California's Shift to Medicaid Managed Care Doesn't Save Money or Improve Outcomes
From 2002 to 2004, Mark Duggan, Ph.D., at the University of Maryland after doing initial work at the University of Chicago evaluated how county-level mandates requiring most Medicaid recipients to enroll in a managed care plan affected government spending and health outcomes in 20 California counties.
This project was part of the Robert Wood Johnson Foundation (RWJF) national program Changes in Health Care Financing and Organization (HCFO) (for more information see Grant Results).
- Following the shift from fee-for-service to managed care, Medicaid spending increased an average of 17 percent, an effect that lasted well after the shift, suggesting that startup costs were not the cause of the increase.
- Counties with only one managed care plan experienced significantly greater spending increases than those with multiple plans, suggesting a benefit to competition.
- Significant improvements in health outcomes did not result.
- Infant health outcomes showed little change.
- Although the study did not produce administrative cost data, anecdotal evidence suggests that the shift to managed care may result in increased state administrative costs.
RWJF supported this project through two grants totaling $69,754.
In the 1990s, in an effort to control rapid growth in Medicaid spending, the state of California shifted most Medicaid recipients county by county from fee-for-service Medicaid to managed care plans. As a result, Medicaid managed care enrollment increased from under 12 percent in 1993 to 51 percent in 1999. The state paid plans a fixed monthly amount per recipient and the plans paid hospitals, physicians and other health care providers. The recipients had no out-of-pocket expenses.
After initial work Duggan conducted at the University of Chicago, he and researchers at the University of Maryland studied the effect of California's changes to its Medicaid program on spending and health outcomes among Medicaid recipients in the state.
They examined mandates in 20 California counties that required particular categories of Medicaid recipients to enroll in managed care plans. Researchers calculated individual-level Medicaid spending in six-month intervals between 1993 and 1999, for a random sample of 20 percent of Medicaid recipients in the 20 counties studied. Most recipients were from the Aid to Families with Dependent Children (now CalWORKS, California Work Opportunity and Responsibility to Kids) population.
The sample totaled about 1.2 million. Project staff also investigated differences across three managed care models in place in the 20 study counties:
- Geographic managed care several competitive commercial plans.
- A county-organized health system a nonprofit community-based plan.
- Two-plans competition between one commercial plan and one nonprofit, private Medicaid-only plan.
Project staff reported the following findings in its findings brief:
- Following the shift from fee-for-service to managed care, Medicaid spending increased an average of 17 percent, an effect that lasted well after the shift, suggesting that startup costs were not the cause of the increase. The principal investigator offered several explanations for the increase in Medicaid spending. First, because of low market share, managed care plans paid significantly more to hospitals, physicians and other health care providers, according to anecdotal evidence. Second, because commercial plans only contract with Medicaid if they anticipate a profit, such contracts may add to Medicaid expenditures. Finally, administrative costs per enrollee may be significantly higher in small managed care plans than in the much larger fee-for-service plan with its likely economies of scale.
- Counties with only one managed care plan experienced significantly greater spending increases than those with multiple plans, suggesting a benefit to competition. However, significant spending increases occurred under all three types of managed care arrangements.
- Significant improvements in health outcomes did not result. While hospitalizations for avoidable conditions decreased, the declines were no more than for other conditions (such as accidents) not affected by medical care quality, leading the principal investigator to conclude "that Medicaid managed care plans did not improve health; they were simply less likely to hospitalize individuals with a certain condition."
- Infant health outcomes showed little change.
- Although the study did not produce administrative cost data, anecdotal evidence suggests that the shift to managed care may result in increased state administrative costs. Study findings, therefore, may understate the increase in Medicaid spending.
- The principal investigator noted the difficulty in measuring health given the many dimensions, observed and unobserved, of health outcomes and cautioned that the California results may not be generalizable to other states' Medicaid programs or to the federal Medicare program.
GRANT DETAILS & CONTACT INFORMATION
Effects of Managed Care on Government Spending and Health Care Quality: Evidence from California's Medicaid Mandate
University of Chicago Department of Economics (Chicago, IL)
Dates: October 2002 to September 2003
University of Maryland College of Behavioral and Social Sciences (College Park, MD)
Dates: September 2003 to March 2004
Mark Duggan, Ph.D.
Report prepared by: Karin Gillespie
Reviewed by: Marian Bass
Reviewed by: Molly McKaughan
Program Officer: Nancy L. Barrand