Medicare+Choice: Current role and near-term prospects

Health Affairs

Volumn 22, Issue SUPPL., 2003, Pages

  Thorpe, Kenneth E ^a,b   Atherly, Adam ^c  

^a EMORY UNIVERSITY   (United States)

^b EMORY UNIVERSITY   (United States)

^c Department of Health Policy and Management ^*

Author keywords

[No Author keywords available]

Indexed keywords

EID: 27544447574     PISSN: 02782715     EISSN: 02782715     Source Type: Journal    
DOI: None     Document Type: Review

References (24)

1. G. Riley et al., "Health Status of Medicare Enrollees in HMOs and Fee-for-Service in 1994," Health Care Financing Review (Summer 1996): 65-76.
2. M. Gold, "Medicare+Choice: An Interim Report Card," Health Affairs (July/Aug 2001): 120-138. Our analysis projects that on average, M+C plan reimbursement will be 108.9 percent of FFS spending by 2003.
3. M. Gold and J. McCoy, "Choice Continues to Erode," Mathematica Policy Research Fast Facts no. 7, January 2002, www.mathematica-mpr.com/PDFs/ fastfacts7.pdf (24 May 2002).
4. In 2002 the supplemental premium for a "typical" M+C plan is approximately $30 per month in our data, compared with $130-$150 per month for the most popular Medigap plan (Plan F), which does not provide any outpatient prescription drug coverage. These quotes come from the Quotesmith Web site, www.quotesmith.com (5 May 2002).
5. Letter from Dan Crippen, director, Congressional Budget Office, to members of Congress, 11 June 2001, www.cbo.gov/showdoc.cfm?index= 2873&sequence=0&from=1 (10 June 2002).
6. K.E. Thorpe and A. Atherly, "Reforming Medicare: Impacts on Federal Spending and Choice of Health Plans," 10 October 2001, www.healthaffairs. org/Web_Exclusives/Thorpe_WE_101001.htm (10 June 2002).
7. Centers for Medicare and Medicaid Services, "Medicare Managed Care Contract (MMCC) Plans - Monthly Summary Report," www.hcfa.gov/stats/mmcc. htm (10 March 2002).
8. CMS, "Medigap Compare," www.medicare.gov/MGCompare/ MPPFRedirect.asp (7 August 2001).
9. This model runs each M+C plan's design through an expenditure distribution of M+C enrollees. It then calculates the actuarial value of each plan (the ratio of payments paid by the plan to total spending - plan payments plus out-of-pocket expenses). The model also calculates the actuarial value of the core set of Medicare services. We subtract each plan's actuarial value from the actuarial value of a plan that provides just the core Medicare services. This difference is multiplied by a mean monthly total "possible" (that is, total beneficiary spending for all services covered - core Medicare services plus prescription drugs, hearing aids, eyeglasses, dental care, and so forth) spending for M+C enrollees in each county. This provides a dollar value of the additional benefits.
10. An example of this calculation can be found in Medicare Payment Advisory Commission, Report to the Congress: Medicare Payment Policy (Washington: MedPAC, March 1999), 139, Table A-3.
11. The dollar value of the supplemental benefits over time is calculated as the difference between the M+C payment rate in 2003-2005 (which varies by each of our scenarios) and the growth in the cost of providing the core Medicare benefits by the managed care plan. We assume that plans with negative supplemental benefits will exit the market.
12. Our method is identical to that used in Thorpe and Atherly, "Reforming Medicare." Briefly, the model takes the change in benefits (as described in Note 11, with a dollar of benefits valued at 80 cents) less the expected increase in FFS premiums. We then use a switching elasticity of -0.2. This is based on our own unpublished work but is similar to results in T. Buchmueller, "The Health Plan Choices of Retirees under Managed Competition," Health Services Research (December 2000, Part 1): 949-976.
13. We originally estimated a mixed logit, but the model violated the Independence of Irrelevant Alternatives (IIA) assumption, so we instead used a nested logit. Complete regression details are available on request from the author; contact Ken Thorpe at kthorpe@sph.emory.edu.
14. In the predictions we controlled for age, sex, ethnicity, health status, region, and veteran status plus the county M+C payment rate, per capita income, total population, number of physicians per capita, and the estimated 2002 Medigap premium.
15. For additional information concerning the characteristics of Medicare beneficiaries selecting supplemental coverage in selected local markets, see K.E. Thorpe, A. Atherly, and K. Howell, "Medicare+Choice: Who Enrolls?" 25 April 2002, www.bcbshealthissues.com/relatives/19526.pdf (10 June 2002).
16. This definition looks at beneficiaries who are not provided supplemental coverage by a third party (the government or a former employer). That is, this analysis examines those that do not have access to an employment-based plan or Medicaid. On the other hand, our results likely include some that are enrolled in the M+C program through a retiree health plan. The extent of such enrollment is not known, however.
17. The market-level averages here rely on our estimate of the dollar value of supplemental benefits provided by each M+C plan in a county weighted by their (county) share of M+C enrollees during 2001.
18. MedPAC, Report to the Congress: Medicare Payment Policy (Washington: MedPAC, March 2002), chap. 4.
19. The MedPAC proposal recommends a financially neutral payment system for M+C plans that sets rates equal to 100 percent of local FFS spending in each county. As the ratios of FFS spending and M+C payments vary widely across counties, MedPAC recommends a four-year transition toward these rates. For a brief description of the Bush proposal, see "Strengthening Medicare's Coverage Options: Affordable Health Care to Improve Lives," 17 May 2002, www.whitehouse.gov/news/releases/2002/05/20020517-2.html (10 June 2002).
20. The CMS recently revised its estimated growth in U.S. per capita costs (USPCC), with the most recent estimate suggesting a negative update for floor and blend counties. CMS, Memo to Medicare+Choice Organizations and Other Interested Parties, 1 March 2002, www.hcfa.gov/stats/hmorates/cover03/cover2003. pdf (10 April 2002). The CMS projects that the USPCC will rise by 0.9 percent in 2003, 3.77 percent in 2004, and 4.49 percent in 2005. Payments in the floor and blend counties in each year will be a 2 percent update or the growth in the USPCC, whichever is greater. Using these data, the CMS projects that floor payments (in the larger urban markets) will total $547.54 - lower than the floor amounts in 2002 ($553). Thus, the 2 percent update for 2003 and 2004 will be higher than the update using the USPCC until 2005. As a result, the 2 percent minimum update will result in a larger update than the revised projection from the USPCC through 2004 for virtually all floor and blend counties.
21. Our analysis reaches similar conclusions. As of 2003, in the floor counties, we find that M+C payments (for aged enrollees) average 116 percent of FFS spending per aged Medicare enrollee. Data on FFS spending by county for the aged are from CMS, "Medicare Preferred Provider Organization Demonstration," www.hcfa.gov/research/ppodemo.htm (30 May 2002). These averages are weighted by the number of M+C enrollees in each plan. In the blend and minimum-update counties, this ratio is approximately 106 percent of FFS. Nationally, M+C payments average 108.9 percent of FFS spending per M+C beneficiary.
22. The Johnson proposal also eliminates the "budget-neutrality" requirement for blend counties as well. This requires that M+C spending, including the blended payments, equal the plan payments that would occur if rates in each county were trended by the USPCC.
23. A caveat is in order here. Our own work indicates that the MCBS and other surveys (such as the CPS) generate a different income distribution of the elderly. Counts from the MCBS appear to generate a higher count of the elderly in poverty than generated by the CPS. This could result in higher estimates of the elderly eligible for Medicaid.
24. The Medicare, Medicaid, and SCHIP Benefits Improvement Act (BIPA) of 2000 increased the minimum growth in M+C payments to plans from 2 percent to 3 percent in 2001. Payments to plans in floor counties increased from $415 per month to $475 or $525, depending on the local market population. Analysis by the U.S. General Accounting Office found that approximately 25 percent of plans improved their benefits, while another 12 percent placed some of the additional funding in a benefit stabilization fund. Most plans used the additional dollars for stabilizing access to care within their networks. This latter finding reflects the fact that in markets with the greatest M+C enrollment (the minimum-update markets), costs to the plans of providing core Medicare benefits increased 7 to 10 percent, while plan payments increased 3 percent. Thus, virtually all of the additional revenue was used to finance the higher hospital and physician costs facing M+C plans. See General Accounting Office, M+C: Recent Payment Increases Had Little Effect on Plan Availability in 2001, Pub. no. GAO-02-202 (Washington: GAO, November 2001).