Medicare should employ new data sources, methods to ensure accuracy of geographic adjustments to payments

Geographic adjustments to Medicare payments are intended to accurately and equitably cover regional variations in wages, rents, and other costs incurred by hospitals and individual health care practitioners, but almost 40 percent of hospitals have been granted exceptions to how their adjustments are calculated, finds a new report from the Institute of Medicine. The rate of exceptions strongly suggests that the mechanisms underlying the adjustments are inadequate, noted the committee that wrote the report. The rationale for fine-tuning Medicare payments based on geographic variations in expenses beyond providers' control is sound and should be continued, the committee concluded. However, several fundamental changes to the data sources and methods the program uses to calculate the adjustments are needed to increase the accuracy of the payments.

"The Medicare program needs more precise and objective tools and methods to assure the nation that the billions being spent are appropriately and fairly disbursed," said committee chair Frank Sloan, J. Alexander McMahon Professor of Health Policy and Management and professor of economics, Duke University, Durham, N.C. "As the criticism we heard from a range of health care providers indicates, there is significant skepticism about the fairness and accuracy of how adjustments are currently being determined. This report's recommendations will increase the likelihood that the geographic adjustments reflect reasonably accurate measures of regional differences in expenses."

Medicare payments to hospitals and health professionals working in private practice topped $500 billion in 2010, according to Congressional Budget Office estimates. Federal law requires geographic adjustments to be budget neutral, meaning any increase in the amount paid to one hospital or practitioner must be offset by a decrease to others.

Salaries and benefits make up one of the largest costs of providing care. The Medicare program should use health sector data from the Bureau of Labor Statistics (BLS) to develop its indexes for calculating wage adjustments for hospitals and private practice health professionals, the report says. BLS data are a more accurate, independent, and appropriate source than the hospital cost reports, physician surveys, census data, and other information currently used, the committee said. Congress will have to revise a section of the Social Security Act to enable this change.

Medicare should take into account median wage data for all types of workers in private practice settings and hospitals to calculate payments, the report adds. Currently, regional wage differences are based on data for registered nurses, licensed practical nurses, health technicians, and administrative staff only, which does not reflect the full work force in many practices or hospitals.

Medicare also adjusts payments according to which labor market a hospital or practitioner operates in and competes for workers. Because hospitals and health professionals in a given area tend to function within the same local market, there is no reason for the program to use one set of 441 markets to determine hospital payments and a different set of 89 markets for practitioner adjustments, the report says. Instead, the program should employ the metropolitan statistical areas (MSAs) developed by the Office of Management and Budget for both. MSAs reflect information on where people live and work and decisions made by employers and employees that define labor markets' boundaries, the report notes.

Although MSAs reasonably approximate local labor markets, hospitals and clinics on the borders of neighboring MSAs may compete for the same pool of workers yet receive significantly different adjustments based on the average wages in their respective labor markets. Commuting patterns of health care workers can capture the economic blurring of labor market boundaries and should be used to smooth out any dramatic differences, the report says.

Commercial rent information would provide a more accurate assessment of variations in the price of office space than information on median subsidized rents for a two-bedroom apartment, which is what the Medicare program currently relies on, the report adds. Because no sources of commercial rent data have the broad geographic coverage necessary, a new source should be developed.

The report is the first of three to be issued by the committee. A supplemental report that discusses physician payment issues further will be issued this summer. A final report to be released in 2012 will present the committee's evaluation of the effects of the adjustment factors on health care quality, population health, and the distribution of the health care work force.

Source: National Academy of Sciences