the performance of the us health care system what s right
play

The Performance of the US Health Care System: Whats Right, Whats - PowerPoint PPT Presentation

The Performance of the US Health Care System: Whats Right, Whats Wrong and What (If Anything) Can Fix It? Robert Town University of Texas-Austin / NBER October 10, 2019 Goal of the Talk Well-known that the US health care system


  1. The Performance of the US Health Care System: What’s Right, What’s Wrong and What (If Anything) Can Fix It? Robert Town University of Texas-Austin / NBER October 10, 2019

  2. Goal of the Talk Well-known that the US health care system underperforms Health policy is front and center in the Democratic primaries Goal of the Talk: Level set by discussing the evidence on the dimensions of underperformance and its underlying causes Conclude by examining the likely impact of Democratic proposals What problems do they address and at what costs?

  3. We Spend an Enormous Amount on Health Care

  4. Implies that Per-Capita Expenditure is Equal to a Buying Good Used Car Every Year

  5. As Percentage of GDP , Health Care Spending has Grown Dramatically

  6. Directly Impacts Premiums

  7. Uninsured Rates are High

  8. Total Expenditure = P × Q . Is it Price or Quantity or both? Source: HCCI (2019)

  9. Or, as Observed in 2003

  10. Is High (and Increasing) Health Care Spending Necessarily Bad? Not necessarily Depends on what benefits come with the increase in health care expenditures

  11. Life Expectancy Has Been Increasing

  12. US Cause of Death Trends

  13. Infant Mortality Declines Important in Explaining Life Expectancy Increases

  14. Does the Reductions in Mortality offset the Increase in Health Expenditures? Increase in life expectancy over last decade is 10 months Back of the envelope calculation suggests ROI in health spending is roughly between 0 and 6% Implies that providers/medtech/insurers appropriating most of the gains from innovation

  15. Large Variation in Medicare Per-Capita Expenditures Across Geography: Dartmouth Atlas Results

  16. Variation in Expenditures Is Not Correlated with Better Outcomes (Tsunga, et al. (2017))

  17. Comparison to Other Countries

  18. OECD Health Outcomes Comparisons

  19. OECD Workforce Comparisons

  20. OECD Access and Quality Comparisons

  21. Why are Prices So Much Higher in US? OCED provider reimbursement rates are highly regulated US privately insured reimbursement rates are determined by negotiations between insurers and providers Prices primarily determined by relative bargaining leverage Leverage depends upon both provider and insurer market structure – how competitive is the market? Provider incentives (“agency”) also lead to inefficient utilization – difficult to generate payment incentives that induces providers to always do the right thing at the right time in the right place

  22. US Provider Markets are Concentrated

  23. How To Increase the Performance of the Health Care Sector? 5 approaches to improving health care system performance: 1 Create more provider competition and/or introduce more provider price regulation 2 Reduce delivery of low value care Difficult to set up such a system that providers would prefer to the current system 3 Increase the productivity of care/medtech where benefits are primarily captured by patients 4 Expand public programs for uninsured 5 Improve population health Meaningfully improving health sector performance likely requires reducing provider income

  24. Democratic Presidential Proposals Cottage industry of health policy interventions: Accountable care organizations, electronic medical records, bundled payments, value-based payments, pay-for-performance, etc Most of these have had little to no impact Democratic health reform proposals Public Option Offer a plan that looks like Traditional Medicare – and price it at approximately cost Expand Affordable Care Act provisions Medicare-For-All Enroll all eligible US residents in an enhanced Medicare plan Mostly eliminate private health insurance Pay providers at Medicare rates (which may have to adjust) Medicare reimbursements approximately 70% of commercial insurance reimbursements

  25. Democratic Presidential Proposals Impact of Democratic Healthcare Proposals Impact Public Option Medicare-for-All Competition and/or price regulation Modest Yes, but not to current Medicare Rates Reduce low value care No No Impact on technological change Modest Unclear Improve Population Health No No Impact on Uninsurance Modest to Signficant Eliminate Disruption Little Large Medical Care Industry Response Negative Going to War!

  26. Medicare-for-All Projected Costs

  27. Summary US health system problems are many fold but the cost problem is central and first-order Solutions that do not address the underlying prices of health care will not meaningfully impact the cost problem Even Medicare-For-All likely will not fully address the cost issue What about Medicare-Advantage-For-All?

  28. The Performance of Interventions to Improve US Health Care Sector Performance A cottage industry has emerged to address health care system inefficiencies Examine 3 important initiatives: Accountable Care Organizations (Medicare Shared Saving Program) Health Information Technology (e.g. EMRs) Managed Care Return to the recent trend in health care costs shift

  29. Accountable Care Organizations ACOs are organizations that are formed and tasked with managing the care of assigned patients – payments to the organization are tied to quality and cost metrics Generally, savings are split between the payer and the ACO Medicare, Medicaid, and Private ACOs The Affordable Care Act authorized a Medicare ACO demonstration (Medicare Shared Savings Program) Cost and quality improvements were modest and it was revenue negative from Medicare’s perspective

  30. Medicare Shared Savings Demonstration Results

  31. Health Information Technology Paper records were very common in health care sector until early-2000 Lots of potential efficiencies from having a centralized, electronic store of medical information (reduce duplication of tests, promote best practices, flag at-risk patients, identify drug/drug interactions, etc) The turn of the century saw an increased diffusion of HIT in hospitals (e.g. EMRs, CPOE, PACS, eMAR) McCullough, Parente and Town (2015) analyze hospital adoption of HIT on patient outcomes

  32. Impact of HIT on Patient Outcomes 0.015 0.015 PN AMI Change in 60-Day Mortality Rate Change in 60-Day Mortality Rate 0.01 0.01 0.005 0.005 0 0 1 2 3 4 5 6 7 8 9 10 -0.005 1 2 3 4 5 6 7 8 9 10 -0.005 -0.01 -0.01 -0.015 -0.015 -0.02 -0.02 -0.025 -0.03 -0.025 Severity Deciles Severity Deciles 0.01 0.15 CHF STEMI Change in 60-Day Mortality Rate Change in 60-Day Mortality Rate 0.005 0.1 0 0.05 1 2 3 4 5 6 7 8 9 10 -0.005 0 -0.01 1 2 3 4 5 6 7 8 9 10 -0.05 -0.015 -0.02 -0.1 Severity Deciles Severity Deciles 0.01 CA Change in 60-Day Mortality Rate 0.005 0 1 2 3 4 5 6 7 8 9 10 -0.005 -0.01 -0.015 -0.02 -0.025 Severity Deciles

  33. Rise of Managed Care From 1993 through 1999, personal health care price growth averaged 2.5 percent – much lower than any other period

  34. Why Did 2 out of 3 of these Programs Fail to Achieve Significant Cost Reductions / Quality Improvements? ACOs focused on the MD incentive contract alone – the easy way to make $ in these settings is to game risk-adjustment not invest in cost reduction/quality improvement HIT affects information acquisition cost and access but does not address bargaining power or incentive problems Managed care addresses both bargaining power and contract form – Providers hated it (and it lead to a hospital merger wave) and the cost benefits are opaque to patients

  35. To Summarize: Or, as Observed in 2019

  36. Additional Slides

  37. OECD GDP Spending on Health

  38. Growth Rates of Expenditures by Source Over Time GDP ¡ Prescrip8ons ¡ Hospitals ¡ Physicians/Clinics ¡ 16.0% ¡ 14.0% ¡ 14.0% ¡ 12.8% ¡ 12.8% ¡ 12.8% ¡ 11.6% ¡ 12.0% ¡ 10.3% ¡ 9.6% ¡ 10.0% ¡ 8.2% ¡ 7.8% ¡ 7.6% ¡ 8.0% ¡ 7.0% ¡ 6.2% ¡ 6.0% ¡ 5.6% ¡ 6.0% ¡ 5.2% ¡ 4.8% ¡ 4.2% ¡ 3.8% ¡ 3.9% ¡ 4.0% ¡ 1.9% ¡ 2.0% ¡ 0.0% ¡ 1970s ¡ 1980s ¡ 1990s ¡ 2000s ¡ 2010-­‑2013 ¡

  39. OECD Utilization Comparisons

  40. But the Trend Is Flattening for the Elderly Source: Cutler et al. (2019)

  41. Bending the Trend for the Elderly Source: Cutler et al. (2019)

  42. Bending the Trend for the Elderly Source: Cutler et al. (2019)

  43. OECD Pharma Comparisons

  44. OECD Utilization Comparisons

  45. US v OECD Infant Mortality

Recommend


More recommend