WELShift Evaluation


FINANCING A NEW INTEGRATED CARE MODELThe Senior Healthcare & Housing Initiative for Transformation or WELShift * . . . Download “Financing a New Integrated Care Model” prepared by Avalere Health LLC




The United States spent more than $200 billion in 2006 on a Long-Term Care (LTC) system that provides inadequate financial protection for the elderly and relies heavily on government funding. Most individuals pay for LTC out of their personal savings until they are poor enough to qualify for Medicaid, which offers few choices other than institutionalization. In addition, the healthcare system inadequately manages the healthcare costs of people with LTC needs. As these patients move across multiple health and LTC settings, their care lacks the consistent coordination and management necessary for the efficient and effective use of healthcare resources. The impending growth of the elderly population will only exacerbate these problems and create significant federal and state budgetary pressures.


Policymakers and others have begun to develop proposals that would create incentives for better care coordination and provide individuals with new LTC financing options. Various health reform proposals have included policies such as post-acute and hospital bundling, which would increase provider accountability for care provided across multiple care settings, and the medical home model, which would pay providers directly for care coordination. Proposals have also included a new social insurance program to finance LTC that would be available to working-age Americans.


In order to contribute to these efforts and address the need for provider accountability and better care coordination, Wesley Enhanced Living (WEL), a nonprofit owner and operator of multiple residential LTC communities called “continuing care retirement communities” (CCRC) in Pennsylvania, has proposed the Senior Healthcare and Housing Initiative for Transformation (SHIFT), a demonstration program that would coordinate and integrate acute and LTC services in combination with housing.


The WELShift program would pool a wide range of funding sources, including home equity, Medicare, and Medicaid. In order to integrate health, LTC services, and housing across these sources of funding, an on-campus, multidisciplinary team of providers would provide care coordination and management for all enrollees, targeting services and supports in a manner that would reduce unnecessary healthcare use.


WEL designed WELShift to achieve the following goals through co-locating health, LTC and housing:

  • Reduce Healthcare Costs.  By offering intensive care coordination in the same location as resident homes and community activities, WEL intends to target healthcare services appropriately and reduce overall Medicare spending and healthcare use for its enrolled population.
  • Maintain Enrollee Independence and Avoid Institutionalization.  Through early intervention, WELShift will provide services necessary to support independent living for as long as possible.
  • Provide Affordable Long Term Care.  By combining multiple funding sources, WEL intends to make non-institutional LTC services available to individuals who might not otherwise be able to afford them.




Avalere found that the WELShift demonstration design would require both Medicaid and Medicare funding to operate. We found that Medicaid funding would not be available under current law, thus requiring a waiver or change in Medicaid law to receive the funding. Alternatively, we found that Medicare funding would be available under current law as long as WELShift becomes certified as a Medicare Advantage (MA) plan or partners with an existing plan.


Avalere estimated that, should WELShift secure Medicare and Medicaid funding, Medicaid would spend about 20 percent less on an initial cohort of 1,000 WELShift enrollees than it would for that population under the current Medicaid program. This would equate to a net savings from the beginning of the program through the end of the current 10-year budget window of $4.107 million.


Although WEL expects to achieve significant Medicare savings, we did not estimate these savings. WELShift’s approach to integrated care is new and thus no documented healthcare savings from other integrated care programs could support assumptions we might make about healthcare savings. Furthermore, since we determined that WELShift can operate without a change in Medicare payment rules or law, the program would not generate Medicare savings compared to current law.




The WELShift proposal would fully integrate housing, LTC services, and medical care for elderly participants in a CCRC setting. The initial WELShift demonstration sites (not yet constructed) would be in Pennsylvania. Residents would join the program by paying an entrance fee and subsequent monthly fees that would stay the same over time even if resident’s service needs increase. In exchange, enrollees would receive housing as well as coordinated LTC and medical care for the remainder of their lives.


The following are key features of the proposed program:

  • Expand the CCRC model to include fully integrated LTC and medical care.  WELShift would offer an onsite, intensive, interdisciplinary team approach to the coordination and management of acute and LTC services. Current CCRC models do not offer medical care as part of the community, nor do they integrate LTC and acute care services.
  • Combine available funding streams.  WELShift would accept payment from Medicare and residents. For residents who exhaust assets and whose income alone is insufficient to cover the monthly fee, WELShift would request a Medicaid payment, at an amount less than or equal to the monthly fee. As described above, WELShift would require a change in law in order for Medicaid to provide this supplement.
  • Allow individuals, regardless of income, to age in place in a “life care” CCRC.  “Life care” CCRCs serve elderly individuals who are relatively healthy when they enter the CCRC, which then provides other types of housing and LTC services, such as assisted living and nursing home care, as residents become frail and need more assistance. Care is provided for the same monthly fee, regardless of the amount of care needed (independent housing, assisted living, or nursing home). Because WELShift is combining funding streams and using Medicaid as a source of funding for residents whose financial resources are insufficient to cover the full costs of care, it can make this model available to individuals who would not otherwise be able to afford it.


Through the proposed structure, the WELShift demonstration will test whether its care management approach and the provision of appropriate LTC services can reduce both overall healthcare spending relative to the Medicare managed care payments it receives and Medicare fee-for-service spending.



Avalere Health developed two models to estimate the impact of WELShift on Medicaid spending. The first model projected what Medicaid would spend on individuals enrolled in the WELShift program (i.e., WELShift residents) and the second estimated what Medicaid would spend on those individuals in the absence of WELShift (referred to as the “baseline” scenario). We used each model to estimate the number of individuals who would receive Medicaid, their per capita Medicaid costs, and their overall Medicaid spending. We projected costs over the next 10 years, the time frame traditionally used in federal budgetary estimates.



We developed a model that enabled us to track enrollees by their income and assets as well as their expenses over a ten year period while enrolled in WELShift. When enrollees’ income and assets were no longer sufficient to cover the cost of living in WELShift’s CCRC, we identified them as Medicaid recipients and calculated the amount that Medicaid would spend to cover the difference between their financial resources and the program’s monthly fee. In this way, we estimated the total number of Medicaid-eligible enrollees each year and the total yearly amount that Medicaid would spend on those WELShift participants.


This estimation required making the following assumptions:

  • Demographic and Financial Profiles of the WELShift Population. We used the Health and Retirement Survey (HRS) to obtain the income and asset levels of the population eligible for WELShift. We identified this population by age, ability to afford the entrance fee, and lack of disabilities on entry.
  • Expenses in WELShift. We based our estimate of the WELShift entrance fee and monthly fees on the fees of comparable life care CCRCs in Pennsylvania. Under guidance from WEL, we inflated the fees to adjust for capital costs for the construction of new WELShift CCRC sites. We also assumed a modest monthly expense to cover additional living expenses based on WEL’s experience with their current CCRCs.


Using the above data points, we projected the spending patterns for individuals with different levels of income and assets. Enrollees would qualify for Medicaid if they spent all of their assets and did not have enough income to cover the ongoing costs of the WELShift CCRC, with Medicaid paying the difference between the two amounts. For each year, we calculated the total number of WELShift enrollees that would receive Medicaid and the total amount that Medicaid would spend on them.



Avalere developed a model that uses demographic data from the HRS and national survey data on rates of nursing home use by age to estimate how many WELShift enrollees would receive Medicaid LTC services absent the WELShift program. We then used data on Medicaid spending in Pennsylvania to estimate how much Medicaid would spend on those individuals for nursing home services, home- and community-based services, and other acute care services each year if they were not enrolled in WELShift.


This estimation required making the following calculations:

  • Calculating the percent of Pennsylvania elderly receiving Medicaid-covered nursing home stays. We used data on nursing home usage from the National Nursing Home Survey and Census Bureau projections of the elderly population in the state to estimate the number of elderly in Pennsylvania nursing homes and the total elderly population in Pennsylvania.
  • Calculating the percent of WELShift-eligible elderly receiving Medicaid-covered nursing home stays. Since enrollment in WELShift assumes that individuals are able to afford the entrance fee and have no disability on entry, the “WELShift-eligible” population would use Medicaid at a different rate than the elderly population at large. We used the HRS data to estimate the subset of the total Pennsylvania elderly population as well as the subset of the elderly receiving Medicaid-covered nursing home stays that would have been eligible for WELShift. This calculation allowed us to alter the percent of Pennsylvania elderly receiving Medicaid-covered nursing home stays to reflect the slightly lower rate of use for the WELShift-eligible population. We then applied this percent to the 1,000 individuals enrolled in WELShift, decreasing the total population over time to account for mortality.
  • Calculating Medicaid spending on nursing home stays for the WELShift-eligible population. We used data from the Medicaid Statistical Information System (MSIS) to calculate the amount that Medicaid spends historically per capita on nursing home care in Pennsylvania, decreasing this amount slightly due to the slightly higher average income of the WELShift population. We then multiplied this per capita Financing A New Integrated Care Model amount by the number of WELShift individuals in nursing homes to estimate Medicaid spending for nursing home services for the WELShift population.
  • Calculating Medicaid spending on other services including home- and community based services. We obtained data from the Office of Long Term Living in Pennsylvania on spending on home- and community-based services (HCBS) and data from MSIS on spending on acute care for the elderly population. We calculated the ratio of spending for HCBS and acute care to spending on nursing home services. We then multiplied this ratio by our estimate of Medicaid spending on nursing home care for the WELShift population under baseline. We also took into account that individuals eligible for WELShift are more likely to have sources of coverage other than Medicaid.



While we based our assumptions on data when possible, our analysis was limited by the available data sources and any changes in our model assumptions noted above may impact the results of our analysis. When solid evidence from research or other programmatic experience was lacking, we used our best analytic judgment about the likely behavior of individuals and providers in response to the proposal’s incentives.


Key limitations to our analysis include the following:

  • Enrollment - Our models required assumptions surrounding enrollment and demographics, which we based on available evidence and analytic judgment. However, since WELShift would enroll a relatively small population in Pennsylvania, the demographic and income and asset profiles of the actual WELShift population could vary from the population of the United States as a whole.
  • Economics and Policy - Both of the models incorporate data on personal assets, like home equity and stocks, which predate the recent financial crisis and recession. Many of those assets have obviously declined significantly in value and will recover their value slowly, if at all. While we would have preferred to account for this drop in asset values in our estimates, it would be extremely difficult in the model that we used for the baseline. We decided not to make an adjustment to either model in order to ensure an equal comparison between our estimates.
  • WELShift Costs - We made assumptions on the amounts of the monthly fee, entrance fee, and cost-of-living expenses based on data from a comparable life care CCRC in Pennsylvania and direction from WEL. We did not analyze the actuarial or operational viability of these program costs. To the extent that any of these costs WELShift due to operational necessities, this will impact the budgetary requirements of the WELShift demonstration. Further, little data exists to support assumptions surrounding whether out-of-pocket spending would vary in a retirement community by income and, if so, the degree it would vary. To the extent that individuals spend more than the estimated living expenses, they could become eligible for Medicaid payments under the demonstration sooner than we estimated.
  • Scope of Impact - Additionally, this study did not analyze the impact of WELShift on total Medicaid nursing home spending. Therefore, we did not estimate any impact from “bed backfill,” which can occur when nursing home bed supply is tight and beds become available.


* Prepared for: Wesley Enhanced Living - Prepared by: Avalere Health LLC  October 2009

WEL engaged Avalere Health to analyze the proposed demonstration program. Avalere analyzed whether the demonstration design would require a change in law and estimated the budgetary impact of the proposed program. Wesley Enhanced Living provided financial support for this research. Avalere maintained sole discretion with regards to methods and interpretation of findings and the authors are solely responsible for the content of this analysis. 

 . . . Download “Financing a New Integrated Care Model” prepared by Avalere Health LLC

HOME   |   ABOUT   |   NEWS   |   BLOG   |   LEGISLATION   |   CONTACT   |    SITEMAP                                                                                          Internet Marketing