The first study of its kind to examine regional and state differences in Medicaid spending shows that there is wide variation in spending per beneficiary, and that some large states are spending more than twice as much per beneficiary as other states of similar size. What’s more, some of the interstate variation in Medicaid spending per beneficiary is due to differences across states in the case-mix of Medicaid enrollees. But after controlling for that case-mix, the variations are influenced more by the volume of services provided to patients than by prices for those services.

The study, published in the newly released July issue of Health Affairs, shows that states in the mid-Atlantic region of the country account for the highest Medicaid spending. Per beneficiary spending in the ten highest-spending states was $1,650 above the national average, most of which ($1,186 or 72 percent) was because of the greater number of services patients received. Spending in the 10 lowest-spending states was $1,161 below the national average, with service use contributing to only $672 or 58 percent of that amount.

Although other studies — most notably those from the Dartmouth Atlas of Health Care — have documented geographic variations in Medicare spending and use, this is the first time researchers have studied the effects of differences in service volume and price on state and regional Medicaid spending.

In the mid-Atlantic region (New Jersey, New York, and Pennsylvania), a combination of high service volume and, to a lesser extent, high prices resulted in the most expensive regional care. In contrast, lower prices and volume of services in the South Central region (Alabama, Arkansas, Kentucky, Louisiana, Mississippi, Oklahoma, Tennessee, and Texas), produced the least expensive care

The authors, who examined Medicaid cash assistance claims data for the period 2001-05 for inpatient hospital services, outpatient services, and prescription drugs, say these large geographic variations in spending offer important clues for states to improve Medicaid quality without increasing costs—and possibly even reducing them. For example, the authors found that the supply of primary care physicians in specific areas was associated with reduced rates of admissions for diabetes, lung disease, and adult asthma. This finding suggests that increased access to primary care providers may result in improved management of common chronic diseases for people on Medicaid.

“Several states are using their Medicaid resources in a way that’s helping to reduce the need for more expensive hospital care,” said lead author Todd P. Gilmer, professor of health economics in the Department of Family and Preventive Medicine at the University of California-San Diego. “This suggests that there is a great deal of room for innovation in Medicaid. By increasing access to primary care and experimenting with team-based delivery models and low-cost providers, states may be able to improve quality while reducing Medicaid spending.”

Washington State provided the best example of how states can lower spending by reducing hospital care and expanding access to primary care providers. Acute care spending there was 18 percent below the national average. Inpatient stays per beneficiary were 35 percent below the national average; in contrast, outpatient visits and prescription fills were each 15 percent above the national average.

Mixed Results Across Regions

The study found significant variation across regions and states:

  • Although Washington State stood out among states for its low hospitalization rates, as a region New England (Connecticut, Massachusetts, Maine, New Hampshire, Rhode Island, and Vermont) had the best results. A higher volume of prescription fills and higher prices for outpatient visits were entirely offset by fewer hospital days and lower prices for pharmaceuticals.
  • Fewer hospital days also offset higher outpatient prices and pharmaceutical fills in several other states, including Hawaii, Iowa, Missouri, North Carolina, Oregon, and Virginia.
  • In the North Central region (Iowa, Illinois, Indiana, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and Wisconsin), higher pharmacy fills and hospital prices were only partially offset by lower pharmacy prices, resulting in higher than average Medicaid costs.
  • In the South Atlantic region (Delaware, the District of Columbia, Georgia, Florida, Maryland, North Carolina, South Carolina, Virginia, and West Virginia), higher hospital volume was offset by lower hospital prices. Whereas Medicaid programs in the Mountain region (Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, and Wyoming) provided lower volumes of services at higher prices.
  • Spending per beneficiary was nearly twice as high in New York as it was in California ($21,195 versus $11,200).
  • The ten highest-spending states were Alaska, Connecticut, the District of Columbia, Illinois, Indiana, Maine, Nebraska, New Jersey, New York, and Ohio. The ten lowest-spending states were Arkansas, Kentucky, Louisiana, Massachusetts, Mississippi, Oklahoma, South Carolina, Tennessee, West Virginia, and Wisconsin.

Market Supply, Program Factors Also Play A Role

The study also explored the influences of market supply factors on hospital admissions. Higher numbers of hospital beds and specialists were associated with higher numbers of hospital admissions, while higher numbers of primary care physicians were associated with reduced hospital admissions for diabetes, lung disease, and adult asthma.

Medicaid program characteristics also influenced hospital admissions. Higher average daily hospital payments were linked with higher admission rates, while higher outpatient visit payments, as well as a higher number of outpatient visits, were linked with lower admission rates.

Gilmer said the findings should help Medicaid directors be more cost effective. “By looking at service mix, access and price, states can find ways to make their programs work better,” he said.