Value-Based Insurance Design in Medicare Advantage Demonstration


CMMI announced a five-year Value-Based Insurance Design (VBID) model demonstration in Medicare Advantage (MA) that will begin on January 1, 2017. Under the demonstration, eligible MA plans may implement interventions falling within four categories (further details below). The model seeks to evaluate whether offering clinically nuanced VBID benefit designs in MA plans would lead to higher quality and more cost-effective care for the targeted enrollees. Organizationally, the model is open for participation to MA and MA-PD plans at the individual plan benefit package (PBP) level.

In support of the demonstration, CMS has indicated that the HHS Secretary intends (as authorized by Section 1115A of the Social Security Act regarding the testing of innovative health care payment and service delivery models) to grant a limited waiver of certain Medicare uniformity requirements to allow MA organizations to offer interventions to clinically targeted enrollee populations within a PBP rather than to their entire membership.

Test States

The following seven states were determined eligible to participate in the demonstration: Arizona, Indiana, Iowa, Massachusetts, Oregon, Pennsylvania, and Tennessee. These states were selected by CMS to be nationally representative of the MA population for the purposes of evaluation.

Clinical Categories

The demonstration limits the benefit design flexibility to enrollees with certain chronic conditions defined by CMS. Plans can alter benefit designs for one or more of the identified clinical categories, or for individuals with a combination of the conditions outlined. The clinical categories in the demonstration include: diabetes, congestive heart failure, chronic obstructive pulmonary disease (COPD), past stroke, hypertension, coronary artery disease, and mood disorders.

Eligibility Criteria for Plans

To be eligible to participate, MA plans must meet certain criteria and have their interventions approved by CMS. Qualified participants must be an HMO, HMO-POS, or local PPO and must not be consistently low performing (3-Star). Additionally, eligible plans must not be under sanction, have no past performance outlier rating, operate for 3 years prior to CY 2017, cover a minimum of 2,000 enrollees, and be offered in no more than two states with 50% of their enrollees in a test state. CMS has indicated there may be some flexibility in eligibility criteria and encourages plans to engage CMS as appropriate as part of the application process.

Eligible Intervention Types

While MA organizations have the flexibility to develop their own interventions for the target populations by PBP, benefit changes must be categorized as: (1) reduced cost sharing for high-value services; (2) reduced cost sharing for high-value providers; (3) reduced cost sharing for enrollees participating in disease management or related programs; or, (4) clinically targeted supplemental benefits.

1. Reduced cost sharing for high-value services. This could include elimination or reduction of co-pays, co-insurance, or exemption of a service from the plan deductible, among other options.

2. Reduced cost sharing for high-value providers. Flexibility around cost sharing can be applied when organizations identify providers (which may be drawn from across all Medicare provider types) as being “high-value.” The methodology for determining high-value providers as they relate to each clinical category must be proposed in detail in the application. Organizations have the flexibility to either adjust cost sharing for an identified high-value provider regardless of the service provided or to adjust cost sharing only when the identified high-value provider provides certain high-value services.

3. Reduced cost sharing for enrollees participating in disease management or related programs. While organizations have the flexibility to adjust cost sharing for targeted enrollees who participate in such programs, cost sharing cannot depend on enrollees achieving any clinical goals.

4. Clinically targeted supplemental benefits. Coverage of additional supplemental benefits can be made available to the targeted populations. However, organizations that propose supplemental benefit interventions must not discriminate within the target population and make these benefits available to every enrollee within the targeted clinical category.
It is important to note that enrollees (including those outside the target population) cannot receive fewer benefits or pay higher cost sharing than other enrollees because of the model. Additionally, any reductions in cost sharing must be available to all enrollees within the target population.

Application Process and Timeline

The deadline to respond to the Request for Applications is 4PM EST on January 8, 2016. Tentative timeline:

January 8, 2016: Model applications due to CMS
March 2016: Provisionally selected model participants identified
June 2016: 2017 MA plan bids due
September 2016: Contract addenda for model participation executed
January 2017: Model implementation begins

Actuarial Guidance

CMA released actuarial guidance to provide clarity on pricing considerations, as well as instructions for completing the Bid Pricing Tools (BPTs) for both Part C and Part D and other supporting documentation related to the application. The guidance discusses a number of important points.

Looking at savings broadly. Although there is a general rule that these demonstrations must result in net savings to Medicare and cannot increase costs to beneficiaries, the guidance suggests that savings may be looked at broadly over the 5-year course of the demonstration. Thus, even if a plan’s bid pricing tools would reflect a net increase of costs to enrollees and/or Medicare for CY 2017 (for either Part C or Part D), the applicant can provide a 5-year projection that demonstrates net savings over the five-year life of the model.

Demonstrating impact on each targeted population. The guidance suggests that applicants would provide estimates of expected impact on costs and utilization, both for the entire plan population as a whole, as well as for each of the targeted VBID populations. Similarly, if the applicant intends to use multiple interventions, the applicant should be prepared to quantify the impact of each intervention on a stand-alone basis.