

13. Conditions involving preference-sensitive surgical decisions Condition Treatment Options Condition Treatment Options Silent Gall stones Surgery versus watchful waitingSilent Gall stones Surgery versus watchful waiting Chronic Stable Angina PCI vs. surgery vs. other methodsChronic Stable Angina PCI vs. surgery vs. other methods Hip and Knee arthritis Joint replacement vs. pain medsHip and Knee arthritis Joint replacement vs. pain meds Carotid Artery stenosis Surgery vs. aspirinCarotid Artery stenosis Surgery vs. aspirin Herniated Disc Back surgery vs. other strategiesHerniated Disc Back surgery vs. other strategies Early prostate cancer Surgery vs. radiation vs. waitingEarly prostate cancer Surgery vs. radiation vs. waiting Enlarged prostate Surgery vs. other strategiesEnlarged prostate Surgery vs. other strategies

15. Which rate is right? Impact of improved decision quality on surgery rates: BPH Knowledge of relevant treatment options and outcomes Concordance between patient values and care received

16. Bottom Line Implication: Clinical Appropriateness should be based on sound evaluation of treatment options (outcomes research) Medical Necessity Should be based on Informed Patient Choice among Clinically Appropriate Options (high quality shared decision making)

17. Thesis # 1: Primary Care Physicians Are the Best Situated Among Health Care Professionals to Take Responsibility for Ensuring that Informed Patient Choice becomes the Standard of Practice for Determining Medical Necessity

18. Supply-Sensitive Care The frequency of use is governed by the assumption that resources should be fully utilized, i.e. that more is betterThe frequency of use is governed by the assumption that resources should be fully utilized, i.e. that more is better Specific medical theories and medical evidence play little role in governing frequency of useSpecific medical theories and medical evidence play little role in governing frequency of use In the absence of evidence and under the assumption that more is better, available supply governs frequency of useIn the absence of evidence and under the assumption that more is better, available supply governs frequency of use

19. Preference Sensitive Care Effective Care Supply Sensitive Care Proportion of Medicare Spending Attributed to Each Category of Unwarranted Variation

20. Association between hospital beds per 1,000 and discharges per 1,000 among Medicare Enrollees: 306 Hospital Regions

21. Association between cardiologists and visits per person to cardiologists among Medicare Enrollees: 306 Regions

22. Supply-Sensitive Care Days in Hospitals per Decedent during Last Six Months of Life Among Patients assigned to the 77 “best” U.S. Hospitals

23. Supply-Sensitive Care Days in Hospitals per Decedent during Last Six Months of Life Among Patients assigned to Academic Medical Centers 8.0 12.0 16.0 20.0 24.028.0 NYU Medical Center27.1 Mount Sinai Hospital22.8 NY Presbyterian Hospital21.6 Cedars-Sinai Medical Center21.3 Mass. General Hospital16.5 UCLA Medical Center16.1 Boston Medical Center15.6 Brigham & Women's Hospital13.9 Beth Israel Deaconess12.2 UCSF Medical Center11.5 Stanford University Hospital10.1

24. Supply-Sensitive Care Physician Visits per Decedent during Last Six Months of Life Among Patients assigned to Academic Medical Centers 10.0 20.0 30.0 40.0 50.0 60.0 70.080.0 NYU Medical Center76.2 UCLA Medical Center43.9 NY Presbyterian Hospital40.3 Mass. General Hospital38.8 Cedars-Sinai Medical Center66.2 Mount Sinai Hospital53.9 Brigham & Women's Hospital31.9 Boston Medical Center31.5 Beth Israel Deaconess29.2 UCSF Medical Center27.2 Stanford University Hospital22.6

25. Days in Hospital and Inpatient Visits per Patient Last 2 years of life for those with Chronic illness (Selected U.S. Regions) Region Days in Hospital Days in Hospital Inpatient Visits Inpatient Visits Los Angeles 28.0 28.0 76.9 76.9 Pittsburgh 24.8 24.8 54.4 54.4 San Francisco 18.7 18.7 34.5 34.5 Denver 14.6 14.6 26.3 26.3 Seattle 13.7 13.7 20.2 20.2 Salt Lake City 11.6 11.6 15.4 15.4

26. Contrasting Practice Patterns in Managing Chronic Illness in Regions (HRRs) Ranked in highest and Lowest Utilization Quintile; (patients in their last 2 years of life) Resource input/Utilization Low HRRs High HRRs High HRRs Ratio H/L Medicare $ per capita Medicare $ per capita$38.300$60,8001.59 Physician Labor/1,000 Physician Labor/1,000 All Physicians All Physicians 16.6 16.6 29.5 29.5 1.78 1.78 Medical Specialists Medical Specialists 5.6 5.6 13.1 13.1 2.35 2.35 Primary Care Doctors Primary Care Doctors 7.4 7.4 11.5 11.5 1.55 1.55 Ratio: MS/PCP Ratio: MS/PCP 0.76 0.76 1.14 1.14 1.50 1.50

27. Contrasting Practice Patterns in Managing Chronic Illness in Regions (HRRs) Ranked in highest and Lowest Utilization Quintile (patients in their last 2 years of life) Low HRRs High HRRs High HRRs Ratio H/L End of Life Care End of Life Care Hospital Days (L6M) Hospital Days (L6M) 8.5 8.5 15.6 15.6 1.83 1.83 Hospital MD Visit (L6M) Hospital MD Visit (L6M) 12.9 12.9 36.3 36.3 2.82 2.82 % Seeing 10 or more MDs % Seeing 10 or more MDs 20.8 20.8 43.7 43.7 2.16 2.16 % Deaths in ICUs % Deaths in ICUs 14.3 14.3 23.2 23.2 1.63 1.63

28. Ratio of medical specialist to primary care physician labor input during the last six months of life 1.21 to to1.93 (51) (51)1.03 to < to <1.21 (60) (60) 0.92 to < to <1.03 (52) (52) 0.80 to < to <0.92 (61) (61) 0.42 to < to <0.80 (82) (82) Not Populated

29. Association between the ratio of medical specialist to primary care physician labor input during the last six months of life and Medicare spending during the last two years of life

30. Per Capita Resource inputs and Health Outcomes: High versus Low Quintiles of Spending Cohort Health Outcomes Survival: Worse Functional Status: Same Satisfaction: worse Perceived Access: Worse Objective Quality: Worse

31. Bottom Line Implication: Reducing overuse of acute care hospitals and growth of organized care should be a goal of public policy as well as clinical policy Bottom Line Implication: Reducing overuse of acute care hospitals and growth of organized care should be a goal of public policy as well as clinical policy

32. Thesis # 2: Primary Care Physicians Are the Best Situated Among Health Care Professionals to Take Responsibility for Coordinating Chronic Care among Providers and Across Sectors of Care: Ambulatory Care, Home Health Care and Institutional Care, including Acute Care Hospitals Thesis # 2: Primary Care Physicians Are the Best Situated Among Health Care Professionals to Take Responsibility for Coordinating Chronic Care among Providers and Across Sectors of Care: Ambulatory Care, Home Health Care and Institutional Care, including Acute Care Hospitals

33. Dartmouthatlas.org Thank You!!!!! Thank You!!!!!