Summary

The purpose of this analysis is to present potential savings from the introduction of an innovative treatment for Alzheimer’s Disease (AD) as applied to the Medicare population over a ten-year time horizon, 2022-2031. The years 2020 and 2021 serve as a baseline for our 2022-2031 estimates. In 2020, 5.8 million Americans aged 65 and older (10%) were diagnosed with AD[1]. The Alzheimer’s Disease Continuum consists of four stages: Mild Cognitive Impairment (MCI) due to AD, Mild AD, Moderate AD and Severe AD. Our study attributes savings to the population in the first two stages of this continuum. Our estimated cost savings assume the introduction of an AD medication that delays disease progression by six months for year 1, twelve months for year 2, eighteen months for year 3, and twenty-four months for all following years until 2031. We build upon a model of annual Medicare costs attributable to AD over the patient’s years before death[2]. We model six-month totals to better estimate potential savings. We assume a “take-up rate” that starts out at 25% in 2022 and increases to 50% of a carefully determined population by 2031. For Medicaid savings, we specifically focus on costs and savings generated from Nursing Home (NH) care for Medicare beneficiaries. Our informal caregiving cost estimates account for both a population that enrolls in nursing home care and a population that never enrolls in nursing homes Our total savings estimates for 10 years for each cost category’s expenditures are shown in Table ES-1. Potential Medicare savings are limited because we focus only on those costs associated with Medicare spending (including Beneficiary copayments) that are directly associated with AD or ~11% of the total Medicare expenditures for AD patients[3]. Greater savings are achieved through reductions in formal Medicaid long-term care expenditures and informal caregiving and support expenditures. A drug that delays AD disease progression by two years promises significant social returns relative to drugs that extend life for much shorter time frames. This study was commissioned by Biogen, Inc.

Background

In 2020, 5.8 million Americans aged 65 and older (10 percent) were diagnosed with Alzheimer’s Disease (AD)[1]. The formal expenditures for AD and other dementias, which include Medicare, Medicaid, and Medicare Out-of-Pocket Copay costs, for these patients were estimated in 2019 to be $305 billion[2]. Informal caregiving costs were also significant. Over 18.6 billion hours were spent in informal unpaid caregiving in 2019, equating to a nearly $244 billion national contribution[3]. This paper is limited to an analysis of savings to Medicare and Medicaid from reductions in expenditures for beneficiaries as well as reductions in total societal expenditures associated with the clinical introduction of a Potential AD treatment to Medicare program beneficiaries. To be consistent with a 10-year CBO-type Medicare savings estimate of the introduction of a Potential AD Treatment, we present 10-year timeframe estimates. We specifically focus on the Medicare beneficiary population. We do not account for the value of additional years of life, if any.

Findings

1.A potential AD treatment would have a likely target population of 1,283,970 Medicare beneficiaries included in the first year of analysis (2022) or ~22% of the AD population. The Mild AD subsection of the potential treatment population is greater than the MCI subsection of the population with 941,650 and 340,298 Medicare beneficiaries respectively in 2022 and 1,206,613 and 432,576 Medicare beneficiaries respectively in 2031. This potential treatment population is then stepped down by annual AD incidence rate and potential AD treatment take-up rate to the final population shown in Exhibit ES-2.

2. A potential AD treatment could result in an estimated $1.24 billion in 10-year Medicare (including Beneficiary copayments) savings as seen in Exhibit ES-3.

3. A potential AD treatment could result in an estimated $26.42 billion in 10-year Medicaid Nursing Home savings as seen in Exhibit ES-4.

4. A potential AD treatment could result in an estimated $11.86 billion in 10-year Informal Caregiving & Support cost savings as seen in Exhibit ES-5.

Methodology

In order to determine the possible Medicare, related Medicaid  societal economic impacts of a potential AD treatment for Medicare beneficiaries, our analytic methodology comprised three steps: 1) determine the number of beneficiaries in the potential treatment population within the AD patient disease spectrum; 2) for Medicare, determine costs attributable solely to AD for each year before death, 3) determine delay of progression attributable to the treatment, and 4) determine savings to this population attributable to this delay.

To determine the potential treatment population for this analysis, we started with the Weldon Cooper population projections[1] per the ages 65-84 and stepped down this population twice, once to find the potential MCI population and once to find the Mild AD population. For each of these step downs, we used four separate factors: 1) Prevalence of MCI and dementia due to any reason, 2) Proportion of MCI & dementia due to any reason presenting to healthcare, 3) Clinical Suspicion of AD etiology, and 4) Amyloid Beta confirmed AD Diagnosis[2]. Then we added the potential MCI population and the potential mild AD population together to determine the potential treatment population. We then stepped the population down using an incidence rate of 25% as the MCI /Mild AD portion of our analysis for a four-year period and a steadily increasing “take-up rate” from 25% in 2022 to 50% in 2031.

For this analysis, we assumed a delay of disease progression of six months for year 1, twelve months for year 2, eighteen months for year 3, and twenty-four months for all following years would be attributable to the potential AD treatment. Our estimates for the cost of Medicare related to AD in the years before death is taken from previous literature[3]. Similarly, our analysis of Medicaid savings due to reduction in the use of Nursing Home care and savings related to the reduction in need for informal care and support were drawn from available data.   

Discussion

Assessing only the Medicare health care cost savings impact understates the possible savings of a potential AD treatment given the broader cost burden of the disease. The potential treatment population for this innovative potential AD treatment is targeted and focused. Possible Medicare savings are constrained since the expenditures (including beneficiary copayments) associated with AD are ~11%[1] of total Medicare expenditures for AD patients. Our savings estimates are primarily achieved through reductions in Medicaid nursing home expenditures and informal caregiving & support expenditures. The most important benefit of a potential AD treatment is likely not monetary, but rather quality of life due to a delay in the debilitating features of AD.

You can download the full analysis here.

1. Alzheimer’s Association. (2020), 2020 Alzheimer's disease facts and figures. Alzheimer's Dement., 16: 391-460. https://doi.org/10.1002/alz.12068

2. Pyenson, B., Sawhney, T. G., Steffens, C., Rotter, D., Peschin, S., Scott, J., & Jenkins, E. (2019). The real-world Medicare costs of Alzheimer Disease: Considerations for policy and care. Journal of managed care & specialty pharmacy, 25(7), 800-809.

3. Ibid.

4. Alzheimer’s Association. (2020), 2020 Alzheimer's disease facts and figures. Alzheimer's Dement., 16: 391-460. https://doi.org/10.1002/alz.12068

5. Ibid.

6. Ibid.

7. University of Virginia Weldon Cooper Center, Demographics Research Group. (2018). National Population Projections. Retrieved from https://demographics.coopercenter.org/national-population-projections.

8. Derived from PPT slides provided by Biogen Team on 10/30/2020.

9. Pyenson, B., Sawhney, T. G., Steffens, C., Rotter, D., Peschin, S., Scott, J., & Jenkins, E. (2019). The real-world Medicare costs of Alzheimer Disease: Considerations for policy and care. Journal of managed care & specialty pharmacy, 25(7), 800-809.

10. Ibid.