Using authority in the Affordable Care Act, the Centers for Medicare and Medicaid Services (CMS) has launched demonstrations that seek to improve care and control costs for people who are dually eligible for Medicare and Medicaid. Nearly 379,000 beneficiaries in nine states are enrolled in capitated models in these demonstrations as of November, 2015. Implemented beginning in July 2013, the demonstrations are changing the care delivery systems through which beneficiaries receive medical and long-term care services and the financing arrangements among CMS, the states, and providers. The demonstrations initially were approved for three years, but in July, 2015, CMS announced that states may extend their demonstrations for an additional two years. (New York’s demonstration for people with developmental disabilities (DD) is approved for four years.)

As of December, 2015, CMS has finalized memoranda of understanding (MOUs) with 13 states to implement 14 demonstrations:

• Ten states (California, Illinois, Massachusetts, Michigan, New York, Ohio, South Carolina, Rhode Island, Texas, and Virginia) are testing a capitated financial alignment model (New York is approved for two capitated demonstrations that target different populations and involve different health plans);
• Two states (Colorado and Washington) are testing a managed fee-for-service (FFS) financial alignment model; and
• One state (Minnesota) is testing the integration of administrative functions without financial alignment (Figure 1).

The status of past proposals from other states is detailed in the Appendix. This issue brief compares key provisions of the approved demonstrations, summarized in Table 1 on the next page.

Background

Dual eligible beneficiaries include seniors and non-elderly people with significant disabilities, some of whom are among the poorest and sickest beneficiaries covered by either Medicare or Medicaid. The predominant existing service delivery models for these beneficiaries typically involve little to no coordination between the two programs. Dual eligible beneficiaries account for a disproportionate share of spending in the Medicare and Medicaid programs. In the case of Medicare, this is mainly due to their relatively poorer health status, which requires higher use of medical services compared to other program beneficiaries. In the case of Medicaid, dual eligible beneficiaries’ relatively high spending is generally attributable to their greater need for LTSS.

Key Demonstration Provisions

Geographic Area and Target Population

Four states (Colorado, Minnesota, Rhode Island, and South Carolina) have statewide demonstrations, while the others are limited to certain regions.

The states’ target populations for their demonstrations vary, with ten states (California, Colorado, Illinois, Michigan, New York, Ohio, Rhode Island, Texas, Virginia, and Washington) including both elderly and non-elderly beneficiaries. Among the states targeting sub-populations:

• Massachusetts targets non-elderly people with disabilities;
• Minnesota’s administrative alignment demonstration targets elderly beneficiaries;
• South Carolina targets elderly beneficiaries who live in community-based settings at enrollment;
• One of New York’s demonstrations focuses on elderly and non-elderly beneficiaries who receive nursing facility services or nursing facility diversion and transition home and community-based waiver services or who require more than 120 days of community-based LTSS, while New York’s other demonstration focuses on beneficiaries with DD; and
• Texas targets elderly and non-elderly beneficiaries with disabilities who qualify for Supplemental Security Income (SSI) benefits or certain Medicaid home and community-based waiver services for seniors and adults with physical disabilities.

Michigan and New York are the only states with capitated demonstrations to include both beneficiaries with DD and DD services.

Enrollment

Estimated Number of Eligible Beneficiaries

CMS has stated that it plans to limit enrollment in the demonstrations to no more than two million dual eligible beneficiaries nationally. As of December, 2015, CMS has approved 14 demonstrations in which an estimated over 1.4 million beneficiaries are eligible to enroll. (Not all beneficiaries who are eligible to participate in the demonstrations are expected to enroll.)  The estimated number of beneficiaries eligible for California’s demonstration is just under 30 percent of the total number of beneficiaries eligible for all demonstrations approved to date and exceeds the number of eligible beneficiaries in each of the other states with approved demonstrations. Enrollment in Los Angeles County, capped at 200,000 beneficiaries, will be greater than the number of beneficiaries eligible to participate in any other state’s demonstration (Figure 2).

Enrollment Timeline

Enrollment has begun in all of the states with demonstrations approved to date except Rhode Island, which will begin its demonstration no sooner than December, 2015, and New York’s demonstration for beneficiaries with DD, which will begin no sooner than April, 2016. States vary in the amount of time that beneficiaries have been enrolled in their demonstrations:  some states have enrolled beneficiaries for more than two years (Washington, Massachusetts), while other states began enrolling beneficiaries in early 2015 (New York, South Carolina, Texas, Michigan) (Figure 3).

Enrollment Process and Beneficiary Choices

Nearly all of the capitated demonstrations (except for some counties in California, in which beneficiaries are automatically enrolled in the demonstration without an initial voluntary enrollment period) began with a voluntary enrollment period in which beneficiaries could “opt in” to the demonstration and select a managed care plan (Rhode Island also plans to begin with a voluntary opt in period). The voluntary enrollment period in nearly all states is followed by passive enrollment periods in which the remaining beneficiaries will be automatically assigned to a managed care plan; however, enrollment in New York’s demonstration for beneficiaries with DD will remain voluntary only. To effectuate passive enrollment, states are developing “intelligent assignment” algorithms to preserve continuity of providers and services when assigning beneficiaries to plans.

Beneficiaries retain the right to opt out of the demonstration at any time but must take affirmative action to do so. In all states, beneficiaries can opt out of the demonstration and choose another delivery system (i.e., FFS, Medicare Advantage, Program of All-Inclusive Care for the Elderly) for their Medicare benefits. However, states may seek CMS approval to require beneficiaries to enroll in Medicaid managed care even if they opt out of the financial alignment demonstration for their Medicare benefits, and five states with capitated demonstrations (California, Illinois, New York’s LTSS demonstration, Ohio, and Texas) have indicated that they are doing so. By contrast, six states with capitated demonstrations (Massachusetts, Michigan, New York’s DD demonstration, Rhode Island, South Carolina, and Virginia) allow beneficiaries who opt out of the demonstration to remain in the FFS delivery system for both their Medicare and Medicaid benefits (Table 2).

Given the complexities of the enrollment decision, beneficiaries are likely to need individual in-person options counseling to make their choice. Seven states (California, Illinois, Massachusetts, Michigan, New York, Virginia, and Washington) have received CMS funding to date to support beneficiary outreach, education, and options counseling in their demonstrations through their State Health Insurance Program and Aging and Disability Resource Centers.

In Washington’s managed FFS demonstration, beneficiaries are automatically enrolled in a health home network but retain the choice about whether to receive Medicaid health home services; other Medicare and Medicaid services will continue to be provided on a FFS basis. Similarly, in Colorado’s managed FFS demonstration, beneficiaries will be automatically assigned to the Regional Care Collaborative Organization in their geographic area to access care coordination services but may disenroll from the demonstration at any time.

Minnesota’s administrative alignment demonstration does not involve passive enrollment; instead enrollment in Senior Health Options plans remains voluntary, although the demonstration will test an integrated enrollment system.

Care Delivery Model

The ten states with capitated demonstrations are using managed care plans to coordinate services for beneficiaries through a person-centered planning process. Person-centered planning focuses on the strengths, needs, and preferences of the individual beneficiary instead of being driven by the care delivery system.

Some states require or allow their managed care plans to contract with other entities to provide services in their demonstrations (Table 3). Massachusetts requires its plans to contract with community-based organizations to provide Long-Term Supports coordinators as independent members of the beneficiary’s care team, Michigan requires its plans to contact with existing Medicaid Prepaid Inpatient Health Plans (PIHPs) to provide behavioral health services, and Ohio requires its plans to contract with Area Agencies on Aging to coordinate home and community-based waiver services for enrollees over age 60. (Illinois, New York, Rhode Island, South Carolina, Texas, and Virginia’s MOUs do not include any similar requirements). California requires its plans to establish MOUs with county behavioral health agencies to provide specialty mental health services and with county social services agencies to coordinate In Home Supportive Services. Demonstration health plans in Los Angeles County are subcontracting with other Medicare Advantage plans to offer a variety of benefit packages to enrollees in California’s demonstration.

Washington’s managed FFS demonstration is using Medicaid health home care coordination organizations to manage services among existing Medicare and Medicaid providers, and Colorado is using its existing Medicaid Accountable Care Collaborative program to coordinate Medicare and Medicaid services for beneficiaries in its demonstration.

Minnesota’s administrative alignment demonstration maintains the existing Senior Health Options program delivery system in which Medicaid managed care plans contract with the state and also are qualified as Medicare Advantage Special Needs Plans focused on dual eligible beneficiaries (D-SNPs) under contract with CMS.

Financing

Ten states (California, Illinois, Massachusetts, Michigan, New York, Ohio, Rhode Island, South Carolina, Texas, and Virginia) are testing CMS’s capitated financial alignment model, in which managed care plans will receive capitated payments from CMS for Medicare services and the state for Medicaid services.

Anticipated program savings in the capitated financial alignment demonstrations are deducted up-front from CMS’s and the state’s respective baseline contributions to health plans by a savings percentage for each year (Table 1). CMS will contribute the Medicare portion of the capitated rate and recently announced rate adjustments for 2016, to better align capitated payments with fee-for-service costs for full benefit dual eligible beneficiaries. The Medicaid portion of the capitated rate includes both the federal and state funding. None of the MOUs explicitly states the basis for the savings percentages, although Illinois’ MOU notes that it currently has one of the highest rates of potentially avoidable hospital admissions among dual eligible beneficiaries nationally and one of the highest proportions of spending on institutional services compared to HCBS. While California’s MOU specifies minimum savings percentages of 1% in year one, 2% in year two, and 4% in year three, it also includes maximum savings percentages of 1.5% in year one, 3.5% in year two, and 5.5% in year three, making the maximum savings percentages in California among the highest of the approved demonstrations to date. Texas’s MOU specifies savings percentages of 1.25% in year 1.a (March to December 2015), 2.75% in year 1.b (2016), 3.75% in year 2, and 5.5% in year 3.

All 10 states with capitated financial alignment demonstrations include provisions to withhold a portion of the capitated rate that plans can earn back if specified quality measures are met. California also requires its plans to provide incentive payments from the quality withhold funds to county behavioral health agencies based on achievement of service coordination measures, and Michigan requires its plans to reward the PIHPs that will provide behavioral health services when the plan earns the withheld payment. South Carolina plans must provide financial incentives to providers that achieve NCQA patient-centered medical home certification.

Two states (Colorado and Washington) are testing CMS’s managed FFS model in which providers will continue to receive FFS reimbursement for both Medicare and Medicaid-covered services. Any savings in these demonstrations will be determined retrospectively, with the state eligible to share in savings with CMS if savings targets and quality standards are met.

Minnesota’s administrative alignment demonstration will not test one of CMS’s financial alignment models. Instead, Minnesota’s Senior Health Options program will maintain its existing capitated integrated payment and delivery system arrangements involving Medicaid MCOs that also qualify as Medicare Advantage D-SNPs. Plans will be allowed to integrate Medicare and Medicaid primary care payments to promote care coordination through health care homes and improved coordination among primary, acute, and LTSS and among physical and behavioral health services.

Benefits

The 10 states with capitated financial alignment demonstrations include nearly all Medicare and Medicaid services in the plans’ benefits package and capitated payment. All states include nursing facility services in the plans’ capitated payment and benefits package. Eight of the 10 states testing the capitated model (Illinois, Michigan, New York (both demonstrations), Ohio, South Carolina, Rhode Island, Texas, and Virginia) include beneficiaries who receive certain Medicaid home and community-based waiver services, while two states (California and Massachusetts) exclude all HCBS waiver enrollees from their demonstrations. Plans are allowed to offer additional benefits, outside the traditional Medicare and Medicaid benefits packages, as appropriate to beneficiary needs. All states require their health plans to offer beneficiaries the option to self-direct their LTSS (Table 3).

Five of the capitated states require plans to offer additional benefits as part of the demonstration. Massachusetts plans must offer certain diversionary behavioral health and community support services that are not otherwise covered as well as expanded Medicaid state plan benefits. Ohio’s  § 1915(b)/(c) Medicaid waiver, which operates concurrently with the demonstration, includes expanded Medicaid state plan benefits and additional HCBS. California’s demonstration includes vision and non-emergency medical transportation benefits, and its plans may offer additional HCBS. South Carolina’s demonstration includes a palliative care benefit for enrollees with a serious, chronic or life-threatening illness who may not meet hospice criteria. Michigan’s health plans must offer adaptive medical equipment and supplies, community transition services, fiscal intermediary services to support self-direction, personal emergency response systems, and respite services (Table 3).

Washington’s managed FFS demonstration adds Medicaid health home services but does not otherwise change the existing Medicare and Medicaid benefits packages. Similarly, Colorado offers care coordination services but otherwise does not change the existing Medicare and Medicaid benefits packages.

Minnesota’s administrative alignment demonstration will continue to provide Medicare benefits at least equivalent to the basic benefit levels included in Medicare Parts A, B, and D and Medicaid benefits based on existing Medicaid MCO contracts.

Demonstration Ombudsman

CMS has announced a funding opportunity for states with approved MOUs to support the planning, development, and provision of independent ombudsman services in the demonstrations, with ten states (California, Colorado, Illinois, Massachusetts, Michigan, Ohio, Rhode Island, South Carolina, Virginia, and Washington) awarded funding to date. Seven states (Colorado, Illinois, Michigan, Ohio, South Carolina, Texas, and Washington) are using an ombuds office within state government, although independent of the state Medicaid agency, while four states (California, Massachusetts, New York, and Virginia) are using an ombuds entity outside of state government. Minnesota’s MOU provides that the state’s managed care ombudsman will provide input on plan and system-wide performance but does not provide further details.

Appeals

All of the capitated demonstrations provide beneficiaries with a single integrated notice of appeal rights, and the existing Medicare Part D appeals process for prescription drugs will continue to apply in all demonstrations. Minnesota’s administrative alignment demonstration is building on the integrated appeals system already established in its Senior Health Options program by adding a single integrated notice of appeal rights and standardizing the timeframes to request Medicare and Medicaid appeals. By contrast, Colorado and Washington’s managed FFS demonstrations do not make any changes to the existing Medicare and Medicaid appeals systems.

New York’s two capitated financial alignment demonstrations  include a fully integrated four level appeals process for all services traditionally covered by Medicare Parts A and B and Medicaid. New York requires its demonstration health plans to continue providing benefits while appeals are pending for both prior-approved Medicare and Medicaid services if the beneficiary so requests within 10 days of the date of the notice. (Continued benefits pending appeal is currently available under federal law for Medicaid services but not for Medicare services.)

Five of the capitated demonstration states (Illinois, Massachusetts, Rhode Island, South Carolina, and Virginia) require beneficiaries to first exhaust an internal health plan appeal before proceeding to external appeals, while four of the capitated demonstration states (California, Michigan, Ohio, and Texas) allow beneficiaries to choose whether to first file an internal health plan appeal or instead to proceed directly to a fair hearing for Medicaid-covered services.

All of the capitated demonstrations require health plans to continue Medicare and Medicaid benefits while internal health plan appeals are pending; beneficiaries may request that Medicaid benefits continue while fair hearings are pending, but Medicare benefits will not continue during external appeals. (California’s aid pending appeal provision is contained in its three-way contract instead of its MOU.)  California’s MOU provides that the existing Medicare and Medicaid appeals processes will continue at least through demonstration year one, and the state will work to create a more integrated appeals process in future years.

Looking Ahead

As the demonstrations are implemented, additional details about several features are emerging, including how beneficiaries are being notified, counseled, and enrolled; how the demonstrations are being monitored and overseen; how beneficiary ombuds programs are being implemented; and how the demonstrations are being evaluated. CMS has contracted with RTI International to conduct an overall evaluation of the demonstrations as well as state-specific evaluations. The MOUs provide that the evaluations will include site visits, analysis of program data, focus groups, key informant interviews, analysis of changes in quality, utilization, and cost measures, and calculation of savings attributable to the demonstrations. The evaluation findings are to be reported quarterly, although there is likely to be a lag in data availability.

The approved MOUs provide additional information about how CMS and the states envision the demonstrations working and insight into the framework and policy decisions that CMS may apply when developing MOUs with other states that submitted proposals. Additional details are specified in the three-way contracts between CMS, the state, and demonstration plans in the capitated model, in the states’ final demonstration agreement with CMS in the managed FFS model, and in policy guidance. Key areas to continue to consider as the demonstrations are implemented include:

• how beneficiaries are making their enrollment choices;
• what the actual sources of program savings will be;
• how beneficiaries’ access to medically necessary services and supports is being ensured;
• how the demonstrations are affecting beneficiary access to HCBS;
• how beneficiaries are navigating the demonstrations’ grievance and appeals processes;
• whether continuity of care and intelligent assignment provisions are sufficient to prevent care disruptions and the extent to which beneficiaries’ current providers are participating in demonstration health plan networks;
• how plans and providers are accommodating the needs of beneficiaries with disabilities; and
• what impact the demonstrations are having on care quality and health outcomes.

While the demonstrations offer the potential opportunity to improve care coordination, lower program costs, and achieve outcomes such as better health and the increased use of HCBS instead of institutional care, at the same time the high care needs of many dual eligible beneficiaries increases their vulnerability when care delivery systems are changed.

In 2015, more than 17 million Medicare beneficiaries (31%) are enrolled in Medicare Advantage plans,¹  such as health maintenance organizations (HMOs) or preferred provider organization (PPOs). Medicare Advantage plans are offered as an alternative to the traditional Medicare program. Medicare beneficiaries can enroll in a Medicare Advantage plan, change Medicare Advantage plans, or switch from Medicare Advantage to traditional Medicare during the annual open enrollment period. Changes in the Medicare Advantage marketplace have always been closely watched, and since 2010 when the Affordable Care Act (ACA) was enacted, many have been interested in the effects of the ACA phasing down federal payments to Medicare Advantage plans. More recently, proposed mergers between health insurance firms with large footprints in Medicare Advantage have raised questions about how the mergers could affect beneficiaries.

This spotlight analyzes publicly available data to review the Medicare Advantage plans offered in 2016. It provides updated information describing how Medicare Advantage plan choices are changing in 2016, includes new information on premiums, out-of-pocket limits and other plan features, and examines the role of large firms offering plans in the marketplace. Findings include:

• The average Medicare beneficiary will be able to choose from 19 plans in 2016, a number which has been relatively stable since 2012. Relatively few plans are entering or exiting the Medicare Advantage market, and for the most part, the same plans that were available in 2015 will be available in 2016.
• While the average Medicare beneficiary can choose from many plans, these plans will be offered by a handful of firms. The average beneficiary will be able to choose from plans offered by six firms; one-quarter of beneficiaries nationwide will have a choice of plans offered by three or fewer firms in 2016.
• If enrollees in Medicare Advantage plans with prescription drug coverage (MA-PDs) stay in the same plan between 2015 and 2016, their premiums will increase by 8 percent, on average. Similar to past years, about four-fifths of beneficiaries (81%) will have access to an MA-PD with no premium in 2016.
• Plans’ limits on out-of-pocket costs for Part A and B benefits will rise in 2016, as they have in prior years. Almost two-fifths (39%) of plans will have limits equal to the maximum allowed ($6,700 per year) in 2016, up from 17 percent in 2013.
• Nearly half (45%) of MA-PDs will impose a deductible for Part D prescription drugs in 2016, an increase from 2015, and the average Part D drug deductible will be higher in 2016 than 2015.
• The average quality star rating for plans will be higher in 2016, with substantial growth in the number of contracts with 4 and 4.5 stars.

Number of Plans Offered

In total, 2,001 Medicare Advantage plans will be available nationwide for individual enrollment in 2016, 56 more than in 2015 (Figure 1). In 2016, 203 plans that were available in 2015 will be exiting the Medicare Advantage market, and 259 plans will be offered for the first time. An earlier analysis describes the nature of change at the national and state level as plans enter and exit the market in 2016.²  While virtually every state will see some change in the number of plan offerings, most of the plans exiting the market in 2016 attracted relatively few enrollees in 2015. Nationwide, the number of plans offered has been relatively stable since 2011.

HMOs continue to account for the majority of plans available, and will account for two-thirds of all plans offered in 2016 (Table A1). Since 2013, the number of HMOs has increased each year, with 76 more HMOs in 2016, while the availability of other plan types has either remained similar or decline. In 2016, there will be limited changes in the number of local PPOs, regional PPOs, and private fee-for-service (PFFS) plans.

Locally, the number of plans available to Medicare beneficiaries in 2016 will not be very different from in 2015. The average beneficiary will be able to choose from among 19 plans, on average, in 2016 – as compared with 18 plans in 2015 (Figure 2). Beneficiaries in metropolitan areas will be able to choose from 21 plans, on average, and beneficiaries in non-metropolitan areas will be able to choose from 11 plans, on average, an increase of 1 plan from 2015. The average number of plans available in the lowest cost counties (i.e., lowest quartile of counties) will remain the same (15 plans) while the average number of plans available in other counties will increase by 1 plan (from 22 plans to 23 plans in the highest cost counties; Table A2).

Number of Firms Offering Medicare Advantage Plans

The average Medicare beneficiary will be able to choose from many plans, but in some areas, these plans will be offered by a handful of firms. The average Medicare beneficiary will be able to choose from plans offered by 6 firms, on average, in 2016, with sizeable variation across counties (Figure 3). One-quarter of beneficiaries will be able to choose from plans offered by three or fewer firms, while another quarter of beneficiaries will be able to choose from plans offered by 8 or more firms. The number of firms offering Medicare Advantage plans will be highest in the New York City area, with at least 19 firms offering plans in Bronx County, Kings County, New York County, and Queens County. In contrast, in 445 counties in 28 states accounting for 4 percent of Medicare beneficiaries, only one firm will offer Medicare Advantage plans in 2016.

Access to Medicare Advantage Plans

As in recent years, virtually all Medicare beneficiaries (99%) will continue to have access to a Medicare Advantage plan as an alternative to traditional Medicare (Figure 4). All beneficiaries in metropolitan areas (100%) and the vast majority of beneficiaries in non-metropolitan areas (97%) will continue to have access to at least one Medicare Advantage plan – the same percentages as in 2015. Consistent with prior years, a smaller share of beneficiaries in non-metropolitan counties than in metropolitan counties will have access to HMOs or local PPOs, although these plans will be available to at least 70 percent of beneficiaries in non-metropolitan counties.

Premiums are the most visible and the easiest factor for beneficiaries to compare across plans. While other factors, such as limits on out-of-pocket spending, cost-sharing, prescription drugs, extra benefits, and provider networks, can have an even larger impact on beneficiaries’ out-of-pocket spending on health care and access to providers, information on them is less readily accessible and often more difficult to interpret. For cost-sharing and provider networks in particular, it is nearly impossible from publicly available data to construct an apples-to-apples comparison across plans that appropriately weights all factors. As a result, many people use the plan premiums to choose among plans.

Medicare beneficiaries enrolled in Medicare Advantage plans pay the Part B premium like other beneficiaries (less any rebate provided by the Medicare Advantage plan), and may also pay an additional monthly premium charged by the Medicare Advantage plan for benefits and prescription drug coverage. This analysis of premiums includes only Medicare Advantage plans that offer prescription drug coverage (MA-PDs). The minority of Medicare Advantage plans (13%) that do not cover prescription drugs are not included in the analysis, in order to better compare premiums across plan types and years.

Monthly Premiums (Unweighted by Enrollment)

To examine how premiums for all plans available to beneficiaries will change between 2015 and 2016, this spotlight first examines the change in premiums for all MA-PDs, without weighting them by the number of enrollees in each plan. The analysis includes plans with no premiums as well as plans with monthly premiums. In 2016, the average premium for MA-PDs (unweighted by plan enrollment) will be $53 per month – similar to 2015 (Table A3). Premiums for HMOs will average $39 per month, up $1 from 2015. Similar to prior years, HMOs will continue to have lower premiums than regional PPOs ($75 per month), local PPOs ($79 per month) and PFFS plans ($91 per month). Similar to 2015, regional PPOs will have the largest change in premiums, on average, and average premiums will increase $7, on average, from $68 per month in 2015. In contrast, premiums for local PPOs will decrease by $2, on average, from $81 per month in 2015.

Monthly Premiums (Weighted by Enrollment), Assuming Enrollees Remain in the Same Plan

To examine how premiums will change for Medicare Advantage enrollees, assuming they remain in the same plan between 2015 and 2016, this spotlight examines the premiums among MA-PDs offered in both years and weight the premiums by plan enrollment in 2015. These enrollee “weighted average” premiums are a clearer indication than unweighted premiums of the amount paid by the average Medicare Advantage enrollee because enrollment across plans is uneven. Similar to the prior analysis, this analysis includes plans with no premiums as well as plans with monthly premiums. Enrollees may be able to prevent or reduce increases in premiums by switching to another plan. After the open enrollment period, we will analyze the premiums for plans selected by all Medicare Advantage enrollees in 2016.³  In recent years, average premiums, weighted by actual enrollment, have tended to be lower than estimates of weighted premiums based on prior year enrollment. The difference reflects the net impact of beneficiaries making enrollment changes between one year and the next, and plan choices among new Medicare Advantage enrollees.

Based on plan enrollment in the Fall of 2015, the average Medicare Advantage enrollee in a plan that will continue to be offered in 2016 will pay about $41 per month, an increase of $3 per month (or 8%) compared to 2015, if they stay in the same plan (Figure 5). In percentage terms, enrollees in PFFS plans will experience a larger increase in premiums (13% on average) than enrollees in other types of Medicare Advantage plans, and average premiums for PFFS plan enrollees will rise from $52 per month to $58 per month, assuming no change in enrollment. The average increase in enrollment weighted premiums is lower than it has been since 2012 when there was a 3 percent increase. Premiums could be rising at a somewhat slower pace for many reasons including the fact that the changes in payments to plans as a result of the ACA are fully implemented in most counties. Monthly premiums, weighted for 2015 enrollment will be lower for enrollees in HMOs ($31 per month) than for enrollees in other types of Medicare Advantage plans, and higher for enrollees in local PPOs ($68 per month), on average.

In general, enrollees in MA-PDs that will continue to be offered in 2016 had somewhat lower premiums in 2015, on average, than enrollees in plans that will be exiting the Medicare Advantage market in 2016 (Table A1). The 2015 premiums for plans remaining in the market in 2016 averaged $38 per month as compared to $41 per month for those exiting the market. However, the difference varies across plan types and these changes are difficult to interpret because plan premiums are a function of many plan-specific factors (e.g., benefits, cost-sharing, provider networks), as well regional-specific factors (e.g., local practice patterns, payments from the Medicare program).

Access to Plans with No Premium

Medicare Advantage plans with no additional premium (other than the Medicare Part B premium) – so called “zero premium plans” – have been a feature of the Medicare Advantage landscape for many years. About four-fifths (81%) of all beneficiaries will have access to a zero-premium MA-PD in 2016, an increase from 78 percent in 2015 (Figure 6). The continued availability of zero-premium plans mainly reflects offerings among HMOs since zero-premium plans are less common among other plan types.

People on Medicare have said that out-of-pocket spending is important to them when selecting their plan.⁴  Yet, it is challenging for beneficiaries to estimate their future out-of-pocket costs for all plans available in their area because it is difficult to predict their health care needs and virtually impossible to compare cost-sharing for specific services. One alternative for beneficiaries is to use the plans’ limits on out-of-pocket spending to gauge the maximum amount they could pay out-of-pocket in a plan for services covered under Medicare Part A and Part B, assuming they use only in-network providers.

The traditional Medicare program does not include a limit on out-of-pocket spending for services covered under Parts A and B, leaving beneficiaries in traditional Medicare with unlimited financial liability – an impetus for many beneficiaries to acquire supplemental coverage. In contrast, Medicare Advantage plans are required to limit enrollees’ out-of-pocket expenses for services covered under Parts A and B to no more than $6,700, with higher limits allowed for services received from out-of-network providers. Even before the 2011 implementation of this requirement, most Medicare Advantage plans (79% in 2010) voluntarily included a limit on enrollees’ out-of-pocket spending.⁵  CMS encourages plans to limit enrollees’ out-of-pocket expenses to no more than $3,400 per year, by allowing these plans to charge higher cost-sharing for some services. These out-of-pocket limits do not include expenses for prescription drugs covered under MA-PDs (which have separate Part D catastrophic threshold), nor do they include expenses for extra benefits or for services not covered by the plan.

Out-of-pocket limits for MA-PDs will be higher in 2016 than in 2015, on average. In 2016, two-fifths of plans (39%) will have limits equal to the maximum ($6,700), an increase from 17 percent of plans in 2013 (Figure 7). One-quarter (24%) of plans will have a limit of $3,400 or less in 2016, a sharp decline since 2013. In 2016, the average plan’s limit will be $5,257, up from $4,352 in 2013. When weighted by 2015 enrollment, out-of-pocket limits will increase by $193 between 2015 and 2016, on average (Table A1).

Average out-of-pocket limits are increasing across all plan types but a smaller share of HMOs than local PPOs or regional PPOs will have the maximum out-of-pocket limit allowed in 2016 (Figure 8). About three-quarters (73%) of regional PPOs, half (49%) of local PPOs, and one-third (35%) of HMOs have out of pocket limits equal to the maximum limit of $6,700. Limits at or below $3,400 now are rare in across all plan types.

In 2016, the vast majority (87%) of Medicare Advantage plans will offer prescription drug coverage. Historically, Medicare Advantage plans without a drug benefit were developed to meet the needs of beneficiaries who may have access to other sources of prescription drug coverage, such as retiree health coverage from former employers or the Veterans Health Administration, as well as those who for any other reason do not want to purchase such coverage.

Coverage of the Part D Deductible

The standard design of the drug benefit includes an initial coverage deductible which is updated each year, and is $360 in 2016. Plans can vary the benefit design as long as they provide cost-sharing that is at least actuarially equivalent to the standard benefit.

In 2016, nearly half (45%) of MA-PDs will impose a deductible for prescription drug coverage. Since 2013, the share of MA-PDs imposing a Part D deductible has increased three-fold from 14 percent to 45 percent while average MA-PD deductibles have quadrupled from $29 in 2013 to $118 in 2015 (Figure 9). Similar to 2015, 16 percent of MA-PDs will charge the maximum deductible in 2016 ($360 in 2016 compared to $320 in 2015).

MA-PDs are better situated than stand-alone prescription drug plans (PDPs) to offset the costs of drugs with other benefits because they combine Part D with other Medicare-covered benefits. The use of Part D deductibles is increasing among MA-PDs, although they still are used more commonly by stand-alone PDPs. In 2016, two-thirds of PDPs will charge a deductible, and a larger share of PDPs than MA-PDs will charge the maximum amount allowed.⁶  Further research is needed to assess how cost sharing between MA-PDs and PDPs varies and whether the two types of plans generally include the same drugs on the same formulary tier, as this has cost implications for beneficiaries.

Coverage within the “Doughnut Hole”

The initial design of the Medicare drug benefit in 2006 included a coverage gap, or “doughnut hole,” which required beneficiaries with relatively high drug costs to pay 100 percent of their expenses in the coverage gap until they qualified for catastrophic protection. The ACA gradually closes the doughnut hole and eliminates the gap by 2020. In 2016, beneficiaries will be responsible for no more than 45 percent of the cost of brand-name drugs and 58 percent of the cost of generic drugs in the gap.

Less than half (44%) of MA-PDs will provide additional coverage in the Part D coverage gap in 2016, beyond that which is required by the ACA, similar to 2015 and down from 50 percent in 2014 (Figure 10). In contrast, less than one-quarter of PDPs offer any additional coverage in the gap.⁷  Little is known about the extent of additional coverage MA-PDs provide in the gap, or whether the additional coverage is for only a handful of drugs or is more extensive.

For many years, the CMS has posted quality ratings of Medicare Advantage plans to provide Medicare beneficiaries with additional information about plans offered in their area. All Medicare Advantage plans are rated on a 1 to 5 star scale, with 1 star representing poor performance, 3 stars representing average performance, and 5 stars representing excellent performance. CMS assigns quality ratings at the contract level, rather than for each individual plan. This means that every Medicare Advantage plan covered under the same contract receives the same quality rating, and most contracts cover multiple plans.

Since 2012, Medicare Advantage plans have received bonus payments based on the quality ratings, and plans with higher ratings also receive higher rebate amounts. In 2016, plans with 4 or more stars will receive bonuses of 5 percent, and new plans and plans with low enrollment will receive bonuses of 3.5 percent; these bonus percentages are doubled in some counties.⁸  Additionally, beneficiaries can enroll in a plan with 5 stars at any time during the year, not just during the annual open enrollment period.

In 2016, the majority (40%) of Medicare Advantage contracts will have ratings of four or more stars – an increase from 33 percent in 2015 (Figure 11). In 2016, as in prior years, few contracts (4%) receive the top star rating (5 stars), a slight increase from 2 percent of contracts in 2015. The vast majority of Medicare beneficiaries (94%) will have access to at least one plan with four or more stars in 2016, including one-quarter (24%) of beneficiaries who will have access to a plan with five stars (data not shown).

Slightly more than one-third (35%) of contracts will have average ratings (3 or 3.5 stars), and 3 percent of contracts will have below average ratings (2 or 2.5 stars) in 2015; 23 percent of contracts will be too new or have too few enrollees to be rated.

While many organizations offer Medicare Advantage plans, a handful of firms and affiliates have historically accounted for the majority of all Medicare Advantage enrollment nationally. This spotlight examines trends among and across these firms and affiliates that account for large shares of Medicare Advantage enrollment nationally: UnitedHealthcare, Humana, Blue Cross and Blue Shield (BCBS) affiliated companies (including Anthem BCBS plans), Kaiser Permanente, Aetna, Cigna, and Wellcare. Together, these seven firms and affiliates accounted for almost three-quarters of all Medicare Advantage enrollment in 2015.⁹    The potential impact of proposed mergers between some of these large firms (Aetna’s proposed acquisition of Humana and Anthem’s proposed acquisition of Cigna) is not yet clear.

Change in Number and Availability of Plans Offered by Firm

The firms and affiliates profiled in Table A4 account for 61 percent of available plans being offered in 2016 (Table A4). The types of plans offered vary across these firms and affiliates. Firm strategies are reflected both in the types of plans they offer and the share of beneficiaries to whom they are available (Figure 12; Table A5). This analysis focuses only on plans offered for broad based individual enrollment, and excludes special needs plans, employer-sponsored group plans for retirees and other types of plans only available to certain beneficiaries.

UnitedHealthcare. UnitedHealthcare has historically had a large Medicare Advantage footprint and it will continue to do so in 2016. Seven in ten Medicare beneficiaries (70%) will have access to a UnitedHealthcare plan in 2016. UnitedHealthcare has historically offered a mix of plan types, but HMOs comprise the vast majority (84%) of its plan offerings, with a smaller share local PPOs (11%), regional PPOs (5%) and PFFS plans (1%). More than half of beneficiaries (55%) will have access to a UnitedHealthcare HMO in 2016, up from 50 percent of beneficiaries in 2015. Average premiums (not weighted by enrollment) in HMOs offered by UnitedHealthcare will decline between 2015 and 2016, while premiums for the firm’s local PPOs, PFFS plans, and regional PPOs will increase (Table A6). This is a slight reversal of the firm’s strategy in 2015 when it introduced premiums across most of its plans that previously had zero-premiums. In early 2015, 20 percent of Medicare Advantage enrollees were in a UnitedHealthcare plan.¹⁰ 

Humana. Humana also has had a large footprint in Medicare Advantage for over a decade, and Medicare Advantage likewise comprises a large share of Humana’s business. Humana will continue to offer a broader mix of plan types than any other large firm or affiliate in 2016, but the total number of plans the firm offers will decline from 395 plans to 362 plans, continuing the firm’s trend of offering fewer plans each year since 2013. Similar to prior years, about half (51%) of Humana’s plans will be HMOs, about three in ten will be local PPOs (29%), and about one in ten will be PFFS plans (11%) or regional PPOs (9%). The vast majority of Medicare beneficiaries (85%) will have access to a plan offered by Humana in 2016, similar to 2015. HMOs, local PPOs, and regional PPOs offered by Humana will be available to more than half of beneficiaries (58%, 53%, and 61%, respectively) and Humana PFFS plans will be available to more than four in ten beneficiaries (43%). Average premiums (not weighted by enrollment) for local PPOs offered by Humana will decline between 2015 and 2016, while average premiums for other plan types offered by Humana will increase, with the largest average increases in premiums among PFFS plans. In early 2015, 19 percent of Medicare Advantage enrollees were in a Humana plan.

Blue Cross and Blue Shield Affiliates (BCBS). BCBS affiliates are separate companies that do not directly compete with each other and are affiliated nationally.¹¹  Some have converted from non-profit to for-profit status in recent years, most notably Anthem BCBS. We analyze the BCBS affiliates together because the affiliation results in some similarities in approach. In 2016, more than six in ten of BCBS affiliates’ Medicare Advantage offerings will be HMOs (61%) while 35 percent will be local PPOs. Nearly three out of four beneficiaries nationwide (72%) will have access to a BCBS affiliated plan, including almost one-quarter (24%) who will have with access to an Anthem BCBS plan and 55 percent who will have access to plans offered by other companies under the BCBS trademark. In early 2015, 3 percent of Medicare Advantage enrollees were in an Anthem plan, including Anthem BCBS plans, and another 13 percent were in plans offered by other BCBS affiliates.

Kaiser Permanente. Similar to prior years, fewer beneficiaries (17%) will have access to a Kaiser Permanente plan than other firms included in this analysis. Kaiser Permanente has historically been geographically concentrated in California, Colorado, Georgia, Hawaii, Oregon, Washington, and the metropolitan area surrounding the District of Columbia, but its plans tend to have many enrollees in the places in which it operates. In contrast to other firms, Kaiser Permanente only offers HMOs (83%) and cost plans, which are somewhat similarly structured. Only 13 percent of beneficiaries will have access to a Kaiser Permanente HMO and only 2 percent of beneficiaries will have access to a Kaiser Permanente cost plan in 2016. In early 2015, 8 percent of Medicare Advantage enrollees were in a plan offered by Kaiser Permanente.

Aetna. Aetna’s footprint in the Medicare Advantage market has expanded since its acquisition of Coventry, and will become even larger if its proposed acquisition of Humana is approved by the Department of Justice. In 2016, Aetna plans will be available to almost half (49%) of all beneficiaries, up slightly from 45 percent in 2015 and 33 percent in 2014 prior to the acquisition of Coventry. Between 2015 and 2016, Aetna will increase the number of local PPOs it offers, rising from 74 plans in 2015 to 82 plans in 2016. Accordingly, a larger share of beneficiaries will have access to an Aetna local PPO (40%) in 2016, up from one-third (33%) of beneficiaries in 2015. Beginning in 2016, Aetna will begin offering regional PPOs in New Jersey and Ohio, and 7 percent of beneficiaries nationwide will have access to these new regional PPOs in 2016. In prior years, the only firms and affiliates that offered regional PPOs were UnitedHealthcare, Humana, and BCBS affiliates. In early 2015, 7 percent of Medicare Advantage enrollees were in a plan offered by Aetna.

Other Firms. About one-third (32%) and one-fifth (20%) of beneficiaries will have access to a plan offered by Wellcare and Cigna, respectively. Both firms heavily emphasize HMOs, with Wellcare exclusively offering HMOs and Cigna offering predominantly HMOs. Wellcare will increase the number of HMOs it offers from 40 plans in 2015 to 49 plans in 2016; Cigna will slightly increase the number of HMOs it offers by 3 plans, increasing from 40 plans in 2015 to 43 plans in 2016.

Beyond these firms, many other firms and organizations sponsor Medicare Advantage plans, usually in limited geographical markets. Taken together, these firms provide plan options to 79 percent of beneficiaries.

Plan choices for beneficiaries will be relatively stable between 2015 and 2016, with very little change in the number or type of plans available to beneficiaries or the firms offering those plans. In 2016, there will be relatively few plan exits that could cause disruption in coverage. Firms continue to offer diverse products and most beneficiaries have a choice of HMOs and PPOs, while fewer beneficiaries have access to PFFS plans. Medicare Advantage plans’ low premiums (relative to Medigap) and their ability to package traditional Part A and B benefits, prescription drugs, and some aspects of supplemental coverage into a single offering have proven popular both with insurers and a sizeable subgroup of Medicare beneficiaries.

Medicare beneficiaries enrolling in a Medicare Advantage plan will find that the plans available to them, on average, have somewhat higher quality ratings in 2016 than in 2015 but provide less financial protection with higher out-of-pocket costs. While premiums are rising at a somewhat slow rate, a higher share of plans will set their out-of-pocket limit for Part A and B benefits at the maximum amount allowed ($6,700), with few plans setting the limit at the lower standard encouraged by CMS ($3,400). Additionally, more plans are requiring enrollees to pay a deductible for Part D drugs, exposing Medicare beneficiaries with relatively high health care costs to greater financial risk. The reality is that both traditional Medicare and Medicare Advantage plans face challenges in providing financial protection to beneficiaries, in that traditional Medicare does not have an out-of-pocket limit for Part A and Part B covered services.

The Department of Justice is considering proposed mergers among some of the leading national firms in the Medicare Advantage market—Aetna’s proposed acquisition of Humana and Anthem’s proposed acquisition of Cigna. While the four firms’ products differ from one another, their combined share of the Medicare Advantage market means that the acquisitions could have important implications for the Medicare Advantage program. Aetna and Humana together account for about one-quarter (26%) of all Medicare Advantage enrollees, while Anthem and Cigna together account for about six percent of all enrollees in 2015.¹² 

The potential effect of the proposed acquisitions on Medicare beneficiaries is not yet clear. For the individual Medicare Advantage market which we focus on here, mergers may affect the number of plan choices available to beneficiaries if the merging firms decide to consolidate plans, but divestiture requirements could offset some of this reduction. Mergers could also reduce the number of different firms offering plans and competing for enrollees, and this effect is likely to vary across the country. While the average beneficiary in 2016 will have plan choices offered by six firms, one-quarter of all beneficiaries in 2016 will have only three firms competing for their business. While mergers will affect the national marketplace, they also have very specific effects on individual markets that require analysis that anticipates the potential short and long term consequences, as well as any related effects on the characteristics of plans that are offered, and implications for beneficiaries’ costs and access.¹³ 

1. See Jacobson G, Damico A, Neuman T, and Gold M. “Medicare Advantage 2015 Spotlight: Enrollment Market Update.” June 2015. https://modern.kff.org/medicare/issue-brief/medicare-advantage-2015-spotlight-enrollment-market-update/ ↩︎
2. Jacobson G, Damico A, and Neuman T. “What’s In and What’s Out? Medicare Advantage Market Entries and Exits for 2016.” October 2015. https://modern.kff.org/medicare/issue-brief/whats-in-and-whats-out-medicare-advantage-market-entries-and-exits-for-2016/ ↩︎
3. For example, see Gold M, Jacobson G, Damico A, and Neuman T, “Medicare Advantage 2014 Spotlight: Enrollment Market Update.” May 2014. https://modern.kff.org/medicare/issue-brief/medicare-advantage-2014-spotlight-enrollment-market-update/ ↩︎
4. For more on the factors affecting seniors’ plan enrollment decisions, see Jacobson G, Swoope C, Perry M, Slosar M, “How are Seniors Choosing and Changing Health Insurance Plans?” May 2014. https://modern.kff.org/medicare/report/how-are-seniors-choosing-and-changing-health-insurance-plans/ ↩︎
5. See Gold M, Hudson M, Jacobson G, and Neuman T, “Medicare Advantage 2010 Data Spotlight: Benefits and Cost-Sharing.” February 2010. https://modern.kff.org/medicare/issue-brief/medicare-advantage-2010-data-spotlight-benefits-and/ ↩︎
6. Hoadley J, Cubanski J, and Neuman T. “Medicare Part D: A First Look at Plan Offerings in 2016.” October 2015. https://modern.kff.org/medicare/issue-brief/medicare-part-d-a-first-look-at-plan-offerings-in-2016 ↩︎
7. Hoadley J, Cubanski J, and Neuman T. “Medicare Part D: A First Look at Plan Offerings in 2016.” October 2015. https://modern.kff.org/medicare/issue-brief/medicare-part-d-a-first-look-at-plan-offerings-in-2016 ↩︎
8. For a list of the counties, see Jacobson G, Neuman T, Damico A, and Huang J. “Medicare Advantage Plan Star Ratings and Bonus Payments in 2012.” November 2011. https://modern.kff.org/medicare/report/medicare-advantage-2012-star-ratings-and-bonuses/ ↩︎
9. See Jacobson G, Damico A, Neuman T, and Gold M. “Medicare Advantage 2015 Spotlight: Enrollment Market Update.” June 2015. https://modern.kff.org/medicare/issue-brief/medicare-advantage-2015-spotlight-enrollment-market-update/ ↩︎
10. Jacobson G, Damico A, Neuman T, and Gold M. “Medicare Advantage 2015 Spotlight: Enrollment Market Update.” June 2015. https://modern.kff.org/medicare/issue-brief/medicare-advantage-2015-spotlight-enrollment-market-update/ ↩︎
11. Blue Cross and Blue Shield Association. 2015 Medicare Advantage and Prescription Drug Plans Offered by Blue Cross and Blue Shield Affiliated Companies, Updated January 2015. http://www.bcbs.com/about-the-companies/ ↩︎
12. G. Jacobson, A. Damico, and T. Neuman. “Data Note: Medicare Advantage Enrollment, by Firm, 2015.” July 2015. https://modern.kff.org/medicare/issue-brief/data-note-medicare-advantage-enrollment-by-firm-2015/ ↩︎
13. Altman D. “Amid Merger Talk, a Look at Health Insurers’ Medicare Business.” Wall Street Journal. July 1, 2015. http://blogs.wsj.com/washwire/2015/07/01/amid-merger-talk-a-look-at-health-insurers-medicare-business/ ↩︎

In thousands of medically underserved communities across the U.S., community health centers enroll low-income people in health coverage and provide care to millions of patients. Against the backdrop of significant health center expansion over several years and a full year of expanded health coverage under the Affordable Care Act (ACA), this brief examines change between 2013 and 2014 in the volume and health coverage profile of health center patients, and health center enrollment activities and service capacity, comparing states that implemented the ACA Medicaid expansion in 2014 and states that did not expand Medicaid in 2014. The study is based on 2014 data from the federal Uniform Data System and a 2014 national survey of health centers.

Several key findings emerged from the analysis:

• In 2014, health centers served 22.5 million patients – an increase of nearly 1.2 million patients over 2013. From 2013 to 2014, the number of health center patients covered by Medicaid rose by 1.85 million, or 22%.
• Between 2013 and 2014, the share of health center patients covered by Medicaid rose from 40% to 46%, the share with private insurance rose from 14% to 16%, and the share who were uninsured declined from 35% to 28%.
• Changes in health center patient coverage were sharply different between Medicaid expansion and non-expansion states (Figure ES-1). In expansion states, the share of patients with Medicaid rose by 20% and the share who were uninsured fell by 29%. In non-expansion states, the share with Medicaid rose by 3% and the share who were uninsured fell by 8%. Private coverage rates remained low in both groups of states because most health center patients have income below the poverty level and are thus ineligible for Marketplace subsidies.
• During the first ACA open enrollment period, the vast majority of all health centers provided Medicaid application assistance, but expansion states were more likely to assist with renewing Medicaid and selecting Medicaid plans, while non-expansion states were more likely to assist with Marketplace applications and Qualified Health Plan selection.
• Health centers in expansion states were significantly more likely than those in non-expansion states to report having expanded their capacity for dental and mental health services since the start of 2014. They were also more likely to report increased wait times for appointments, possibly reflecting greater increases in demand for services associated with larger gains in coverage among health center patients in these states.

Introduction

Community health centers play a major role in providing health care for millions of people who reside in areas designated by the federal government as medically underserved, based on high rates of poverty and infant mortality and other measures of health care need.¹  In 2014, 1,249 community health centers in the 50 states and DC served nearly 22.5 million people.²  Another 81“look-alike” health centers, which meet all health center requirements but receive no federal funding, served nearly 900,000 additional patients.³  Health center patients are much more likely to be poor or low-income than Americans overall, and they are much more likely to be uninsured.⁴  Six in ten health center patients are nonelderly adults – a population largely excluded from Medicaid before the ACA.

The ACA expanded Medicaid eligibility to nonelderly adults up to 138% of the federal poverty level (FPL) and provided income-related subsidies for the purchase of private insurance through new Marketplaces for people  with income between 100% and 400% FPL. Both these new avenues of coverage, but particularly the Medicaid expansion, have the potential to lower the uninsured rate among health center patients, a largely low-income population. Following the Supreme Court ruling that effectively made the Medicaid expansion to adults optional for states, many, but not all, states implemented it. In states that did not expand Medicaid, adults with incomes between 100% and 138% FPL are eligible for Marketplace subsidies. But in these states, many adults below 100% FPL fall into a coverage gap because they are above the Medicaid income cut-off in their state but below the income eligibility threshold for Marketplace subsidies. In Medicaid expansion states, there is a continuum of coverage options for low-income adults, with Medicaid for those up to 138% FPL and Marketplace subsidies for those with higher income.

The ACA coverage expansions have important implications for health centers, which anchor health care in underserved communities, providing comprehensive primary care, case management, and enabling services such as medical transportation, and serving as an entry point for access to more specialized care. In addition, health centers have experience providing onsite outreach and enrollment assistance to uninsured patients who may be eligible for Medicaid or other coverage. During the first ACA open enrollment period, health centers provided enrollment assistance to more than 10 million people.⁵ 

To investigate and compare the historical and recent experience of health centers in Medicaid expansion and non-expansion states, the Kaiser Family Foundation and the Geiger Gibson RCHN Community Health Foundation Research Collaborative at George Washington University’s Milken Institute School of Public Health used federal data to assess changes in health center patient volume and insurance coverage in the two groups of states. Findings from a nationwide survey of health centers, fielded in Fall 2014, provide insight into additional differences between health center experiences in Medicaid expansion and non-expansion states, related to their enrollment assistance efforts and service capacity. This survey followed on a Fall 2013 survey to assess health centers’ outreach and enrollment activities on the eve of the first year of ACA coverage expansions.⁶ 

Data and Methods

We analyzed data from the federal Uniform Data System (UDS), into which all health centers report annually, to assess growth in the volume of health center patients between 2013 and 2014 and over the 2004-2014 decade; 2014 UDS data became available for analysis in late August 2015. We also used UDS data to examine 2013-2014 changes in the health coverage profile of health center patients overall and separately in Medicaid expansion and non-expansion states.⁷ 

We used data from the 2014 Follow-up Survey of Health Centers’ Outreach and Enrollment Activities to analyze health centers’ enrollment assistance activities during the first ACA open enrollment period (October 1, 2013 through March 31, 2014); major barriers to their current enrollment efforts; and increases in selected measures of service capacity since January 1, 2014. We emailed the survey between late September and November 2014 to all health centers listed in the 2013 UDS (2014 UDS data were not yet available). A total of 679⁸  health centers in all 50 states and DC responded, yielding an overall response rate of 57%; the response rate was 50% or higher in 44 states and DC, and the lowest rate in any state was 40%. The survey responses were almost evenly distributed between the Medicaid expansion states (including DC) (n=364) and the non-expansion states (n=315). To adjust for observed differences between respondents and non-respondents, we weighted the survey data by total health center patients, the proportion reported as racial/ethnic minorities, and total revenue per patient, using the 2014 UDS data.

To investigate differences between the experiences of health centers in Medicaid expansion and non-expansion states, we sorted the UDS and survey data based on the Medicaid expansion status of the state in which the health center grantee was located. In most of our analyses, we counted states as Medicaid expansion states if they implemented the expansion at any time during 2014 (Figure 1). However, in our analysis of health center enrollment activities during the first ACA open enrollment period, we counted states as expansion states only if their Medicaid expansion was in effect on January 1, 2014. Therefore, two states that expanded Medicaid later in 2014 (Michigan and New Hampshire) were considered non-expansion states for this purpose.⁹ 

Findings

The number of people served by health centers and the number covered by Medicaid both grew substantially from 2013 to 2014.

• In 2014, health centers served 22.5 million patients, an increase of 1.2 million patients over 2013. Over the course of the decade 2004-2014, the number of health centers rose by more than one-third, from 887 to 1,249, and the volume of patients served by health centers nearly doubled (Figure 2).

• Between 2013 and 2014, the number of health center patients with Medicaid coverage rose by 1.85 million, or 22%. Over the decade 2004-2014, the number of health center patients covered by Medicaid more than doubled, rising from 4.4 million to 10.4 million. The Medicaid increase during the decade is attributable to multiple factors, including the impact of demographic, programmatic, and economic changes (e.g., population growth, state Medicaid eligibility policies, and the 2007-2009 economic recession). It is also attributable to the ACA Medicaid expansion in states that chose to implement it.

health coverage among health center patients changed markedly between 2013 and 2014, the year the ACA Coverage expansions took effect.

• The share of health center patients with Medicaid increased sharply. From 2013 to 2014, the first year the Medicaid expansion was in effect in the states that adopted it, the share of health center patients with Medicaid grew by 15%, from 40% to 46% (Figure 3). To put this one-year increase in perspective, the average annual increase share of health center patients with Medicaid during the 10-year period 2004-2014 was 3% (data not shown).

• The uninsured rate among health center patients declined sharply. The uninsured rate among health center patients declined by 20% between 2013 and 2014, dropping from 35% to 28%. Once more, for perspective, the average annual decline in the uninsured rate among health center patients over 2004-2014 was approximately 3.5%.
• The share of health center patients with private insurance increased, but remained small. The share of health center patients with private coverage rose from 14% to 16% between 2013 and 2014. The private coverage rate among health center patients remains low due to health center patients’ low income, lack of access to employer-sponsored insurance, and limited access to subsidies for Marketplace coverage.

health center patients’ coverage profile is different in medicaid expansion versus non-expansion states.

• Even before the ACA Medicaid expansion took effect in 2014, Medicaid coverage rates were higher and uninsured rates were lower in the expansion states than in the non-expansion states. In 2013, 44% of health center patients were covered by Medicaid in states that expanded Medicaid in 2014, compared to 33% of health center patients in the states that did not expand Medicaid. At least in part, this difference reflects higher pre-ACA Medicaid eligibility in the expansion states. In 2013, the median Medicaid income eligibility threshold for adults with dependent children was 106% FPL in states that expanded Medicaid in 2014, but just 48% FPL in states that did not expand Medicaid.¹⁰ 
• Between 2013 and 2014, the Medicaid coverage rate among health center patients increased much more in Medicaid expansion states than non-expansion states, and to a higher level. In the 2013-2014 timeframe, the share of health center patients covered by Medicaid rose by 20%, from 44% to 53%, in the expansion states. In the non-expansion states, the share with Medicaid rose by 3%, from 33% to 34% (Figure 4).

• The share of health center patients with private insurance grew more in Medicaid non-expansion than expansion states, but remained low even in non-expansion states. Between 2013 and 2014, private coverage among health center patients increased from 15% to 17% in the non-expansion states and from 14% to 15% in the Medicaid expansion states. While small, the increase indicates that some patients who obtain private insurance continue to rely on health centers for care. Still, the relatively small share of health center patients with private coverage even in non-expansion states indicates that Marketplace subsidies have limited potential as a coverage pathway for the health center patient population – a large majority of health center patients (71%) have incomes below 100% FPL, making them ineligible for Marketplace subsidies.¹¹ 
• The uninsured rate among health center patients declined much more in the Medicaid expansion states, and the gap between expansion and non-expansion states widened. Between 2013 and 2014, the uninsured rate for health center patients in expansion states fell by 29%, from 32% to 22%; it fell by 8%, from 41% to 38%, in non-expansion states. In short, for health center patients, insurance coverage in non-expansion states fell even further behind coverage in expansion states. In expansion states, there is a continuum of coverage options for low-income adults, with Medicaid eligibility for those up to 138% FPL and Marketplace subsidies for those with income above that level. On the other hand, in non-expansion states, many uninsured adults below poverty remain in a coverage gap, largely excluded from Medicaid in their states but also ineligible for Marketplace subsidies.¹² 

Health Centers in Medicaid Expansion and Non-expansion States Invest Differently in Enrollment Assistance.

• During the first ACA open enrollment period, health centers in Medicaid expansion states were more likely than those in non-expansion states to provide Medicaid enrollment assistance. All health centers received federal grants to provide outreach and enrollment assistance during the first ACA open enrollment period (October 1, 2013 – March 31, 2014). In the health center survey, more than 80% of health centers in both Medicaid expansion and non-expansion states reported providing assistance with Medicaid applications during this first open enrollment period. However, health centers in expansion states were significantly more likely than those in non-expansion states to assist patients with Medicaid renewal (78% vs. 66%) and Medicaid plan selection (69% vs. 49%) (Table 1).
• Health centers in non-expansion states were more likely than those in expansion states to provide Marketplace enrollment assistance. By a small but statistically significant increment, a larger share of health centers in non-expansion states provided assistance with applications for Marketplace subsidies in the first ACA open enrollment period, compared to health centers in Medicaid expansion states (94% vs. 89%). Health centers in non-expansion states were also significantly more likely to provide assistance with a Qualified Health Plan selection compared to health centers in expansion states (87% vs. 79%). Precluded from securing Medicaid for most of their poorest uninsured patients, health centers in non-expansion states may have been more inclined to invest in assistance to help those with income of at least 100% FPL obtain coverage through the Marketplace.

Health centers in Medicaid Expansion and Non-expansion States Report Some Common and Some Different major Barriers to Current Enrollment Activities.

• Among health centers in non-expansion states, the state decision not to expand Medicaid was, by far, the most frequently reported major barrier to enrollment assistance. Nearly all (94%) health centers in non-expansion states ranked their state’s decision not to expand Medicaid as a major barrier to their enrollment assistance activities (Table 2).
• About half of health centers in both groups of states reported patient confusion about eligibility for coverage as a major barrier. The large share of all health centers citing this issue indicates that patients’ understanding of their coverage options was unrelated to state Medicaid expansion status and, rather, that confusion was widespread everywhere in the early post-ACA implementation period studied.
• Large shares of health centers in both Medicaid expansion and non-expansion states ranked problems related to documentation requirements and inadequate state information as major barriers to enrollment assistance. About 40% of health centers in Medicaid expansion states ranked incomplete or missing income, residency, or citizenship information as a major barrier, as did a significantly smaller but still large share (30%) of health centers in non-expansion states. Likewise, inaccurate or inconsistent answers from the state were considered a major barrier by 40% of health centers in expansion states and 30% of those in non-expansion states. One-quarter of health centers in expansion states and more than one-third in non-expansion states ranked inadequate consumer information provided by the state as a major barrier.
• Inadequate outreach by the state was also cited as a major barrier to enrollment assistance. One in three health centers in non-expansion states, and one in five in expansion states, ranked inadequate state outreach as a major barrier to their current enrollment activities. Roughly one in five health centers in both groups of states considered the lack of a state procedure manual and inadequate funding major barriers; more than one in 10 health centers in both groups cited the lack of culturally and/or linguistically appropriate materials.

Health Center Service Capacity is More Likely to have Expanded in Medicaid Expansion states.

• Health centers in Medicaid expansion states were significantly more likely than those in non-expansion states to report expanded service capacity since January 2014. Health centers in expansion states were more likely than those in non-expansion states to have increased dental service capacity (37% vs. 31%) and mental health service capacity since January 2014 (42% vs. 35%) (Figure 5). These results are consistent with a finding from previous work that higher Medicaid income eligibility for adults and greater health center capacity were correlated.¹³  It is reasonable to surmise that increased patient revenues generated by increased coverage among low-income populations help health centers to expand their service capacity.

• At the same time, wait times for health center appointments increased more in expansion states than in non-expansion states. Health centers in expansion states were more likely to report increased wait times to obtain an appointment compared to health centers in non-expansion states (35% vs. 20%). Increased wait times for appointments may reflect greater demand for health center services in Medicaid expansion states and increased strains on health center capacity at least in the short term.

Discussion

The findings from this study demonstrate that expanded health coverage under the ACA has lowered the uninsured rate among health center patients in all states. The largest gains in coverage among health center patients have occurred in Medicaid expansion states, a reflection of the deep poverty of the communities that health centers serve. In non-expansion states, the subgroup of health center patients with incomes equal to at least 100% FPL can qualify for Marketplace subsidies. The uptick in private coverage rates among health center patients in non-expansion states suggests that this coverage pathway – premium subsidies for Marketplace coverage – make a difference to health center patients as well as other low income people in non-expansion states.¹⁴  However, as important as the Marketplace subsidies are, most health center patients are too poor to qualify for them; furthermore, those who do qualify can face premiums and significant out-of-pocket burdens for deductibles, coinsurance, and copayments, depending on the plan they select. As coverage continues to expand under the ACA, health center grants will remain essential to keep care affordable for privately insured patients through income-related fee reductions, and to support the fundamental safety-role health centers play in serving the uninsured, who account for more than one in five health center patients even in states that have expanded Medicaid and almost 40% of health center patients in non-expansion states.

The finding that health centers in Medicaid expansion states were more likely to have expanded service capacity suggests that new patient revenues associated with the Medicaid expansion improved health centers’ financial position to invest. The health centers in non-expansion states lagged behind, but the grant funding provided to all health centers through the special ACA trust fund has provided crucial resources to help them build capacity to provide care, too.

Regardless of their state’s Medicaid expansion status, health centers play a crucial role in providing outreach and enrollment assistance to their patients and communities. Substantial gains in health coverage among health center patients provide new evidence of the impact of their efforts. Health centers are well-equipped to assist patients who are very poor, new to navigating a complex system of coverage, enrollment, and plan selection, and often without access to technology necessary to enroll online. Even so, despite streamlined enrollment systems under the ACA, patient confusion about eligibility and documentation requirements pose major challenges to health centers’ current enrollment activities. Health center grant funding will remain important to sustaining health centers’ ability to link their patients and communities to coverage.

Looking Ahead

As ACA implementation proceeds, a key ongoing role for health centers will be to identify and enroll uninsured health center patients who are eligible for coverage. Some patients remain ineligible for Medicaid or ACA coverage because of their immigration status. But health centers’ ability to connect patients with coverage is most powerfully influenced by state Medicaid expansion decisions, and at this writing, 20 states have not moved forward, leaving millions uninsured and straining health centers’ capacity to serve them. These non-expansion states are mostly in the South,¹⁵  where poverty is deepest, communities of color are more concentrated,¹⁶  and gaps in access to preventive care and treatment, leading to avoidable complications of disease and poor health outcomes¹⁷  are particularly great. It will be important to continue to assess the implications of Medicaid expansion for health center primary care capacity, including access to pediatric care, women’s health services, general adult medicine, and services for special needs populations, such as patients with physical and mental health disabilities. Understanding the broader impacts of Medicaid expansion on low-income communities, for example, on employment opportunities and local economies, would also help to inform state policy choices.

The high uninsured rate even among health center patients in Medicaid expansion states underscores the continuing importance of federal grant funding for health centers, as well as coverage expansion, to sustain and improve access to care in underserved communities. In addition, considering the significant role of health centers in serving Medicaid beneficiaries and other low-income populations, the extent to which they are integrated into emerging delivery systems that involve new provider alignments and performance-driven payment systems will have important implications for health care access and quality, the future of the health care safety-net, and efforts to improve population health.

1. Health Resources and Services Administration. Negotiated Rulemaking Committee on the Designation of Medically Underserved Populations and Health Professional Shortage Areas: Final Report to the Secretary, Table 5. 2011. http://www.hrsa.gov/advisorycommittees/shortage/nrmcfinalreport.pdf ↩︎
2. The total number of health centers in 2014 was 1,278 when health centers in the U.S. territories are included. Bureau of Primary Health Care, Health Resources and Services Administration. National 2014 Health Center Data, 2015. http://bphc.hrsa.gov/uds/datacenter.aspx?q=tall&year=2014&state= ↩︎
3. Bureau of Primary Health Care, Health Resources and Services Administration. National 2014 Look-Alikes Data, 2015 http://bphc.hrsa.gov/uds/lookalikes.aspx?state=national ↩︎
4. Shin P et al., A Profile of Community Health Center Patients: Implications for Policy, 2013, Kaiser Commission on Medicaid and the Uninsured, Kaiser Family Foundation and the RCHN Community Health Foundation. https://modern.kff.org/medicaid/issue-brief/a-profile-of-community-health-center-patients-implications-for-policy/ ↩︎
5. Sharac J et al., How has the Affordable Care Act Benefitted Medically Underserved Communities? National Findings from the 2014 Community Health Centers Uniform Data System, 2015, Policy Research Brief No. 42, Geiger Gibson/RCHN Community Health Foundation Research Collaborative, George Washington University. http://www.rchnfoundation.org/?p=4998 ↩︎
6. Shin P et al., Assessing the Potential Impact of State Policies on Community Health Centers’ Outreach and Enrollment Activities, 2014, Policy Research Brief No. 35, Geiger Gibson/RCHN Community Health Foundation Research Collaborative, George Washington University. http://www.rchnfoundation.org/?p=3814 ↩︎
7. Data on health centers in the U.S. territories were excluded from this analysis. ↩︎
8. The number of responses including health centers in U.S. territories was 693. ↩︎
9. Kaiser Family Foundation. State Health Facts: Status of State Action on the Medicaid Expansion Decision, November 2015. https://modern.kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/ ↩︎
10. Kaiser Family Foundation. Medicaid Eligibility for Adults as of January 1, 2014, https://modern.kff.org/medicaid/fact-sheet/medicaid-eligibility-for-adults-as-of-january-1-2014/ ↩︎
11. Bureau of Primary Health Care, Health Resources and Services Administration. National 2014 Health Center Data, 2015. http://bphc.hrsa.gov/uds/datacenter.aspx?q=tall&year=2014&state= ↩︎
12. Garfield R et al., The Coverage Gap: Uninsured Poor Adults in States That Do Not Expand Medicaid: An Update. Kaiser Family Foundation, 2015. http://files.kff.org/attachment/issue-brief-the-coverage-gap-uninsured-poor-adults-in-states-that-do-not-expand-medicaid-an-update ↩︎
13. Shin P et al., Medicaid and Community Health Centers: The Relationship between Coverage for Adults and Primary Care Capacity in Medically Underserved Communities. Kaiser Family Foundation, 2012. https://modern.kff.org/health-reform/issue-brief/medicaid-and-community-health-centers-the-relationship/ ↩︎
14. This hypothesis is consistent with recent findings that individuals with income of 100-150% FPL made up nearly half of Marketplace purchasers in non-expansion states, but under one-quarter of those in Medicaid expansion states.  U.S. Department of Health and Human Services. “Health Insurance Marketplaces 2015 Open Enrollment Period: March Enrollment Report.” ASPE Issue Brief (2015): 1-72. http://aspe.hhs.gov/pdf-report/health-insurance-marketplace-2015-open-enrollment-period-march-enrollment-report ↩︎
15. Garfield et al., op. cit. ↩︎
16. Artiga S et al., The impact of the coverage gap in states not expanding Medicaid by race and ethnicity, Kaiser Family Foundation, 2015. https://modern.kff.org/disparities-policy/issue-brief/the-impact-of-the-coverage-gap-in-states-not-expanding-medicaid-by-race-and-ethnicity/ ↩︎
17. Schoen C et al., Health Care in the Two Americas: Findings from the Scorecard on State Health System Performance for Low-Income Populations, 2013, The Commonwealth Fund, 2013. http://www.commonwealthfund.org/publications/fund-reports/2013/sep/low-income-scorecard   ↩︎

Medicare Part D drug plans differ considerably in the drugs they list on their formularies, their use of formulary tiers, and the level and structure of cost sharing applied to those tiers. Plan premiums and the use of deductibles also vary widely. Plan decisions affect different beneficiaries in different ways, depending on the drugs they use. The financial consequences for Part D plan enrollees can be substantial. In this brief, we focus on out-of-pocket drug costs for Part D enrollees in 2016 for specialty, brand, and generic drugs.

The analysis is based on a selected set of specialty drugs and commonly prescribed brand and generic drugs. Out-of-pocket costs are taken from the Medicare Plan Finder, available at medicare.gov, as they apply to Part D enrollees who do not qualify for the Low-Income Subsidy. Drug costs were obtained for 20 prescription drug plans (PDPs) offered on a national or near-national basis, using a location in Baltimore, MD. In addition to examining costs for common drugs, we also examine profiles of multiple drugs for five hypothetical Part D enrollees. More details on the study methods can be found in Appendix 1: Methods.

Findings include:

• Part D enrollees can expect to pay thousands of dollars out of pocket for a single specialty drug in 2016, even after their drug costs exceed the catastrophic coverage threshold.
• Out-of-pocket costs are substantially higher—often ten times higher or more—for specialty drugs when they are not listed on formulary by a Part D plan.
• Out-of-pocket costs for specialty drugs tend to be similar across Part D plans when on formulary and are typically subject to prior authorization.
• Monthly out-of-pocket costs for commonly used brand and generic drugs tend to vary widely across Part D plans, even when included on plan formularies; for five of ten top brands, monthly costs vary by as much as $100 across plans.
• Out-of-pocket costs for commonly used brand and generic drugs are often significantly higher when they are off formulary than when they are on formulary; for six top brands and one top generic drug, costs are at least $200 more per month when off formulary than the median cost on formulary.
• Based on five hypothetical beneficiaries taking a mix of medications for multiple conditions, total out-of-pocket costs vary by as much as four-fold across Part D plans, taking into account the number and type of drugs they take, whether or not their drugs are on formulary, cost-sharing amounts, and monthly premiums.

Part D enrollees can expect to pay thousands of dollars out of pocket for a single specialty drug in 2016, even after their drug costs exceed the catastrophic coverage threshold

Specialty drugs, defined by Medicare as those that cost more than $600 per month, can contribute to high out-of-pocket costs for Part D enrollees. For 12 specialty drugs used to treat four health conditions —hepatitis C, multiple sclerosis, rheumatoid arthritis, and cancer—Part D enrollees face at least $4,000 and as much as nearly $12,000 in annual out-of-pocket costs in 2016 for one drug alone (Figure 1). (See Appendix Table 1 for drug-specific cost information.)

Among the studied 12 specialty drugs median out-of-pocket costs are lowest for the three drugs used to treat rheumatoid arthritis. But these costs still range from $4,413 to $4,872 for a year’s treatment. Median out-of-pocket costs for Medicare beneficiaries taking any one of three hepatitis C specialty drugs range from $6,516 to $7,153 in 2016. Some cancer drugs cost Part D enrollees even more. Revlimid, used to treat cancer of the blood, including multiple myeloma and a form of lymphoma, has a median out-of-pocket cost of $11,538, while Gleevec (for certain types of leukemia) has a median out-of-pocket cost of $8,503.

At least one-third of total out-of-pocket costs for each of these 12 specialty tier drugs are incurred by Part D enrollees after their spending reaches the catastrophic coverage phase of the Part D benefit. For seven of these drugs, enrollees pay more than half of their total out-of-pocket costs above the catastrophic threshold. For example, 58 percent of an enrollee’s out-of-pocket costs for Sovaldi (a treatment for hepatitis C) occur in the catastrophic coverage phase, which translates to $3,828 in costs in the catastrophic phase in 2016. For Revlimid, this share rises to 76 percent, or $8,758 in out-of-pocket costs in the catastrophic coverage phase. These examples assume that the particular specialty drug in question is the only drug taken by a Part D enrollee. In the likely event that they take other drugs, the share of costs in the catastrophic phase will be even greater.

Part D enrollees taking high-cost specialty tier drugs often incur significant costs beyond the catastrophic threshold because the threshold is not an absolute limit on out-of-pocket spending. By law, Part D enrollees pay 5 percent of drug costs after exceeding the catastrophic threshold (equivalent to $7,515 in total drug costs in 2016 under the standard benefit).¹  Therefore, despite having catastrophic coverage, Part D enrollees can face thousands of dollars in annual out-of-pocket costs if they take expensive drugs (or, as described later, if they take multiple high-cost brand-name drugs).

Part D enrollees who take specialty drugs face especially high costs at the start of the year (Figure 2). To illustrate, beneficiaries enrolled in the SilverScript Choice PDP face out-of-pocket costs of $6,221 in 2016 for Copaxone, a drug used to treat multiple sclerosis with a full cost of about $74,000. But they are liable for $2,404—over a third of the full year’s cost—in January. In this example, costs drop to $704 in February and then to $311 per month for the rest of the year. Similar patterns of front-loaded spending occur in other PDPs and for other high-cost specialty drugs. For enrollees who take specialty drugs over multiple years, the pattern repeats.

This pattern occurs because the Part D benefit includes an initial coverage period during which enrollees pay from 25 percent to 33 percent (and, for two-thirds of plans, a deductible). The monthly cost of many specialty drugs is greater than the initial coverage limit ($3,310 in 2016), so enrollees shift quickly into the gap phase, during which enrollees pay 45 percent of the drug’s cost by statute (in 2016). Once the catastrophic threshold ($7,515 in total drug costs in 2016) is passed, enrollees pay 5 percent of the drug’s cost for the rest of the year.

Out-of-pocket costs are substantially higher—often ten times higher or more—for specialty drugs when they are not listed on formulary by a Part D plan

The most significant driver of out-of-pocket costs for a specialty drug, given the extraordinarily high cost of these drugs, is whether or not the drug is on formulary. Six of the 12 specialty drugs included in this study are on formulary in all plans—including two of the three hepatitis C drugs and all three of the cancer drugs (Figure 3). Cancer drugs are required to be on formulary by CMS guidance as one of six protected classes. Plans are also required to include at least two drugs on formulary in every category or class. Although the drugs in this study are not the only drugs in their drug classes, this requirement is a key factor in formulary design decisions.

The other six specialty drugs in this study are off formulary in at least four PDPs, including all three of the drugs used to treat multiple sclerosis (MS). Although all 20 PDPs include at least one of the three studied MS drugs on formulary (there are other treatments for MS as well), 14 PDPs include just one of the three MS drugs. Plans omit drugs from their formularies for a variety of reasons, and they must use their pharmacy and therapeutics committee to incorporate clinical considerations in their decisions. A key reason for exclusion is price negotiation: plans reportedly get the deepest discounts from a manufacturer when competing drugs are excluded from the plan’s formulary.

The cost to the beneficiary when a specialty drug is off formulary is nearly ten times higher—or more—than the median cost when on formulary (Table 1). This represents an added out-of-pocket cost, from nearly $40,000 (for Orencia or Enbrel) to nearly $90,000 (for Viekira Pak).

Out-of-pocket costs for specialty drugs tend to be similar across Part D plans, when on formulary, and are typically subject to prior authorization

For these 12 specialty drugs, the maximum out-of-pocket cost in PDPs listing the drug on formulary is never more than 10 percent higher than the minimum out-of-pocket cost. The out-of-pocket cost for Harvoni, the most common treatment for hepatitis C, ranges from $7,072 to $7,245, a difference of only 2 percent (Figure 4). The cost of Enbrel, the most commonly prescribed treatment for rheumatoid arthritis, ranges from $4,571 to $4,926 when on formulary, a difference of 8 percent. The pattern for the two drug classes shown in Figure 4 is comparable to the other two drug classes in this study.

While plans differ in setting specialty-tier coinsurance in the initial coverage phase at rates ranging from 25 percent to 33 percent, the vast majority of total costs to the consumers taking specialty drugs is established by law and does not vary by much across plans. This is because the different coinsurance rates apply only in the initial coverage period, and the majority of costs incurred by consumers occur in the coverage gap and catastrophic phases where cost sharing is set by law.² 

When included on plans’ formularies, 11 of the 12 specialty drugs included in this analysis are always covered on the specialty tier. One of the 12 drugs—Orencia, for rheumatoid arthritis—is covered on a non-preferred brand tier by the two PDPs offered by UnitedHealth. This placement results in higher enrollee cost sharing in the initial coverage period than if the drugs were on the specialty tier, because the coinsurance for non-preferred brand drugs in these two plans is higher than for specialty drugs: 40 percent in the AARP MedicareRx Saver Plus PDP and 50 percent in the AARP MedicareRx Preferred PDP (in pharmacies offering preferred cost sharing, these amounts are 30 percent and 40 percent, respectively).

Specialty drugs are regularly subject to prior authorization and quantity limits. For eight of the 12 specialty drugs studied, all plans that list these drugs on formulary also require prior authorization, and most plans do the same for the other four drugs (Copaxone, Orencia, Enbrel, and Zytiga). In only two cases do plans require step therapy for these specialty drugs: one plan for Orencia and another plan for Tecfidera. (See Appendix Table 2 for drug-specific tier placement and utilization management information.)

Monthly out-of-pocket costs for commonly used brand and generic drugs tend to vary widely across Part D plans, even when included on plan formularies; for five of ten top brands, monthly costs vary by as much as $100 across plans

Top Brand Drugs

Cost sharing for a 30-day supply of ten commonly prescribed brand drugs can be anywhere from two to 14 times higher in one PDP versus another (Figure 5). For example, Namenda (a drug used by people with Alzheimer’s disease with a typical price of $345) has median monthly cost sharing of $142 in 2016, but an enrollee in Aetna Medicare Rx Saver PDP or First Health Part D Value Plus PDP pays only $40, whereas an enrollee in WellCare Classic PDP pays $173. (See Appendix Table 1 for drug-specific cost information.)

Compared to specialty drugs, total out-of-pocket costs are more heavily driven by how plans structure their benefits and formularies. Differences across plans in out-of-pocket costs for enrollees are driven by three factors: (1) tier placement (typically, two tiers for generics and two tiers for brands), (2) cost-sharing amounts assigned to the tiers by plans, and (3) whether cost sharing is structured as copayments or coinsurance (mostly coinsurance for non-preferred brands, mixed for preferred brands). Most brand drugs not on the specialty tier are not costly enough by themselves to force the beneficiary into the catastrophic coverage phase.

The impact of benefit design variations is illustrated by the drug with the greatest range in cost sharing (including only PDPs where the drug is on formulary). The cost to the consumer of Spiriva, a drug used to treat emphysema or COPD with a typical price of $950, ranges from $33 to $472 per month in 2016. The highest cost sharing for this drug ($472) is for a PDP that places the drug on a non-preferred brand tier with 50 percent coinsurance. The lowest cost sharing ($33) is for a PDP with a flat copayment amount for a preferred brand tier. In PDPs that place Spiriva on a preferred brand tier with coinsurance for a preferred brand tier, the cost to the consumer is also high, ranging from $153 to $245 depending mostly on variations in the coinsurance percentage, but also in the retail price of the drug.

By contrast, the smallest range in cost sharing ($29 to $56) is for Crestor (for high cholesterol), a drug with a price of about $220 per month. The small range in cost sharing for Crestor partially reflects pricing at a level where coinsurance typically matches the copayments. In general, any brand drug priced higher than $200 per month has higher out-of-pocket costs in PDPs that use coinsurance rather than a copayment. Thus, copayments for Crestor range from $29 to $47 and PDPs with coinsurance charge from $35 to $56.

For a less expensive brand drug such as ProAir (a treatment for asthma priced just over $50), consumers face lower costs when plans use coinsurance. Based on the relatively low price of this drug, PDPs with coinsurance charge their enrollees from $8 to $13 per month, while PDPs with copays charge from $33 to $47.

In most cases, plan enrollees pay more when the drug is on a non-preferred tier as opposed to a preferred tier. But there are exceptions. Synthroid (for hypothyroidism) is typically priced at about $32 per month. For five PDPs that place the drug on a non-preferred brand tier with coinsurance, enrollees pay between $14 and $17 per month in 2016. By contrast, seven PDPs with preferred brand tier use copayments that exceed the price of this drug; thus, the enrollee pays the entire cost of the drug—about $32.

Top Generic Drugs

The range in out-of-pocket costs across Part D plans for a 30-day supply for ten top generic drugs tends to be much narrower than for brands (Figure 6). Most Part D enrollees who use the ten top generic drugs pay no more than $10 per month in cost sharing in 2016. The median cost sharing for all but one of these ten generics is between $3 and $8. Four of the top generics (furosemide, lisinopril, metformin, and metoprolol) are preferred generics for all 20 PDPs with a median copay of $3 or $4 and can be obtained for a zero copayment in some PDPs.

However, in some cases, the differences in costs across plans are greater. Among the ten generic drugs included in this analysis, hydrocodone/APAP, an opioid used for pain relief, is covered by most plans as if it were a brand drug, which leads to higher costs for enrollees. Hydrocodone/APAP is placed on a brand tier for eight of the ten PDPs that have it on formulary. As a result, cost sharing in these ten PDPs is much higher than other generics, varying from $16 to $78 in 2016. (The full price for a month’s supply of hydrocodone/APAP also varies, ranging from $94 to $184 for the PDPs that have it on formulary.)

The two generic statins used to treat high cholesterol are treated somewhat differently. Atorvastatin has a median monthly copayment of $7 as result of being on the non-preferred generic tier on 12 of 20 PDPs (and on the non-preferred brand tier for 1 PDP). Simvastatin is a preferred generic for 17 of 20 PDPs and has a median copayment of $4. But, regardless of the PDP selected, either one of these generic drugs is less expensive for beneficiaries than the brand statin (Crestor), which has cost sharing of $29 to $56.

Out-of-pocket costs for commonly used brand and generic drugs are often significantly higher when they are off formulary than when they are on formulary; for six top brands and one top generic drug, costs are at least $200 more per month when off formulary than the median cost on formulary

Most (but not all) generics are on formulary for all PDPs, whereas many brand drugs are off formulary for a subset of PDPs. Eight of the ten top generic drugs in this study are always on formulary in 2016; hydrocodone/APAP is off formulary for 10 of 20 PDPs, and omeprazole is off formulary for 1 PDP. Just two of the ten top brands in this study are always on formulary. The remaining eight brands are off formulary for as few as 3 PDPs and as many as 7 PDPs of the 20 PDPs in this study.

Beneficiaries can face considerable costs for brands and for some generics when their drug is not listed on the plan’s formulary. For six of the top brands and one generic, the difference between median out-of-pocket cost when a drug is on formulary and the highest off-formulary cost is at least $200 per month in 2016 (Figure 7). A beneficiary with Alzheimer’s disease who takes Namenda will pay $392 per month in the five PDPs that leave this drug off formulary, compared to median cost sharing of $142. A beneficiary using Januvia to treat diabetes will see out-of-pocket costs increasing from $47 to $382.

Even generic drugs can be expensive when off formulary. Beneficiaries may pay up to $114 for a month’s supply of omeprazole, compared to a monthly copayment of $5. Similarly, the price of hydrocodone/APAP is as high as $237 instead of a typical monthly copayment of $36.

Access to some drugs may be further influenced by utilization management restrictions, although these restrictions are applied less frequently for brands and generics than for specialty drugs. Two of the ten top brands in this study have prior authorization requirements in 2016. Six PDPs apply prior authorization to Lyrica, a treatment for nerve and muscular pain, and 13 PDPs have prior authorization requirements for Namenda, a drug used for patients with Alzheimer’s disease. Several drugs also have step therapy requirements: Lyrica (one PDP), Januvia (two PDPs), and both versions of Lantus (one PDP). In addition, one PDP has a step therapy requirement for two generic drugs (atorvastatin and hydrocodone/APAP). (See Appendix Table 2 for drug-specific tier placement and utilization management information.)

Based on five hypothetical beneficiaries taking a mix of medications for multiple conditions, total out-of-pocket costs vary by as much as four-fold across Part D plans, taking into account the number and type of drugs they take, whether or not their drugs are on formulary, cost-sharing amounts, and monthly premiums

Few Medicare beneficiaries take only one drug, and for those who use the Medicare Plan Finder, the total out-of-pocket costs displayed reflect costs for the full array of drugs they take throughout the year along with plan premiums and deductibles, if applicable. Total out-of-pocket costs—cost sharing for all drugs and premiums—for five hypothetical beneficiaries who take at least five medications for various health conditions differ by at least $800 across 20 PDPs available in a given area in 2016 (Figure 8). Alice (Profile A), who takes six drugs for conditions that include asthma and diabetes, would spend $3,451 out of pocket if she enrolls in the SilverScript Choice PDP, whereas it would cost her $14,600 if she enrolls in Transamerica MedicareRx Classic PDP. The contrast is considerably less for Dora (Profile D) who takes five generic drugs for diabetes, hypertension, and three other conditions. Her annual costs would be $951 in the AARP Medicare Rx Saver Plus PDP, compared to $1,771 in the Humana Enhanced PDP.

The wide variation in out-of-pocket costs across Part D plans is due to several factors, including whether the beneficiary’s drugs are on or off formulary, each drug’s placement on a cost-sharing tier, the different cost-sharing amounts associated with each tier, whether cost sharing is structured as coinsurance or copayments, and plan premium amounts. Total costs vary less for Dora because her five generic drugs are all on formulary for the 20 PDPs in this study. For a beneficiary such as Alice who takes drugs that are not always on formulary (in her case, two of six drugs are off formulary in some PDPs), total costs are as much as $10,000 higher.

But even when all of their drugs are on formulary, total out-of-pocket costs across all PDPs can still vary substantially (Table 2). Bob (Profile B) takes five drugs for Alzheimer’s disease and other conditions. His drugs are all on formulary for 19 of 20 PDPs; yet his costs for those plans range from $1,697 (AARP Medicare Rx Saver Plus PDP) to $3,349 (SilverScript Plus PDP). Premium differences ($34 versus $87 in this example) tend to be a greater factor in cost differences when comparing plans that have all drugs on formulary.

Earl (Profile E) has multiple chronic conditions for which he takes 13 drugs, including 9 brand drugs (but no specialty-tier drugs). His out-of-pocket costs add up to $5,433 even in the least expensive plan for his situation (Humana Enhanced PDP), but can be as much as $18,687 in a plan (Transamerica MedicareRx Classic PDP) where four of his drugs off formulary. Notably, the least expensive plan for Earl is the most expensive plan for Dora, illustrating how much costs can vary based on individual needs. Earl’s costs, even in the least expensive plan for him, put him over the catastrophic threshold. Like beneficiaries who take a single specialty drug, a large share (37 percent) of Earl’s total out-of-pocket drug costs occur while he is in the catastrophic phase.

Pharmacy choice can affect total out-of-pocket costs, and pharmacies offering preferred cost sharing can provide savings, but the magnitude of savings varies by plan, pharmacy, and the mix of drugs

There is some variation in out-of-pocket costs by pharmacy—driven by plans’ use of tiered pharmacy networks, by pricing differences across pharmacies, and by the availability of mail order. The recent trend toward greater use of tiered pharmacy networks has raised the issue of pharmacy choice as another factor for enrollees to consider in comparing plans. Under these arrangements, plans make preferred cost sharing available at a limited number of network pharmacies and higher (standard) cost sharing at other network pharmacies.

For three of our five hypothetical beneficiaries (Alice, Carla, and Earl), the typical (median) savings achieved by using a pharmacy with preferred cost sharing is less than 3 percent in 2016—measured across 13 PDPs with a nearby pharmacy that offers preferred cost sharing (Figure 9). For the other two hypothetical beneficiaries, the median savings for using a pharmacy with preferred cost sharing is greater: 11 percent for Bob and 19 percent for Dora.

Savings from pharmacies offering preferred cost sharing can be substantial, especially for drug profiles that are more heavily reliant on generics, but can be quite modest when most drugs are brands. This is because plans that use tiered pharmacy networks typically reduce cost-sharing amounts more for generic drugs than for brands, perhaps because pharmacies have more leverage over discounts for generic drugs.³  For example, cost sharing for generics can drop from $10 down to $1 or even $0 at pharmacies offering preferred cost sharing.

Alice and Earl have the highest overall out-of-pocket costs for the drugs they take—largely because they are more reliant on brand drugs than the three other hypothetical beneficiaries. But they have the smallest rate of savings from using a pharmacy with preferred cost sharing (Table 3). In dollar terms, they have the potential to save less than $200 based on their chosen pharmacy in 2016—a small amount for a set of drugs that cost them at least $3,000 out of pocket. From a beneficiary’s perspective, the selection of a pharmacy or the selection of a plan based on the pharmacy they typically use is more important, in terms of savings, if they take a lot of generic drugs.

Mail order can also offer some savings to enrollees, compared to retail pharmacies, but not for all PDPs. Mail order savings are greater when compared to pharmacies with standard cost sharing (versus preferred cost sharing); all five of our hypothetical beneficiaries save money using mail order compared to retail for most PDPs when the chosen retail pharmacy is one that uses standard cost sharing (Table 4). In many cases, however, the savings are less than 10 percent. Out-of-pocket costs for mail order are closer to costs at retail pharmacies with preferred cost sharing. In fact, for some PDPs, out-of-pocket costs are higher for mail order than for a retail pharmacy with preferred cost sharing.

Discussion

This analysis shows that out-of-pocket costs can exceed thousands of dollars for beneficiaries who take just one specialty drug or a large number of less costly brands and generics. In 2014, 2 percent of Part D enrollees used specialty tier drugs, though not all enrollees who use these drugs will face out-of-pocket costs as high as those identified in this analysis.⁴   Out-of-pocket costs would be lower for enrollees who receive the Low-Income Subsidy and for enrollees who take specialty drugs that are less expensive than the drugs we analyzed.

Even with catastrophic coverage, Part D enrollees are exposed to high costs because the catastrophic coverage threshold is not a hard cap on out-of-pocket costs. In extreme situations, enrollees pay more money after reaching the catastrophic phase of the benefit than before it. Policymakers could strengthen financial protections for beneficiaries with significant drug costs by making the current catastrophic coverage threshold an absolute limit.

The highest costs occur when a prescribed drug is off formulary. Beneficiaries in this situation have the option to consult with their prescriber about therapeutic alternatives and get a different prescription if appropriate, or request an exception. Or they can try to obtain the drug through a source such as a manufacturer’s patient assistance program. Such programs, however, may require meeting income eligibility standards, and they must operate independently of the Part D program.

If the need for a new drug can wait until the first of the following year, after the open enrollment period, the high cost of specialty drugs, and even some brands, creates a strong incentive to enroll in a plan that has their drug on formulary. When a new drug is prescribed midyear, however, this delay may come with serious clinical consequences. Neither a new prescription for a drug that is off formulary nor a newly diagnosed health condition creates the basis for a midyear special enrollment period.

With median income among Medicare beneficiaries at about $24,000 per person, the cost of off-formulary specialty and other high-priced drugs is beyond the reach of many Part D enrollees.⁵  Some might be able to pay directly for off-formulary drugs out of pocket, but if they do, their costs will not count towards reaching their benefit phases, including the catastrophic threshold. Individual purchasers do not benefit from the manufacturer discounts that are available to plan purchasers in the form of post-invoice rebates paid by manufacturers to plans and kept confidential.

The Medicare Plan Finder can be a powerful tool to allow enrollees to predict total annual out-of-pocket costs across plans in their area, based on the drugs they take. Even so, the Plan Finder has inherent limitations. Beneficiaries can use the Plan Finder to search for plans based on drugs they are currently taking, but they cannot factor in their search the cost of a new prescription that comes after the enrollment period has ended. Nor can the Plan Finder easily show the beneficiary others ways to obtain a drug or what therapeutic alternatives are available. Furthermore, when a drug is off formulary, the price displayed may not reflect the actual price a beneficiary sees at the pharmacy.

Choice of plans matters in Part D. Beneficiaries’ health conditions and medication needs continually change, and plans often make changes in their formularies and cost sharing from one year to the next. The Plan Finder allows beneficiaries to compare plans during the open enrollment period, taking into account their current drugs and plan benefit designs for the coming year. A concise and personalized notice to enrollees describing how plan changes will affect them in the coming year might be one way to convince more to shop during the open enrollment period.

Our analysis shows that Part D enrollees can benefit from careful plan comparisons, looking not just at premiums, but also at formulary coverage and tier placement of their drugs, and at how their pharmacy is treated by plans. Yet while it pays to shop, most people do not switch plans during the open enrollment period.^6 ,7  As a result, many Part D enrollees leave money on the table—sometimes thousands of dollars.

Appendix 1: Methods

Data collection, plans, and pharmacies

All data were collected from the Medicare Plan Finder, using zip code 21201 in Baltimore, MD, mostly between October 19 and November 13, 2015. Data were collected for 20 PDPs, all of which are offered by 10 firms that sponsor plans on a national or near-national basis. The one other PDP in the selected zip code was Magellan Rx Medicare Basic PDP, which is offered in 20 of 34 regions and is not considered a near-national plan. Medicare Advantage drug plans were excluded from the analysis because plan participation varies geographically and few plans are offered on a national or near-national basis.

We are able to use one zip code as a reasonable representation of PDP costs nationally because nearly 90 percent of PDPs are offered on a national or near-national basis. These PDPs use the same formulary and the same formulary tier structure in all regions. Some of the national and near-national PDPs have modest variations in cost-sharing amounts from region to region, and there may be modest variations in the full price of drugs across pharmacies and across regions. Although total out-of-pocket costs will vary modestly across regions, the overall patterns reported here should not vary by geography.

Data were collected for two different pharmacies in Baltimore: (1) Rite Aid pharmacy on Lexington Avenue, selected to represent a pharmacy with standard cost sharing for all 20 PDPs, and (2) Community Pharmacy, a Walgreens pharmacy on Howard Street, selected to represent a pharmacy with preferred cost sharing for 13 of 20 PDPs. Three of the other seven PDPs (SilverScript Choice, WellCare Classic, and WellCare Extra) do not use tiered pharmacy networks. The remaining four PDPs (Humana Preferred, Humana Walmart, Symphonix Prime Saver Rx, and Symphonix Value Rx) had no pharmacies with preferred cost sharing within the default mileage distance used on the Medicare Plan Finder for the selected zip code. Data were also collected for mail order, as shown on the Medicare Plan Finder. A selection of data were also collected from two community pharmacies (Best Care and Mt. Vernon pharmacies) to verify that prices were relatively similar across pharmacies.

Drug selection

A set of ten brand-name and ten generic drugs were selected using the Medicare Provider Utilization and Payment Data: Part D Prescriber Public Use File (PUF) for 2013, released by CMS in April 2015. Drugs were sorted by numbers of claims. We used the ten generic drugs with the most claims for 2013. Similarly, we used the ten brand drugs with the most claims for 2013, but including only those brand drugs that have not acquired generic competition since that date.⁸  None of the excluded brands had quantities that would have qualified them for the list of most-used generics. Because of changes in drug utilization and availability, the actual top drugs in 2015 may vary somewhat.

Specialty drugs were selected for four health conditions that are commonly treated by specialty drugs: hepatitis C, rheumatoid arthritis, multiple sclerosis, and cancer. For the latter three conditions, the specific drugs were selected based on usage levels in the 2013 Part D Prescriber PUF. Because key drugs to treat hepatitis C were not approved for the market in 2013, we selected the three drugs that are recognized as significant therapeutic options in 2015. Many cancer drugs and some treatments for other conditions are covered under Part B, rather than Part D; thus, our selection of drugs includes those that are commonly paid for under Part D.

For all drugs, the dosage and the form of drug and the quantity of the medication used per month were taken from the defaults offered by the Medicare Plan Finder.

Data elements on drug costs collected for each drug

For each drug, we collected information on the full cost (price), cost-sharing amounts paid by enrollees, tier placement, and utilization management restrictions. For specialty drugs, cost information is presented on an annual basis because the 12 studied drugs are priced high enough that out-of-pocket costs are determined based on all benefit phases. For brands and generics, cost information is presented on a monthly basis because prices are low enough that beneficiaries do not routinely reach the gap or catastrophic phases based on taking any one drug.

The full cost of the drug is shown on the Medicare Plan Finder. The amount shown if the drug is on formulary is based on the drug’s unit price and dispensing fee as submitted by the plan. For off-formulary drugs, prices are inserted by CMS using a standard formula to approximate cash pricing: the wholesale acquisition cost (WAC) plus 15 percent for brands and WAC plus 20 percent for generics. The WAC is a publicly available list price that approximates what retail pharmacies pay wholesalers for single source drugs and is taken by CMS from the Medispan database, with First Data Bank as a backup. When we characterize a typical drug price in this report, it is based on the median price for plans that include that drug on formulary. For the drugs in this analysis, the CMS-supplied price tends to be about 10 percent to 20 percent higher than the median price for plans with the drug on formulary. But for some drugs it is considerably higher, probably reflecting lags in the data used to update the WAC.

Cost sharing is shown on the Medicare Plan Finder for four phases of the Part D drug benefit. Cost sharing in the deductible phase (where applicable) is equal to the full cost of the drug. Cost sharing in the initial coverage phase is determined based on the tier placement and cost-sharing structure for the particular plan. Cost sharing in the coverage gap phase is based on a statutory formula that takes into account the statutory manufacturer’s discount for most brand drugs and a required coinsurance amount. Cost sharing in the catastrophic phase is based on a statutory rule: the greater of 5 percent of the full cost of the drug or a nominal copayment amount.

Tier placement is also shown on the Medicare Plan Finder for each drug, as is the use of utilization management restrictions, including prior authorization, step therapy, and quantity limits.

Profiles of hypothetical beneficiaries

Five drug profiles were created for this study to reflect different utilization patterns that combine brand and generic drugs at different levels of drug costs. We included only commonly used drugs in the profiles and assembled drug combinations that reflect commonly co-occurring health conditions. No specialty drugs were included in the profiles because most Part D enrollees do not use any specialty drugs and because those drugs were analyzed separately.

Total out-of-pocket costs for these hypothetical beneficiaries were collected from the Medicare Plan Finder as described above for the specialty, brand, and generic drugs. Total costs include plan premiums, deductibles where applicable, and the total cost-sharing amounts owed. They also include the full cost of any off-formulary drugs, based on the CMS-supplied full cost for that drug.

Appendix 2: Data Tables

1. The statutory limit on out-of-pocket spending for 2016 is $4,850. This amount is substantially higher than the out-of-pocket spending shown before the catastrophic threshold (generally about $3,000) in Figure 1. Under the provision in the Affordable Care Act that phases out the coverage gap, manufacturers are required to discount the cost of certain drugs (mostly brand drugs) by 50 percent. The amount of this discount counts toward achieving the $4,850 catastrophic threshold. ↩︎
2. In general, specialty tier rates are limited to 25 percent by CMS guidance. But plans that reduce or eliminate the deductible can raise the specialty tier coinsurance rate as high as 33 percent. ↩︎
3. Jack Hoadley, Juliette Cubanski, and Tricia Neuman, Medicare Part D at Ten Years: The 2015 Marketplace and Key Trends, 2006-2015, Kaiser Family Foundation, October 2015, available at https://modern.kff.org/medicare/report/medicare-part-d-at-ten-years-the-2015-marketplace-and-key-trends-2006-2015/. ↩︎
4. Centers for Medicare & Medicaid Services, Medicare Part D Specialty Tier, April 7, 2015, available at https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovGenIn/Downloads/CY-2016-Specialty-Tier-Methodology.pdf. ↩︎
5. Gretchen Jacobson, Christina Swoope, Tricia Neuman, and Karen Smith, Income and Assets of Medicare Beneficiaries, 2014-2030. Kaiser Family Foundation, September 2015, available at https://modern.kff.org/medicare/issue-brief/income-and-assets-of-medicare-beneficiaries-2014-2030/. ↩︎
6. Jack Hoadley, Elizabeth Hargrave, Laura Summer, Juliette Cubanski, and Tricia Neuman, To Switch or Not to Switch: Are Medicare Beneficiaries Switching Drug Plans To Save Money? Kaiser Family Foundation, October 2013, available at https://modern.kff.org/medicare/issue-brief/to-switch-or-not-to-switch-are-medicare-beneficiaries-switching-drug-plans-to-save-money/. ↩︎
7. Gretchen Jacobson, Christina Swoope, Michael Perry, and Mary C. Slosar, How are Seniors Choosing and Changing Health Insurance Plans? Kaiser Family Foundation, May 2014, available at https://modern.kff.org/medicare/report/how-are-seniors-choosing-and-changing-health-insurance-plans/. ↩︎
8. One of the brands on our list, Synthroid, has generic alternatives (levothyroxine), available long before 2013. But this drug is commonly prescribed as a brand because many clinicians believe that patients whose dosage has been established on the brand version are not readily switched to the generic version without further titration of the dosage. ↩︎

States have long used Medicaid funds as premium assistance to purchase private health insurance for beneficiaries as an alternative to providing coverage directly through the state Medicaid program. States using premium assistance generally must provide wrap-around benefits and cost-sharing protections so that Medicaid beneficiaries receiving private coverage will not have access to fewer benefits or pay higher out-of-pocket costs when private coverage fails to meet Medicaid’s level of coverage or is more expensive. Some states seeking alternative ways to implement the Affordable Care Act’s (ACA) Medicaid expansion have been interested in expanding Medicaid premium assistance programs and adopting new models to purchase individual market coverage. These initiatives can be informed by an understanding of how pre-ACA premium assistance programs are working, particularly regarding wrap-around benefits and cost-sharing protections.

This issue brief examines states’ approaches to administering wrap-around benefits and cost-sharing in long-standing Medicaid premium assistance programs and the information available to beneficiaries about how to access these program features. We present findings based on our survey in 2014 of eight states (AL, LA, NV, RI, TX, UT, VT, and WI) that previously had reported spending dedicated to premium assistance wrap-around benefits to collect updated data and ascertain what states considering premium assistance in the Medicaid expansion context could learn about wrap-around benefits and cost-sharing based on long-standing premium assistance programs. We supplement our analysis by examining these states’ written materials designed to inform beneficiaries about premium assistance coverage and end with a discussion of post-ACA premium assistance programs.

Little data are available about spending on wrap-around benefits in state premium assistance programs.  Key findings from the 8 states include the following:

• States serve relatively small numbers of people in premium assistance programs that pre-date the ACA’s Medicaid expansion, and to the extent that limited data are available, spending on wrap-around benefits as a percent of total premium assistance program spending varies considerably among states.
• Few states report how much is spent on wrap-around benefits and cost-sharing protections in their premium assistance programs which makes it difficult to assess the extent to which beneficiaries are accessing those benefits and whether premium assistance programs are cost-effective.
• The clarity of states’ written materials explaining how beneficiaries can access wrap-around benefits varies.
• States’ written materials do not always clearly convey the availability of wrap-around EPSDT services for children.
• States have different policies for administering wrap-around cost-sharing protections, with most protecting beneficiaries from paying excess cost-sharing upfront. However, wrap-around cost sharing protections in examined states are available only if beneficiaries receive services from a provider who is both in their private insurance plan network and also accepts Medicaid. Beneficiaries may not be aware of this limitation, and it may further restrict provider options rather than expanding them.

As states’ interest in Medicaid premium assistance models continues, further research is needed to examine the beneficiary experience in these programs. This is especially true for access to wrap-around benefits and cost sharing protections, as little data presently is available in this area. Premium assistance approaches could have political and practical advantages for enrollees, and implementation of new programs in the context of alternative Medicaid expansions, can be informed by states’ experiences with long-standing programs. At the same time, wrap-around benefits add a layer of complexity for beneficiaries, providers, and states in premium assistance programs. Explaining how the wrap-around works is a challenge, especially when broad benefits that may not be fully included in private coverage, like EPSDT, are involved. Given the complex nature of these programs and variation in state implementation, the extent to which enrollees have access to the full Medicaid benefit package and cost-sharing protections is an area for continued study.

In addition, to the extent that Medicaid beneficiaries enrolled in premium assistance programs do not have access to wrap-around cost sharing protections unless they see Medicaid providers, beneficiaries are not receiving one of the key advantages often cited by proponents of premium assistance – access to a wider network of providers. Educational materials provided to beneficiaries do not always clearly explain the availability of wrap-around benefits and cost sharing protections, particularly EPSDT, and varied considerably in the states we examined. As more states consider premium assistance models to serve greater numbers of beneficiaries, the need to understand how to best administer wrap-around benefits and cost-sharing protections is of growing importance.

For many years, states have used Medicaid funding to purchase private health insurance for Medicaid beneficiaries as an alternative to providing coverage directly through the state Medicaid program. This approach, known as premium assistance, typically has been used to help people eligible for Medicaid afford the premiums for employer-sponsored insurance and has been a relatively small component of state Medicaid program enrollment prior to the Affordable Care Act (ACA). Since 2014, some states seeking alternative ways to implement the ACA’s Medicaid expansion have been interested in expanding Medicaid premium assistance programs to cover more beneficiaries and adopting new models to purchase coverage in the individual market.¹  These initiatives can be informed by an understanding of how pre-ACA premium assistance programs are working, particularly regarding so-called “wrap-around” benefits and cost-sharing protections, which states generally must provide to supplement the private coverage and make it comparable to Medicaid.

This issue brief examines states’ approaches to administering wrap-around benefits and cost-sharing in long-standing Medicaid premium assistance programs and the information available to beneficiaries about how to access these program features. We sought to collect updated data and ascertain what states considering premium assistance in the Medicaid expansion context could learn about wrap-around benefits and cost-sharing based on long-standing premium assistance programs. We present findings based on data from our survey in 2014 of eight states that had previously reported spending dedicated to premium assistance wrap-around benefits.  The states include Alabama, Louisiana, Nevada, Rhode Island, Texas, Utah, Vermont, and Wisconsin. We identified these states because they had reported such spending in 2009 in a Government Accountability Office (GAO) report.²  Of the 45 premium assistance programs in 37 states in the GAO report, nine states reported spending for wrap-around benefits.  Of these, we excluded Virginia because its premium assistance program is limited to CHIP beneficiaries and only provides wrap-around benefits for immunizations.  Since our survey, Vermont has discontinued its program, and Louisiana recently announced that it intends to discontinue its program as of December 1, 2015.

We supplement our analysis by examining these states’ written materials designed to inform beneficiaries about premium assistance coverage.  We asked state officials to share these materials and accessed others online at the states’ websites.  We did not receive written beneficiary materials from Vermont, and were unable to access materials online as its program was discontinued in 2014; we did receive a limited response to our survey on some of the spending questions.  The brief ends with a discussion of post-ACA premium assistance programs.

Background

Medicaid Premium Assistance Options

State options to use premium assistance to purchase private coverage for Medicaid beneficiaries predate the ACA.³  The most common form of Medicaid premium assistance is the use of Section 1906 authority to purchase private group insurance with Medicaid dollars, authorized in 1990.⁴  These programs are commonly referred to as Health Insurance Premium Payment (HIPP) programs. If premium assistance for group coverage is deemed cost-effective by the state, relative to the cost of providing Medicaid coverage directly, beneficiaries may be required to enroll in Section 1906 premium assistance programs. While these programs can be implemented through a state plan amendment, some states operate premium assistance programs with similar features through Section 1115 demonstration authority. Because relatively few Medicaid beneficiaries have access to employer-sponsored or other private group health insurance, these premium assistance programs have remained small.⁵ 

In addition to subsidizing group coverage, the Centers for Medicare and Medicaid Services (CMS) has used Section 1905(a) to authorize Medicaid premium assistance for individual market coverage.⁶  Prior to the ACA, Medicaid beneficiaries’ access to individual market coverage was limited because the cost was often unaffordable, and insurers could deny coverage based on pre-existing medical conditions. The creation of the Marketplaces under the ACA has generated new interest among some states in using Medicaid as premium assistance for individual market coverage, and in 2013, CMS released regulations and guidance for such programs.⁷  Enrollment in Section 1905(a) premium assistance programs is voluntary, unless the state obtains Section 1115 demonstration authority from CMS to make enrollment mandatory; in such cases, states must offer beneficiaries a choice of at least two health plans.⁸ 

Wrap-Around Benefits and Cost-Sharing in Medicaid Premium Assistance

Medicaid serves people with low incomes and who often have greater health care needs relative to other populations. In light of these characteristics, federal Medicaid law contains minimum benefit standards and maximum cost-sharing limitations. Private insurance typically offers fewer benefits than Medicaid does.⁹  Certain Medicaid services, most notably non-emergency medical transportation, usually are not covered by private insurance, and some private plans may have more restrictive limits on prescription drugs and other services, such as physical therapy, than are available to adults under Medicaid. In addition, private coverage for children is almost certainly less extensive than Medicaid as Medicaid’s Early Periodic Screening Diagnosis and Treatment (EPSDT) benefit requires states to cover any services “necessary. . . to correct or ameliorate. . . physical and mental illnesses or conditions. . . .” ¹⁰  A service that is not covered by private insurance is likely unattainable for people with low-incomes if it is not available through the Medicaid benefit package and therefore must be paid out-of-pocket. Medicaid also limits beneficiaries’ cost-sharing obligations to nominal amounts for adults with income below the federal poverty level and generally prohibits cost-sharing for children,¹¹  while private insurance is likely to have cost-sharing obligations in excess of Medicaid limits. For people with low incomes, a body of research has established that cost-sharing creates a barrier to accessing needed services.¹² 

Given the characteristics of the Medicaid-eligible population, federal law generally requires states using premium assistance to provide wrap-around benefits and cost-sharing protections so that Medicaid beneficiaries receiving private coverage will not have access to fewer benefits or pay higher out-of-pocket costs when private coverage fails to meet Medicaid’s level of coverage or is more expensive.¹³  States need waiver authority from CMS to limit wrap-around benefits and cost-sharing protections. A small number of pre-ACA premium assistance programs operate under such waivers; these programs may have expanded coverage to populations who were otherwise ineligible for Medicaid and include fewer benefits and/or higher cost-sharing. As of 2014, the ACA provides authority for states to cover nearly all adults without a waiver as full Medicaid beneficiaries with enhanced federal matching funds.

Consistent with the ACA’s coverage expansion, CMS’s 2013 regulations governing individual market premium assistance specify that beneficiaries must be provided with the same benefits and cost sharing protections that they would have had under traditional Medicaid.¹⁴  The preamble to the regulations confirms that “[u]nder all premium assistance arrangements, Medicaid and CHIP-eligible individuals remain Medicaid or CHIP beneficiaries and continue to be entitled to all Medicaid/CHIP benefits and cost sharing protections.”¹⁵  Additional guidance issued in light of state interest in alternative Medicaid expansions using Marketplace premium assistance requires states to provide wrap-around benefits and cost sharing protections to “ensure that coverage is seamless [and] that cost-sharing reductions are effectively delivered.”¹⁶  CMS’s regulations and guidance do not prescribe the specific methods that states must use but rather confirm that “states have the flexibility to determine how best to meet these cost-sharing and benefit responsibilities.”¹⁷ 

The costs of providing wrap-around benefits and cost-sharing protections must be included when determining whether Medicaid premium assistance programs are cost-effective.¹⁸  States generally cannot use premium assistance unless the cost is comparable to the cost of providing direct coverage through state’s Medicaid program. In addition to the cost of wrap-around benefits and cost-sharing protections, cost-effectiveness determinations also must include the cost of administering premium assistance programs.¹⁹ 

GAO Report on State Premium Assistance Programs

The GAO identified 47 Medicaid and CHIP premium assistance programs in 39 states in 2009, and received survey responses from 45 programs in 37 states.²⁰  Overall, enrollment in premium assistance programs is relatively small. Among the states reporting spending dedicated to wrap-around benefits, 2009 enrollment ranged from six beneficiaries in Alabama to nearly 8,700 in Texas. The GAO found that at least eight states target their premium assistance programs to people with high health care costs such as pregnant women, premature or low birth-weight infants, or people with HIV/AIDS, diabetes, or cancer.²¹ 

The GAO noted that “a reported issue with premium assistance programs is that there may be disparities in the benefits and cost-sharing protections offered to enrollees in such programs compared with those in direct coverage.”²²  The GAO also identified some potential advantages of premium assistance programs, such as helping families transition to private coverage, expanding coverage to family members who are ineligible for Medicaid or CHIP, and supporting the private insurance market. It also notes that while premium assistance programs could generate cost savings by leveraging employer contributions, these programs may be more expensive than direct coverage through states’ Medicaid and CHIP programs.²³ 

Few premium assistance programs reported spending dedicated to wrap-around benefits or cost-sharing protections in 2009. Nine of the 36 programs that provided at least some wrap-around benefits reported the dollar amount spent for those benefits.²⁴  Four of the 34 programs that paid at least some cost-sharing (including copayments, coinsurance, and deductibles²⁵ ) reported the dollar amount spent for wrap-around cost-sharing.²⁶  Thus, limited data are available to assess state spending on wrap-around benefits and cost-sharing protections in Medicaid premium assistance programs.

Key Findings About Wrap-Around Benefits and Cost-Sharing

Premium Assistance Program Enrollment

States serve relatively small numbers of people in premium assistance programs that pre-date the ACA’s Medicaid expansion. Enrollment in the premium assistance programs in states we examined ranges from 93 people in Nevada in 2014 to over 26,000 people in Texas in 2012 (Table 1). Despite wide state-level variation, enrollment in state premium assistance programs was a very small proportion of total Medicaid enrollment across all of the states. Enrollment in premium assistance accounted for 5% of total enrollment in Rhode Island and Vermont, and was less than 1% in all other states examined by this study.²⁷  This relatively low enrollment in premium assistance programs compared with total Medicaid enrollment is consistent with earlier research on premium assistance programs in Medicaid.²⁸ 

Some states target high cost/high need populations for their premium assistance programs. Similar to the GAO report, our survey found that some premium assistance programs (Alabama and Louisiana) specifically reach out to pregnant women, while others (Nevada) target people with conditions that incur high medical costs such as AIDS and cerebral palsy. Targeting populations with high needs may make sense for states seeking to ensure that their premium assistance programs are cost-effective. However, given their more extensive health care needs, this population may be even more likely to need access to wrap-around coverage for services that are available through Medicaid but not through private insurance.

Premium Assistance Program Spending

Few states report how much is spent on wrap-around benefits in their premium assistance programs, and to the extent that limited data are available, spending on wrap-around benefits as a percentage of total premium assistance program spending varies considerably among states.  Of the eight states we studied, four (Alabama, Louisiana, Rhode Island, Utah) reported spending for wrap-around benefits in 2014, and these data demonstrate variation in this spending among states (Table 1). The other states that responded to the survey noted their inability to break out costs for wrap-around benefits and were only able to report total spending for their premium assistance program.

The lack of data on spending devoted to wrap-around benefits in Medicaid premium assistance programs makes it difficult to assess the extent to which beneficiaries are accessing those benefits. Better data in this area would improve the ability of states and other stakeholders to determine whether beneficiaries are receiving wrap-around services. The availability of spending data broken down into categories, such as wrap-around benefits and wrap-around cost sharing protections, also could improve program monitoring and evaluation efforts. As one example, Utah is able to report the amount that it spends on the single wrap-around benefit that it offers in its premium assistance program, dental services for children (see Table 1 and Box 1).

The lack of data on spending devoted to wrap-around benefits in Medicaid premium assistance programs also makes it difficult to assess whether these programs are cost-effective. Only two states (Louisiana, Texas) were able to break out spending for the administrative costs for their premium assistance programs in their responses to our survey. States are required to evaluate cost-effectiveness as part of determining eligibility for premium assistance, and this determination must include spending on wrap-around benefits and cost-sharing protections as well as administrative costs. Better data about spending devoted to the various components of premium assistance programs would allow for a more accurate assessment of cost-effectiveness.

Overall state spending on premium assistance programs varies considerably by state. Per enrollee spending ranged considerably in 2014, from a low of $1,337 in Rhode Island to a high of $5,388 in Nevada (Table 1). This is likely a result to some degree of the fact that states serve different populations in their programs as well as relatively low program enrollment, which could result in wide variation in per capita costs; however, it also may reflect state challenges in tracking premium assistance costs.

Access to Wrap-around Benefits

Of the seven states we examined,³⁰  the clarity of the written materials provided to beneficiaries about how to access wrap-around benefits varied, and no state clearly conveyed the availability of EPSDT wrap-around services for children. This was despite the fact that all states targeted families with children in their premium assistance program enrollment materials. Federal regulations require states to use clear non-technical language to inform Medicaid-eligible children and their families about EPSDT, including what services are available and where and how to obtain those services.³¹  EPSDT is a broad benefit that includes regular screenings; vision, dental, and hearing services; and any treatment services necessary to correct or ameliorate physical or mental health conditions, regardless of whether such services are covered under the state’s adult benefit package.³² 

Rhode Island’s materials most clearly conveyed the availability of wrap-around benefits and described some of the benefits that EPSDT provides such as dental and vision services. Rhode Island’s premium assistance program booklet for beneficiaries explains that enrollees receive two cards – a private health plan card and a Medical Assistance Card. The booklet goes on to explain that “[t]he Medical Assistance Card is used for a few extra covered benefits listed in this booklet,” and provides details on eye care, dental services, bus passes, interpreter services, additional services that have limits in the employer plan (such as physical, occupational, and speech therapy; and mental health and substance abuse services), and over-the-counter medicine.³³  Other examined states did not provide any explicit or more generalized description (similar to Rhode Island’s) of the kinds of services that children are guaranteed through the EPSDT benefit and should be able to access even when enrolled in employer-sponsored insurance.

Two states we examined, Texas and Alabama, did refer to beneficiaries’ ability to receive Medicaid covered services on a wrap-around basis, but without any explanation of what those might be. Texas’ online description of its premium assistance program states that “[o]f course, Medicaid will pay for services not covered by the employer-sponsored insurance, as long as they are Medicaid-covered services provided by a Medicaid provider.”³⁴  While Alabama’s frequently asked questions for its premium assistance program explain that “[o]nce you are enrolled in the AL HIPP program, you will have access to benefits from both programs,”³⁵  the program flyer that Alabama uses instead highlights that premium assistance includes “[c]overage of group health insurance AND Medicaid … including benefits Medicaid may not cover.”³⁶  Nevada’s beneficiary brochure states that “Medicaid covered services may be included if covered through the employer health insurance,”³⁷  which does not assure beneficiaries that Medicaid benefits outside of the employer-sponsored insurance will be covered (emphasis added).

It is worth noting that in addition to written materials, a number of states also mentioned toll free numbers or third party contractors who were available to answer questions about the program as a way of educating beneficiaries about how to access premium assistance benefits. Utah’s program materials are primarily designed to inform families about the larger Utah Premium Partnership program as dental services are the only wrap-around benefit available to children eligible for CHIP. The state does include a FAQ about dental coverage which clarifies that parents can receive $20 per month towards the cost of their employer’s dental coverage or enroll in the CHIP dental plan.³⁸ 

By contrast, some states’ written materials made the opposite point, underscoring that the private health plan may cover services that are not covered by Medicaid. Beneficiary materials in Alabama, Louisiana, and Nevada all highlight that private insurance may cover benefits that Medicaid does not cover. For example, Louisiana’s materials highlight that employer-sponsored health insurance “may cover services that Medicaid doesn’t cover” rather than the reverse (i.e., that Medicaid is likely to cover services that your employer’s plan does not).³⁹  Wisconsin enrollees receive a lengthy booklet about enrollment and benefits which describes the situations in which a family might be able to enroll in their employer-sponsored insurance but does not describe how premium assistance or the wrap-around benefits work in practice. However, the state indicated in its survey response that state staff calls and details the plan to the member at the time of enrollment.

Access to Wrap-Around Cost-Sharing Protections

States have different policies about how they administer wrap-around cost-sharing protections in their premium assistance programs, with most protecting beneficiaries from making upfront payments in excess of Medicaid limits. According to survey responses, in some states (Louisiana, Nevada, Texas), Medicaid beneficiaries receiving premium assistance do not pay cost-sharing or out-of-pocket charges directly. In Alabama and Wisconsin, beneficiaries are responsible only for paying out-of-pocket co-payments at Medicaid levels. By contrast, in Rhode Island, beneficiaries must pay the full cost-sharing required by the private plan upfront when receiving services and then are reimbursed monthly.

In six states, our survey found that wrap-around cost sharing protections are available only if beneficiaries receive services from a provider who is both in their private insurance plan network and also accepts Medicaid.⁴⁰  If the provider does not accept Medicaid, beneficiaries must pay the entire cost of the service out-of-pocket. Thus, a family who is enrolled in their employer-sponsored insurance (ESI) with the help of Medicaid premium assistance may have to pay cost-sharing that exceeds Medicaid limits to have access to the full range of providers included in the ESI network, if all of those providers do not also accept Medicaid. Some of the premium assistance programs we examined target high-cost populations who may need specialty services and are likely frequent users of health care services. These populations may benefit from access to a private insurance network that is broader than the Medicaid provider network, but they also are unlikely to be able to afford out-of-pocket cost-sharing in excess of Medicaid limits. It is likely that beneficiaries may find that the need to see a provider who accepts both their private coverage and Medicaid further restricts their provider options rather than expanding them.

Beneficiaries may be unaware that they must see a provider who accepts both their private insurance plan and Medicaid to receive wrap-around cost-sharing protections. While written beneficiary materials overall were more clear about how to access wrap-around cost-sharing protections than wrap-around benefits, they did not consistently inform beneficiaries that they needed to see a Medicaid-participating provider to receive wrap-around cost-sharing protections. Some states’ materials are explicit on this point. For example, Rhode Island’s materials clearly state this limitation by informing beneficiaries that “[i]f you go to a provider who does not accept Medical Assistance, you will be required to pay the co-payment.”⁴¹  In addition, Alabama’s materials state that “[q]ualified Medicaid recipients have most out-of-pocket expenses covered by Medicaid when a recipient elects to go to a Medicaid provider,”⁴²  and similarly, Texas’ materials state that “Medicaid pays the co-pays and deductibles when people with HIPP and Medicaid see a Medicaid doctor.”⁴³  By contrast, Louisiana’s beneficiary flyer does not mention that the state will only cover cost-sharing charges for providers who also accept Medicaid,⁴⁴  although it does point out that “[i]f you have both Medicaid and other health insurance coverage, you may get increased access to primary care doctors, specialty care doctors, and hospitals.”⁴⁵  Alabama’s HIPP guide and application brochure state that “[m]embers receive … access to a wider network of doctors through group insurance coverage” while the application mentions that “Medicaid recipients’ out-of-pocket medical costs will be paid by Medicaid if they receive treatment from a Medicaid provider.⁴⁶  Wisconsin provides a lengthy information packet to potential Medicaid beneficiaries, which includes only a small subsection on premium assistance and does not discuss limitations on accessing wrap-around cost-sharing.⁴⁷ 

Post-ACA Premium Assistance Programs

Since 2014, a small number of states have included one or more premium assistance components as part of a Section 1115 waiver using an alternative approach to implementing the ACA’s Medicaid expansion.⁴⁸  These programs differ from the premium assistance programs we examined in that they may cover a greater number of beneficiaries and often they use Medicaid premium assistance to purchase individual Marketplace coverage. Arkansas was the first state to adopt this model and requires all newly eligible Medicaid beneficiaries to enroll in Marketplace premium assistance.⁴⁹  New Hampshire will be implementing a similar approach in 2016.⁵⁰  Iowa initially required Marketplace premium assistance for newly eligible adults with income between 100-138% of the federal poverty level, but the program is now voluntary due to the loss of one of the two Marketplace plans that covered Medicaid beneficiaries.⁵¹ 

Some state Medicaid expansion waivers also include provisions for more traditional premium assistance programs for the purchase of employer-sponsored coverage; these are relatively small parts of the expansions approved in Iowa (which enrollment is required) and Indiana (voluntary enrollment).⁵²  Other states that have debated the Medicaid expansion also have considered gubernatorial or legislative proposals that include premium assistance models.⁵³  For example, Tennessee’s proposal included voluntary premium assistance for employer-sponsored coverage,⁵⁴  and Utah’s proposal required premium assistance for Marketplace or employer-sponsored coverage.⁵⁵  This trend may continue as states that have not yet expanded Medicaid may be more inclined to favor the use of program designs that emphasize private coverage or other features that require waiver authority.

CMS generally has not allowed states with Medicaid expansion waivers using premium assistance to waive benefits in lieu of providing wrap-around coverage. The one exception is limited permission to waive non-emergency medical transportation benefits in Iowa⁵⁶  and Indiana.⁵⁷  For all remaining benefits that are not covered in the private insurance package, states electing to use premium assistance must provide wrap-around coverage. For example, Arkansas is providing both non-emergency medical transportation and EPSDT services for 19- and 20-year-olds through its Medicaid fee-for-service program as a wrap-around benefit for newly eligible adults (see Box 2).

CMS also generally has required states using Medicaid premium assistance for their new adult expansions to impose cost-sharing within Medicaid limits. This means that states electing premium assistance must determine how to administer wrap-around cost-sharing protections. Like most of the pre-ACA premium assistance programs that we examined, Arkansas’ approach to wrap-around cost-sharing protections shields beneficiaries from having to front high out-of-pocket costs and be reimbursed later.⁵⁹  Because Arkansas is using a limited number of Marketplace plans for its premium assistance program, it may be administratively easier to implement this approach to wrap-around cost-sharing than in states that offer premium assistance for anyone with any employer-sponsored coverage.

Conclusion

As states’ interest in Medicaid premium assistance models continues, further research is needed to examine the beneficiary experience in these programs. This is especially true for access to wrap-around benefits and cost sharing protections, as little data presently is available in this area. Premium assistance approaches could have political and practical advantages for enrollees, and implementation of new programs in the context of alternative Medicaid expansions can be informed by states’ experiences with long-standing programs. At the same time, wrap-around benefits add a layer of complexity for beneficiaries, providers, and states in premium assistance programs. Explaining how the wrap-around works is a challenge, especially when broad benefits that may not be fully included in private coverage, like EPSDT, are involved. Given the complex nature of these programs and variation in state implementation, the extent to which enrollees have access to the full Medicaid benefit package and cost sharing protections is an area for continued study.

In addition, to the extent that Medicaid beneficiaries enrolled in premium assistance programs do not have access to wrap-around cost sharing protections unless they see Medicaid providers, beneficiaries are not receiving one of the key advantages often cited by proponents of premium assistance – access to a wider network of providers. Educational materials provided to beneficiaries do not always clearly explain the availability of wrap-around benefits and cost sharing protections, particularly EPSDT, and varied considerably in the states we examined. As more states consider premium assistance models to serve greater numbers of beneficiaries, the need to understand how to best administer wrap-around benefits and cost-sharing protections is of growing importance.

SURVEY FOR KAISER/GEORGETOWN REPORT ON WRAPPED BENEFITS IN PREMIUM ASSISTANCE PROGRAMS

Thank you very much for your willingness to answer a short survey on your premium assistance program. We are mindful of how valuable your time is and have tried to keep this survey brief.

The purpose of this study is to examine the provision of wraparound benefits in state Medicaid premium assistance programs. Wraparound benefits are additional benefits and/or lower cost-sharing provided by Medicaid but not typically covered in a beneficiary’s private insurance plan.

We are approaching you because in a 2010 study conducted by the Government Accountability Office (GAO) entitled Medicaid and CHIP: Enrollment, Benefits, Expenditures, and Other Characteristics of State Premium Assistance Programs, STATE reported MEDICAID OR PROGRAM NAME expenditure data on the costs of wraparound benefits for FY 2009. Your state reported XXXX.

We are interested in both STATE’s provision of benefits covered by Medicaid that are not covered by the beneficiaries’ private insurance plans as well as expenditures to reimburse additional cost-sharing charges (such as deductibles, copayments and coninsurance) when private insurance charges are higher than those permitted by Medicaid.

1. For the most recent fiscal year available, please share enrollment and overall program expenditures for your premium assistance program.
2. For the same fiscal year, please share expenditures on premium subsidies, “wrapped benefits” i.e. benefits not covered by the beneficiary’s private insurance, and cost-sharing charges paid by the state on the beneficiaries’ behalf. We would appreciate if you could be as specific as possible.
3. Please briefly describe for us how the wraparound works in practice –
   1. How do beneficiaries learn about and access benefits that are covered by Medicaid but not by their private insurance plan?
   2. How are providers reimbursed for additional benefits covered by the Medicaid program and on what fee schedule?
   3. How is cost-sharing tracked and reimbursed? Do beneficiaries pay out-of-pocket at the point of service, and if so, are copayments limited to the Medicaid amounts?
   4. Are there different cost-sharing reimbursement rules for providers that are participating Medicaid providers and those that are not?

4. Please share with us any program materials that you provide to beneficiaries to help them understand and access their wraparound benefits and cost-sharing protections. Are there other resources available to beneficiaries (such as call centers) if they have questions?
5. Please share with us any additional information and/or observations you have about issues associated with providing wraparound benefits and cost sharing protections.

1. See generally Kaiser Commission on Medicaid and the Uninsured, A Look at the Private Option in Arkansas (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Aug. 2015), https://modern.kff.org/medicaid/issue-brief/a-look-at-the-private-option-in-arkansas/; Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion Through Marketplace Premium Assistance (Sept. 2013), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-through-marketplace-premium-assistance/; Kaiser Commission on Medicaid and the Uninsured, Premium Assistance in Medicaid and CHIP: An Overview of Current Options and Implications of the Affordable Care Act (March 2013), https://modern.kff.org/medicaid/issue-brief/premium-assistance-in-medicaid-and-chip-an-overview-of-current-options-and-implications-of-the-affordable-care-act/. ↩︎
2. United States Government Accountability Office, Medicaid and CHIP: Enrollment, Benefits, Expenditures, and Other Characteristics of State Premium Assistance Programs (Washington, DC: United States Government Accountability Office, Jan. 19, 2010), http://www.gao.gov/new.items/d10258r.pdf. ↩︎
3. There also are statutory authorities for states to use premium assistance in CHIP programs. See Kaiser Commission on Medicaid and the Uninsured, Premium Assistance in Medicaid and CHIP: An Overview of Current Options and Implications of the Affordable Care Act (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, March 2013), https://modern.kff.org/medicaid/issue-brief/premium-assistance-in-medicaid-and-chip-an-overview-of-current-options-and-implications-of-the-affordable-care-act/. ↩︎
4. 42 U.S.C. § 1396e, https://www.law.cornell.edu/uscode/text/42/1396e. States also have the Section 1906A option to use premium assistance for Medicaid beneficiaries with access to employer-sponsored insurance with a minimum 40% employer contribution to premium costs. Section 1906A was limited to children and their parents when first effective in 2009, and extended to all Medicaid beneficiaries as of 2014. Unlike Section 1906 programs, beneficiaries cannot be required to enroll in premium assistance under Section 1906A. 42 U.S.C. § 1396e-1, as amended by ACA § § 2003 and 10203(b)(2)(B), http://www.ssa.gov/OP_Home/ssact/title19/1906A.htm#ft118. ↩︎
5. The top two challenges in implementing and operating premium assistance programs identified by states surveyed by the GAO in 2009 include the limited number of individuals with access to private health insurance and difficulty identifying those individuals. GAO at 11. ↩︎
6. 42 U.S.C. § 1396d(a) (medical assistance payments “may include. . . other insurance premiums for medical or any other type of remedial care or the cost thereof”), https://www.law.cornell.edu/uscode/text/42/1396d. Only six states reported using Section 1905(a) authority in 2009. GAO at 15-16, Table 3. ↩︎
7. 42 C.F.R. § 435.1015; CMS, Medicaid and the Affordable Care Act: Premium Assistance Frequently Asked Questions (March 2013), http://medicaid.gov/Federal-Policy-Guidance/Downloads/FAQ-03-29-13-Premium-Assistance.pdf. ↩︎
8. 42 C.F.R. § 435.1015(b); CMS, Medicaid and the Affordable Care Act: Premium Assistance Frequently Asked Questions at 2 (March 2013), http://medicaid.gov/Federal-Policy-Guidance/Downloads/FAQ-03-29-13-Premium-Assistance.pdf. ↩︎
9. See, e.g., Kaiser Commission on Medicaid and the Uninsured, Benefits and Cost-Sharing for Working People with Disabilities in Medicaid and the Marketplace (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Oct. 2014), https://modern.kff.org/medicaid/issue-brief/benefits-and-cost-sharing-for-working-people-with-disabilities-in-medicaid-and-the-marketplace/. ↩︎
10. 42 U.S.C. § 1396d(r)(5), https://www.law.cornell.edu/uscode/text/42/1396d. ↩︎
11. 42 U.S.C. § 1396o, https://www.law.cornell.edu/uscode/text/42/1396o.   ↩︎
12. See, e.g., L. Dague, “The effect of Medicaid premiums on enrollment: A regression discontinuity approach,” Journal of Health Economics 37, (Sept. 2014): 1-12, http://www.sciencedirect.com/science/article/pii/S0167629614000642; Kaiser Commission on Medicaid and the Uninsured, Premiums and Cost-Sharing in Medicaid: A Review of Research Findings, (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Feb., 2013), https://modern.kff.org/medicaid/issue-brief/premiums-and-cost-sharing-in-medicaid-a-review-of-research-findings/. ↩︎
13. 42 C.F.R. § 435.1015(b); CMS, Medicaid and the Affordable Care Act: Premium Assistance Frequently Asked Questions at 2 (March 2013), http://medicaid.gov/Federal-Policy-Guidance/Downloads/FAQ-03-29-13-Premium-Assistance.pdf. ↩︎
14. 42 CFR § 435.1015(a)(2) and (3) (making premium assistance subject to the conditions that, inter alia, “[t]he agency furnishes all benefits for which the individual is covered under the State plan that are not available through the individual health plan” and that “[t]he individual does not incur any cost-sharing charges in excess of any amounts imposed by the agency under subpart A of part 447”). ↩︎
15. 78 Fed. Reg.42159, 42184 (July 15, 2013), https://www.federalregister.gov/articles/2013/07/15/2013-16271/medicaid-and-childrens-health-insurance-programs-essential-health-benefits-in-alternative-benefit . ↩︎
16. CMS, Medicaid and the Affordable Care Act: Premium Assistance Frequently Asked Questions at 2 (March 2013), http://medicaid.gov/Federal-Policy-Guidance/Downloads/FAQ-03-29-13-Premium-Assistance.pdf. ↩︎
17. 78 Fed. Reg. 42184. ↩︎
18. 42 C.F.R. § 435.1015(a)(4). ↩︎
19. 42 C.F.R. § 435.1015(a)(4). ↩︎
20. GAO at 2-3. Not all programs responded to all survey questions. All 45 programs offered premium assistance for group coverage, and 21 offered premium assistance for individual market coverage. GAO at 7. Less than half (20 programs) required beneficiaries to enroll in premium assistance. GAO at 8. ↩︎
21. GAO at 8. ↩︎
22. GAO at 2. ↩︎
23. GAO at 2. ↩︎
24. GAO at 9, 10, n.22. An additional three states reported $0.00 in spending. ↩︎
25. GAO at 6. ↩︎
26. GAO at 10, n.22. An additional four states reported $0.00 in spending. ↩︎
27. Kaiser Family Foundation, Total Monthly Medicaid and CHIP Enrollment (Washington DC: Kaiser Family Foundation, Aug. 2015), https://modern.kff.org/health-reform/state-indicator/total-monthly-medicaid-and-chip-enrollment/; Kaiser Family Foundation, Total Monthly Medicaid Enrollment (Washington DC: Kaiser Family Foundation, Dec. 2013), https://modern.kff.org/medicaid/state-indicator/monthly-medicaid-enrollment-in-thousands/. ↩︎
28. Kaiser Commission on Medicaid and the Uninsured, Serving Low-income Families Through Premium Assistance: A Look at Recent State Activity (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Oct. 2003), https://modern.kff.org/medicaid/issue-brief/serving-low-income-families-through-premium-assistance/. ↩︎
29. For more information, see Utah Dep’t of Health, UPP for Health insurance Frequently Asked Questions, http://health.utah.gov/upp/faqs.htm#15; Utah Dep’t of Health, UPP for Health Insurance Comparison Chart, http://health.utah.gov/upp/PDF/Comparison%20Chart2014%20-%20eng%20.pdf. ↩︎
30. We did not receive written materials from Vermont, although we did receive a limited response to our survey on some of the spending questions. We were not able to analyze Vermont’s materials as it discontinued its program in 2014. ↩︎
31. 42 C.F.R. § 441.56(a). ↩︎
32. 42 U.S.C. § 1396d(r)(5). ↩︎
33. Rhode Island Department of Human Services, Rite Share Health Insurance Premium Assistance Program Brochure (Feb. 2011), http://www.eohhs.ri.gov/Portals/0/Uploads/Documents/rs_booklet_eng.pdf. ↩︎
34. Texas Medicaid and CHIP Health Insurance Premium Payment program web description, http://www.hhsc.state.tx.us/Medicaid/hipp/. ↩︎
35. Alabama Health Insurance Premium Payment Program FAQs, http://myalhipp.com/faqs/. ↩︎
36. Alabama Health Insurance Premium Payment Program Brochure, http://myalhipp.com/faqs/. ↩︎
37. State of Nevada Division of Health Care Financing and Policy Health Insurance Premium Payment Program Brochure, dhcfp.nv.gov/uploadedFiles/dhcfpnvgov/content/Pgms/CPT/HIPPEnglishBrochure.pdf. ↩︎
38. See “Frequently Asked Questions” Utah Department of Health Utah Premium Partnership website, health.utah.gov/upp/faqs.htm. ↩︎
39. Louisiana HIPP application and flyer on file with authors. ↩︎
40. In Utah where families only receive a monthly contribution towards the cost of dental coverage cost-sharing is not addressed at all. ↩︎
41. Rhode Island Department of Human Services, Rite Share Health Insurance Premium Assistance Program Brochure (February 2011), http://www.eohhs.ri.gov/Portals/0/Uploads/Documents/rs_booklet_eng.pdf. ↩︎
42. Alabama Health Insurance Premium Payment Program, Fact Sheet, http://www.myalhipp.com/wp-content/uploads/2012/02/Fact-Sheet.pdf . ↩︎
43. Texas Health and Human Services Commission, “Your Guide to HIPP,” (on file with authors). ↩︎
44. The state’s response to our survey indicated that this is Louisiana’s policy on wrap-around cost-sharing. ↩︎
45. Louisiana Health Insurance Premium Payment Program, “Pregnant and on Medicaid,” http://new.dhh.louisiana.gov/assets/medicaid/lahipp/docs/LaHIPP_pregnancy_onepager.pdf. ↩︎
46. Alabama Health Insurance Premium Payment Program Brochure and website materials available as noted above. ↩︎
47. Wisconsin’s member booklet “ForwardHealth: Your Connection to Health Care Coverage and Nutrition Benefits” https://www.dhs.wisconsin.gov/publications/p0/p00079.pdf. Also see a state factsheet describing the program which does not clarify that providers must be Medicaid providers to be reimbursed, https://www.dhs.wisconsin.gov/publications/p1/p10095.pdf. ↩︎
48. See generally Kaiser Commission on Medicaid and the Uninsured, The ACA and Medicaid Expansion Waivers (Feb. 2015), available at https://modern.kff.org/medicaid/issue-brief/the-aca-and-medicaid-expansion-waivers/. As of this writing, 31 states (including DC) have implemented the ACA’s Medicaid expansion to nearly all adults up to 138% of the federal poverty level (FPL, $16,243 for an individual in 2015), and most states have done so as Congress envisioned in the ACA. Kaiser Family Foundation State Health Facts, Status of State Action on the Medicaid Expansion Decision. Data Source: KCMU tracking and analysis of state executive activity (Nov. 2, 2015), https://modern.kff.org/medicaid/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/. As a result of the Supreme Court decision in National Federation of Independent Business v. Sebelius, states may now choose whether they wish to expand Medicaid. Kaiser Family Foundation, A Guide to the Supreme Court’s Affordable Care Act Decision (Washington, DC: Kaiser Family Foundation, June 2012), https://modern.kff.org/health-reform/issue-brief/a-guide-to-the-supreme-courts-affordable/. ↩︎
49. Kaiser Commission on Medicaid and the Uninsured, A Look at the Private Option in Arkansas (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Aug. 2015), https://modern.kff.org/medicaid/issue-brief/a-look-at-the-private-option-in-arkansas/; Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion in Arkansas (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Feb. 2015), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-in-arkansas/. ↩︎
50. New Hampshire initially implemented the Medicaid expansion using its existing Medicaid delivery system. Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion in New Hampshire (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, March 2015), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-in-new-hampshire/. ↩︎
51. Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion in Iowa (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Feb. 2015), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-in-iowa/. ↩︎
52. Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion in Indiana (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Feb. 2015), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-in-indiana/. ↩︎
53. For example the proposal advanced by the Florida Senate (Florida Health Insurance Affordability Exchange Program) included a premium subsidy model that would potentially subsidize employer-sponsored insurance, www.ahealthyfloridaworks.com. ↩︎
54. Kaiser Commission on Medicaid and the Uninsured, Proposed Medicaid Expansion in Tennessee (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Jan. 2015), https://modern.kff.org/medicaid/fact-sheet/proposed-medicaid-expansion-in-tennessee/. ↩︎
55. Kaiser Commission on Medicaid and the Uninsured, Proposed Medicaid Expansion in Utah (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Jan. 2015), https://modern.kff.org/medicaid/fact-sheet/proposed-medicaid-expansion-in-utah/. ↩︎
56. Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion in Iowa (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Feb. 2015), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-in-iowa/. ↩︎
57. Kaiser Commission on Medicaid and the Uninsured, Medicaid Expansion in Indiana (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Feb. 2015), https://modern.kff.org/medicaid/fact-sheet/medicaid-expansion-in-indiana/. ↩︎
58. Data was obtained via personal communication with the Arkansas Department of Human Services, Division of Medical Services, Coordination of Coverage Unit, April 2014, on file with authors. ↩︎
59. Kaiser Commission on Medicaid and the Uninsured, A Look at the Private Option in Arkansas at 11 (Washington, DC: Kaiser Commission on Medicaid and the Uninsured, Aug. 2015), https://modern.kff.org/medicaid/issue-brief/a-look-at-the-private-option-in-arkansas/. ↩︎