Filling the need for trusted information on national health issues

To Switch or Be Switched: Examining Changes in Drug Plan Enrollment among Medicare Part D Low-Income Subsidy Enrollees

INTRODUCTION
  1. A similar analysis of switching for non-LIS Part D enrollees is found in our earlier report. Jack Hoadley et al., “To Switch or Not to Switch: Are Medicare Beneficiaries Switching Drug Plans to Save Money?” October 2013. https://www.kff.org/report-section/to-switch-or-not-to-switch-issue-brief/.

    ← Return to text

  2. Most beneficiaries are deemed automatically eligible for the LIS based on being enrolled in both Medicare and Medicaid (or receiving benefits from Supplemental Security Income as well as Medicare). See Appendix 1: Study Methodology for more information.

    ← Return to text

  3. In 2010 (the end of the period covered in this study), a somewhat larger share of Part D enrollees qualified as LIS beneficiaries: 9.9 million beneficiaries or 36 percent of all Part D enrollees. The share of LIS beneficiaries in PDPs (81 percent) was somewhat higher in 2010 than in 2015. The decline in the share of LIS beneficiaries from 2010 to 2015 is partly a result of employers transferring their retirees into Part D plans, thus increasing the total number of Part D enrollees, as well as growing enrollment in LIS Medicare Advantage enrollment through the Financial and Administrative Alignment Demonstration for Dual Eligible Beneficiaries (308,000 as of April 2015).

    ← Return to text

  4. In the program’s first five years (2006 to 2010), annual enrollment periods ran from November 15 to December 31. Starting in 2011 for enrollments effective in 2012, the dates were moved earlier in the year to ensure that coverage could be effective on January 1 of the next year.

    ← Return to text

  5. Examples of such circumstances are if the enrollee moves outside the plan’s service area or if the plan changes its contract with Medicare in certain ways.

    ← Return to text

  6. Each year, the amount of the benchmark is calculated separately in each of the 34 PDP regions based on the enrollment-weighted average premiums submitted by stand-alone PDPs and MA-PDs, or is set at the premium of the lowest-cost stand-alone PDP if no plan has a premium below the average of PDP and MA-PD premiums.

    ← Return to text

  7. Although LIS beneficiaries must be enrolled in a benchmark plan to avoid paying a premium, they have lower cost sharing and no coverage gap even if enrolled in a non-benchmark plan.

    ← Return to text

  8. Additional LIS beneficiaries were in plans losing their benchmark status in 2008, but were transferred to other plans offered by the same plan sponsor – resulting in less disruption than if reassigned to a plan offered by a different sponsor.

    ← Return to text

  9. These so-called chooser notices were not sent for the annual enrollment period at the end of the program’s first year; the practice was initiated in the fall of 2007 for the 2008 plan year. In this first round of notices, only those who had no premium liability when they joined their plan, but were subject to a premium in 2008, received this notice. In later years, all LIS beneficiaries who were subject to a premium in the next year have received notices.

    ← Return to text

  10. Beneficiaries enrolled in Medicare Advantage drug plans are excluded because the enrollment dynamics and factors affecting plan switches are different for these two segments of the program and because CMS does not reassign LIS beneficiaries to Medicare Advantage plans.

    ← Return to text

  11. There are different types of LIS assistance available in Part D. Beneficiaries who qualify because they are full-benefit dual eligibles or are enrolled in the Medicare Savings Program and those who qualify with incomes below 135 percent of the federal poverty level (FPL) and assets below specified levels pay no premiums if enrolled in a benchmark plan. Other beneficiaries with incomes between 135 percent and 150 percent of FPL qualify for partial subsidies and are generally excluded from the analysis. See Appendix 1: Study Methodology for more information on these other groups.

    ← Return to text

  12. These results are generally taken from our previous report. Jack Hoadley et al., “To Switch or Not to Switch: Are Medicare Beneficiaries Switching Drug Plans to Save Money?” October 2013. https://www.kff.org/report-section/to-switch-or-not-to-switch-issue-brief/. In a few cases, we recalculated results from that report to match the particular definitions used in this report (noted in footnotes).

    ← Return to text

Key Findings
  1. The share of those eligible for reassignment was calculated from the subset of full-subsidy LIS beneficiaries who were enrolled continuously in PDPs for any particular two-year period.

    ← Return to text

  2. See Appendix 1: Study Methodology for an explanation of the coding inconsistencies in the data with regard to beneficiaries’ enrollment status.

    ← Return to text

  3. For plan changes effective in January 2007, the CMS preliminary dataset made available for this project did not include information on whether the changes resulted from beneficiary elections or CMS reassignments (see Appendix 1: Study Methodology).

    ← Return to text

  4. The rate of reassignments was also lower in 2007 (not shown in Figure 4). Most reassignments to new plans made by CMS effective in January 2007 were incorrectly coded as beneficiary elections in the preliminary file available for this analysis. Thus we cannot classify 2007 plan changes as reassigned or voluntary. In addition, some beneficiaries are incorrectly classified as not eligible for reassignment in December 2007 (and to a lesser degree in later years) because they look as if they voluntarily chose their 2007 plan. As a result, the rate of changing plans for 2008 is artificially high for the “not eligible for reassignment” group. In other words, nearly all the 13.2 percent shown as reassigned (as well as some who were not reassigned) likely belong in the “eligible for reassignment” group. As a result, the rate of changing plans may also be affected for the “eligible for reassignment” group. See Appendix 1: Study Methodology.

    ← Return to text

  5. A small number of those not eligible for reassignment to a new plan (0.2 percent in 2010 and 1.9 percent in 2009) were in fact assigned to a new plan by CMS, in most cases because their plan had exited from the market and no comparable plan was available from the same plan sponsor.

    ← Return to text

  6. Figures are not shown for 2007 or 2008 because of the incorrect coding in January 2007 in the preliminary file available for this analysis. In addition, some switches recorded as voluntary (especially in January 2007) may have been switches between different plans offered by the same plan sponsor; these switches may have been facilitated by the plan sponsor and accomplished with minimal input from plan enrollees. See Appendix 1: Study Methodology.

    ← Return to text

  7. These findings are consistent with those from an evaluation of the transition process between 2007 and 2008, which indicated that among the 1.9 million LIS beneficiaries who received letters from CMS indicating that they would be reassigned if they did not choose a new plan, only 10 percent acted on the letter and made a choice. The other 90 percent were randomly reassigned to new plans for 2008. Grecia Marrufo et al., “Evaluation of the Medicare Demonstration to Transition Enrollment of Low Income Subsidy Beneficiaries,” Acumen, LLC, June 2009, http://www.cms.gov/reports/downloads/Marrufo_PartD_LIS_2009.pdf.

    ← Return to text

  8. We count changes in plan enrollment starting in August 2006 in order to avoid the effects of the first annual enrollment period, which was open until May 15, 2006, with late enrollments effective in June 2006. We excluded July 2006 to allow the possibility that some enrollments in the annual enrollment had a delayed effective date in July.

    ← Return to text

  9. This comparison is restricted to the three-year period from 2008 to 2010 to avoid complications resulting from the incomplete coding for January 2007 in the preliminary file available for this analysis.

    ← Return to text

  10. In this analysis, we excluded switches outside the annual enrollment that were accomplished as reassignments by CMS. These mostly occur in situations where a beneficiary moves to a different PDP region, thus triggering a special enrollment period.

    ← Return to text

  11. In 2010, 1.4 percent of those who were assigned to their current plan elected a new plan between February and December compared to 5.1 percent of those who chose their prior plan. Rates were similar in 2008 (2.7 versus 5.5) and 2009 (1.5 versus 4.7).

    ← Return to text

  12. This total also includes some plan changes resulting from specific circumstances such as moving to another state. Other circumstances include one instance in 2010 were a PDP contract was terminated and its enrollees were assigned to other PDPs or chose one on their own.

    ← Return to text

  13. The period of investigation for switching outside of the annual enrollment period starts in enrollment with enrollment decisions effective August 2006 in order to exclude all switches during the longer annual enrollment period in the program’s first year. The 14.4 percent changing plans at least one time includes 3.7 percent who did so two or more times.

    ← Return to text

  14. During the last half of 2010, the share of LIS beneficiaries in PDPs who made a voluntary switch of plans in any month was generally about 0.2 percent. In August 2010, plan switches were effective for 0.6 percent of LIS beneficiaries in PDPs (and about 0.3 percent in September). These shares project to about 30,000 LIS beneficiaries responding to the “nudge notice” in August and another 6,000 in September.

    ← Return to text

  15. Jack Hoadley et al., “To Switch or Not to Switch: Are Medicare Beneficiaries Switching Drug Plans to Save Money?” October 2013. https://www.kff.org/report-section/to-switch-or-not-to-switch-issue-brief/.

    ← Return to text

  16. Neither of these analyses of premium costs fully account for the fact that some beneficiaries may face quite different costs for the set of drugs that they take in one plan compared to another, including both whether their drugs are on formulary and any variations in cost sharing for non-LIS beneficiaries. In some cases these cost-sharing differences may be enough to offset premium savings.

    ← Return to text

  17. In this analysis, beneficiaries are classified whether they were eligible for reassignment as of the start of the measurement period (August 2006). Because some switch classifications over time, mostly from eligible to not eligible, this approach may weaken the differences. But it also has the advantage of mostly avoiding the impact of the misclassification issue in January 2007.

    ← Return to text

  18. Adam Atherly, Curtis S. Florence, and Kenneth E. Thorpe, ‘‘Health Plan Switching among Members of the Federal Employees Health Benefits Program,’’ Inquiry 42(3): 255–65, 2005. Peter J. Cunningham, “Few Americans Switch Employer Health Plans for Better Quality, Lower Costs,” NIHCR Research Brief No. 12, Center for Studying Health System Change, January 2013.

    ← Return to text

  19. Health Insurance Marketplaces 2015 Open Enrollment Period: March Enrollment Report, March 10, 2015. http://aspe.hhs.gov/health/reports/2015/MarketPlaceEnrollment/Mar2015/ib_2015mar_enrollment.pdf.

    ← Return to text

  20. Elizabeth Hargrave, Bhumika Piya, Jack Hoadley, Laura Summer, and Jennifer Thompson, “Experiences Obtaining Drugs under Part D: Focus Groups with Beneficiaries, Physicians, and Pharmacists,” Contractor report submitted to the Medicare Payment Advisory Commission, March 2008. http://www.medpac.gov/documents/contractor-reports/May08_PartDFocusGroup_CONTRACTOR_JS.pdf. Gretchen Jacobson, Christina Swoope, Michael Perry, and Mary C. Slosar, How are Seniors Choosing and Changing Health Insurance Plans? https://www.kff.org/medicare/report/how-are-seniors-choosing-and-changing-health-insurance-plans/.

    ← Return to text

  21. Jack Hoadley et al., “ Medicare Part D in Its Ninth Year: The 2014 Marketplace and Key Trends, 2006-2014,”August 2014, https://www.kff.org/medicare/report/medicare-part-d-in-its-ninth-year-the-2014-marketplace-and-key-trends-2006-2014/.

    ← Return to text

  22. Most plans include at least 90 percent of all pharmacies in their networks, so pharmacy availability is usually not an issue. Even though there has been a rise in plans’ of pharmacy tiers with different cost-sharing levels, LIS beneficiaries pay the same cost sharing regardless of pharmacies’ network status.

    ← Return to text

  23. Laura Summer, Jack Hoadley, and Elizabeth Hargrave, “The Medicare Part D Low-Income Subsidy Program: Experience to Date and Policy Issues for Consideration,” Kaiser Family Foundation, September 2010,” available at https://www.kff.org/medicare/issue-brief/the-medicare-part-d-low-income-subsidy/. Jack Hoadley et al., “The Role of Beneficiary-Centered Assignment for Medicare Part D,” contractor report for MedPAC, June 2007, http://www.medpac.gov/documents/contractor-reports/June07_Bene_centered_assignment_contractor.pdf  Jack Hoadley, Elizabeth Hargrave, Katie Merrell and Laura Summer, “Beneficiary-Centered Assignment and Medicare Part D,” presentation to MedPAC, September 4, 2008, https://georgetown.app.box.com/s/rg9g4wzpqs3drcs5fc1taxtahxopno1d. Yuting Zhang, Chao Zhou, and Seo Hyon Baik, “A Simple Change To The Medicare Part D Low-Income Subsidy Program Could Save $5 Billion,” Health Affairs 33(6):940-945, June 2014.

    ← Return to text

  24. Laura Summer, Jack Hoadley, and Elizabeth Hargrave, “The Medicare Part D Low-Income Subsidy Program: Experience to Date and Policy Issues for Consideration,” Kaiser Family Foundation, September 2010,” available at https://www.kff.org/medicare/issue-brief/the-medicare-part-d-low-income-subsidy/.

    ← Return to text

  25. The 29 percent is a share of all enrollees who renewed coverage, regardless of whether they received a subsidy, but overall 87 percent of all enrollees in the federal marketplace were eligible for subsidies.   “Health Insurance Marketplaces 2015 Open Enrollment Period: March Enrollment Report,” Department of Health and Human Services, ASPE Issue Brief, March 10, 2015, http://aspe.hhs.gov/health/reports/2015/MarketPlaceEnrollment/Mar2015/ib_2015mar_enrollment.pdf.

    ← Return to text

  26. http://news.coveredca.com/2015/01/covered-california-has-taken-steps-to.html.

    ← Return to text

  27. http://www.healthsourceri.com/press-releases/healthsource-ri-releases-enrollment-demographic-and-volume-data-through-february-23-2015/.

    ← Return to text

  28. Marketplace consumers with incomes in the range of Part D LIS beneficiaries do not qualify for marketplace subsidies, but should qualify for Medicaid in the states that have enacted Medicaid expansions.

    ← Return to text

  29. Sabrina Corlette, Jack Hoadley, and Sandy Ahn, “Marketplace Renewals: State Efforts to Maximize Enrollment into Affordable Health Plan Options,” Robert Wood Johnson Foundation, December 2014, http://www.rwjf.org/en/library/research/2014/12/marketplace-renewals--state-efforts-to-maximize-enrollment-into-.html.

    ← Return to text

Appendix
  1. We exclude any enrollees in plans that are open exclusively to beneficiaries who are former employees of a particular employer or union. In 2010, only 0.4 percent of LIS beneficiaries were in employer plans.

    ← Return to text

  2. The continuous enrollment criterion for the 2006-07 file allows inclusion of beneficiaries who entered the program during the full annual enrollment period for 2006, which extended from November 15, 2005 until May 15, 2006.

    ← Return to text

  3. Jack Hoadley et al., “To Switch or Not to Switch: Are Medicare Beneficiaries Switching Drug Plans to Save Money?” October 2013. https://www.kff.org/report-section/to-switch-or-not-to-switch-issue-brief/.

    ← Return to text

  4. The supplemental file included two other variables: the date of this most recent enrollment type and the history of enrollment types for the month in which the decision was made. Thus, if a beneficiary is randomly reassigned, then picks a plan on her own, and then changes her mind and picks another plan (all to go effect in January), the history variable would record “HBB.” We used these variables for testing purposes.

    ← Return to text

  5. These are mostly situations where a plan sponsor has reorganized plan offerings as a result of acquiring plans from another sponsor, dropping some of its plans, or adding new plans.

    ← Return to text

  6. These situations have occurred relatively rarely in the Part D program to date.

    ← Return to text

  7. The value of the LIS depends on income. A few LIS beneficiaries (2.6 percent) are entitled to a reduced premium instead of premium-free coverage. The so-called partial LIS beneficiaries also receive more limited relief for cost sharing. See Appendix 1: Study Methodology for more numbers.

    ← Return to text

  8. See LIS report for a more detailed discussion of benchmark plans. Laura Summer, Jack Hoadley, and Elizabeth Hargrave, “The Medicare Part D Low-Income Subsidy Program: Experience to Date and Policy Issues for Consideration,” Kaiser Family Foundation, September 2010,” available at https://www.kff.org/medicare/issue-brief/the-medicare-part-d-low-income-subsidy/.

    ← Return to text

  9. In some cases, CMS approves a formal crosswalk of enrollment, by which enrollees in one plan are transferred to another plan offered by the same sponsors. We do not count these as plan switches. In other cases, it may be that transfers among plans are made by plan sponsors by other procedures, such as a letter to the enrollees.

    ← Return to text