Geographic Variation in Health Care Spending and Promotion of High-Value Care: Interim Report
Posted Mar 22 2013 12:00am
The Healthcare Economist has a new report out for the Institute of Medicine. The report is co-authored with a number of outstanding individuals at Acumen including: Thomas MaCurdy, Jay Bhattacharya, Daniella Perlroth, Anita Au-Yeung, Hani Bashour, Camille Chicklis, Kennan Cronen, Brandy Lipton, Shahin Saneinejad, Elen Shrestha, and Sajid Zaidi. IOM summarizes the goals of the project as follows:
For over three decades, researchers have documented large, systematic variation in Medicare fee-for-service spending and service use across geographic regions, seemingly unrelated to health outcomes. This variation has been interpreted by many to imply that high spending areas are overusing or misusing medical care. Policymakers, seeking strategies to reduce Medicare costs, naturally wonder if cutting payment rates to high cost areas would save money without adversely affecting Medicare beneficiary health care quality and outcomes.
Yet, many have cautioned that geographically-based payment policies may have adverse effects if higher costs are caused by other variables like beneficiary burden of illness, or area policies that affect health outcomes. Further, if there are substantial differences in provider practice patterns within regions, cutting payments to all providers within a region would unfairly punish low cost providers in high-spending regions and unfairly reward high cost providers in low spending regions. Little work has been conducted to evaluate variation in the private insurance market, which represents around 45 percent of United States health insurance expenditures. Even less is known about geographic variation in the Medicaid program and other vulnerable populations.
In 2009, a group of U.S. House of Representatives members, known as the Quality Care Coalition, asked the Secretary of the U.S. Department of Health and Human Services, Kathleen Sebelius, to sponsor two IOM studies focused on geographic payments under Medicare, independent of final health reform legislation. The first study evaluated the accuracy of geographic adjustment factors used for Medicare payment, which alter physician and hospital payments based on specific, geographically-based input prices. This second study asked the IOM Committee on Geographic Variation in Health Care Spending and Promotion of High-Value Care to investigate geographic variation in health care spending and quality and to analyze Medicare payment polices that might encourage high-value care, including adoption of a geographically-based value index, that would modify provider payments based on composite measures of cost and quality of geographic-area performance.
To meet a firm Congressional deadline, the IOM committee is issuing a brief interim report. This report is designed to provide preliminary Committee observations for the 113th Congress as it considers Medicare reform in its first session. This report contains only key preliminary observations predominantly related to the committee’s analyses of Medicare Parts A, B and D, complemented by other work. It does not contain any observations related to the committee’s analyses of the commercial insurer population, Medicare Advantage, or Medicaid, which will be presented in the committee’s final report, due for release summer 2013.
Key findings from the report are the following:
Within-HRR variation in spending and utilization is significantly larger than the across HRR variation for both Medicare and Medicaid. The average standard deviation of per-capita utilization for beneficiaries within each hospital referral region (HRR) is $1,621, whereas the standard deviation of average Medicare utilization levels across HRRs is $84
Post acute care drives much of the regional variation in Medicare spending. In particular, regional variation in home health care is the key reason why many regions–such as McAllen, Texas–have much higher per capita spending levels than the national average.
An HRR’s average use of medical resources, at an aggregate level, has no relationship to quality. For instance, the correlation between the number of risk-adjusted potentially avoidable complications and iatrogenic eventsmeasured using AHRQ’s Patient Safety Indicators (PSI) composite scoreand Medicare price-standardized risk-adjusted spending is almost zero.
Regions that exhibit high utilization of Medicare services do not necessarily have high utilization levels of Medicaid services. The correlation between an HRR’s Medicare and Medicaid resource use (i.e., risk-adjusted, price standardized spending) is -0.07, indicating that HRRs with high per capita Medicare utilization levels are no more likely to have high per capita Medicaid utilization levels than any other HRR. See scatterplot here .
Variation in Spending Across All Beneficiaries
Average Cost : $958
Standard Deviation: $1,695
Average Cost: $1,094
Standard Deviation: $2,767
Cost Savings from Reducing Utilization to Lowest-Cost HRR
$68 billion per year
$24 billion per year
Correlation: HRR Utilization Levels Over Time (2007 – 2008)
Standard Deviation Within vs. Across HRRs
Spending Correlation Across Cohorts (Range of Correlations)
0.23 – 0.95
0.17 – 0.94
Primary Service Type Driving Regional Variation
(including all nursing home)
Correlation: Utilization and Patient
Safety Indicator composite score
Correlation: Medicare and Medicaid Utilization by HRR