Boards of Trustees for the Federal Hospital Insurance & Federal Supplementary Medical Insurance Trust Funds Issue Annual Report to Congress (Part 9 of 12)

WASHINGTON, June 8 (TNSrep) -- The Boards of Trustees for the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds issued a 263-page annual report to Congress on June 2, 2022.

(Continued from Part 8 of 12)

* * *

D. LONG-RANGE MEDICARE COST GROWTH ASSUMPTIONS

Sections IV.A, IV.B, and IV.C have described the detailed assumptions and methodology underlying the projected expenditures for HI (Part A), SMI (Parts B and D), and private health plans (Part C) during 2022 through 2031. These projections are made for individual categories of Medicare-covered services, such as inpatient hospital care and physician services.

As the projection horizon lengthens, it becomes increasingly difficult to anticipate changes in the delivery of health care, the development of new medical technologies, and other factors that will affect future health care cost increases. Accordingly, rather than extending the detailed projections by individual type of service for all future years, the Trustees use a more aggregated basis for setting cost growth assumptions in the long range. Such increases also reflect the substantial uncertainty associated with payments that are specified through statute, which may present challenges for the Medicare program.

Beginning with the 2013 report, the Trustees used the statutory price updates and the volume and intensity assumptions from the "factors contributing to growth" model to derive the year-by-year Medicare cost growth assumptions for the last 50 years of the projection period.67 The Trustees assume that the productivity reductions to Medicare payment rate updates will reduce volume and intensity growth by 0.1 percent below the factors model projection./68

The output and key assumptions of the factors model that are used in this year's report are similar to those used in the 2021 report. In subsequent reports, the Trustees will determine if additional historical data warrant a re-evaluation of these assumptions and a re-estimation of the factors model output. The remainder of section IV.D discusses the factors model and its role in the Medicare projections. Section V.C explains the methods used to derive the long-range cost growth assumptions underlying the illustrative alternative projection.

1. Long-Range Growth Assumptions for the Overall Health Sector

The first step to estimate the long-range Medicare trends is to determine the long-range assumptions affecting the overall health sector. The Trustees use the factors model to determine the year-by-year growth rates for the overall health sector over the last 50 years of the projection. Based on the factors model, the Trustees assume that the long-range per capita overall health spending growth is GDP plus 0.7 percent (or 4.3 percent) for 2046, gradually declining to GDP plus 0.4percent by 2096 (or 4.1 percent)./69

The per capita increase in overall health care costs is due to the combined effects of general inflation, medical-specific excess price inflation (above general price growth), and changes in the utilization of services per person and the intensity or average complexity per service. The Trustees assume that beginning in 2046 (i) general price inflation will remain constant at 2.05 percent per year, as measured by the GDP deflator; (ii) excess medical price inflation will remain constant at 0.75 percent per year; and (iii) the annual increase in the volume and intensity of services per person will decline gradually from approximately 1.5 percent in 2046 to 1.3 percent in 2096 based on the key economic assumptions and elasticity estimates from the factors model, as described below.

Excess medical price inflation for the overall health sector is assumed to grow at 0.75 percent annually from 2046 through 2096. This assumption is roughly equivalent to the difference between the growth in the personal health care deflator over the past three decades and the growth in the GDP deflator over this same period./70

Combining this assumption with the ultimate assumed growth rate of 2.05 percent per year in the GDP deflator yields the Trustees' estimate of the long-range rate of medical price growth of 2.8 percent annually. Using the relationship between medical price growth and resource-based health sector productivity growth/71 allows for the determination of medical input price growth./72

For resource-based health sector productivity, the Trustees assume that the rate of growth will be equivalent to published research/73 of 0.4 percent per year. Hence, the Trustees' estimate of the long-range rate of growth of medical input prices is 3.2 percent.

As stated earlier, the factors model is based on economic research that separates health spending growth into its major drivers--income growth, relative medical price inflation, insurance coverage, and a residual that primarily reflects the impact of technological development./74

The factors model provides the ability to model the expected behavioral effects associated with a continuing increase in the share of national income devoted to consumption of health care services. In particular, this approach is based on historically estimated income and price elasticities and uses measurable key variables, providing a foundation for developing the long-range growth assumptions./75

In the factors model, the sensitivity of health cost growth to each of the three factors must be estimated. Each sensitivity is measured as an elasticity, which is the percentage change in cost growth that is caused by a 1-percent change in a factor. The first elasticity, the income-technology elasticity, reflects the increase in demand for health care and new medical technologies in response to growth in income. The second elasticity, the relative medical price elasticity, reflects the sensitivity of consumers and purchasers in consuming health care to changes in excess medical price inflation. The final key elasticity is the insurance elasticity, which reflects the change in demand for medical care as the level of insurance coverage changes.

For the income-technology elasticity, the Trustees developed a time-trend-based method for projecting the elasticity that reflects the historical declining trend, produces results consistent with the elasticity implied by the most recent short-range national health expenditure (NHE) projections, and converges to 1.0 within a range of roughly 75 to 150 years. In the resulting projection, the income-technology elasticity is 1.24 in the 25th year of the projection period (2046) and declines at a slowing pace to 1.07 in the 75th year of the period (2096). This methodology results in an income-technology elasticity that reaches 1.0 in 2125. These are the same elasticity assumptions that were used for 2046 and 2096 in the 2021 report.

For the medical price elasticity, the Trustees assume a rising sensitivity of demand for health care to changes in relative medical price as the share of income devoted to health care rises. The medical price elasticity is determined for a given year by subtracting an income effect from a pure substitution effect. The income effect is determined by multiplying the share of income devoted to health care in that year by the estimated yearly income-technology elasticity. The substitution effect is assumed to be equal to -0.2 and represents the change in demand in response to a change in the relative price of health care holding utility constant. For the 2022 report, the Trustees project the price elasticity to be -0.50 for the 25th year of the projection (2046) and assume that it will follow a non-linear path until it reaches -0.56 in the 75th year of the projection (2096). Based on the RAND Health Insurance Experiment, the insurance elasticity was estimated at -0.2 and was assumed to be unchanged over the long range./76

Two additional assumptions are required to complete the factors model determination. First, relative medical price inflation must be estimated over the long-range projection period. As discussed previously, the Trustees assume a relative medical price growth rate of 0.75 percent per year. Second, insurance coverage is assumed to be unchanged over the long range in order to maintain consistency with the concept of a Medicare projection in which the Medicare benefit package is not altered.

2. Long-Range Growth Assumptions for Medicare

The Trustees have assumed since 2001 that it is reasonable to expect over the long range that the drivers of health spending will be similar for the overall health sector and for the Medicare program. This view was affirmed by the 2010-2011 Medicare Technical Review Panel, which recommended use of the same long-range assumptions for the increase in the volume and intensity of health care services for the total health sector and for Medicare. Therefore, the overall health sector long-range cost growth assumptions for volume and intensity are used as the starting point for developing the Medicare-specific assumptions.

Medicare payment rates for most non-physician provider categories are updated annually by the increase in providers' input prices for the market basket of employee wages and benefits, facility costs, medical supplies, energy and utility costs, professional liability insurance, and other inputs needed to produce the health care goods and services. These updates are then reduced by the 10-year moving average increase in economy-wide productivity, which the Trustees assume will be 1.0 percent per year over the long range. The Trustees assume that the full market basket increase would be approximately 3.2 percent annually, or about 0.4 percent greater than the net price increase of 2.8 percent per year described above for the total health sector. The different statutory provisions for updating payment rates require the development of separate long-range Medicare cost growth assumptions for four categories of health care provider services:

(i) All HI, and some SMI Part B, services that are updated annually by provider input price increases less the increase in economy-wide productivity.

The annual increase in Medicare payment rates for these services is reduced by the 10-year moving average increase in economy-wide productivity. These gains are estimated to be 1.0 percent per year over the long-range projection period. Combined with an assumed market basket increase of 3.2 percent, the statutory price update for these services is 2.2 percent per year over the long range. The initial projected increase in the volume and intensity of these Medicare services is assumed to be equivalent to the average projected growth in the volume and intensity of services for the overall health sector. The Trustees believe that the use of a common baseline rate of volume and intensity growth across all Medicare services is reasonable, as there would be only a small likelihood that one part of the health sector could continue to grow indefinitely at significantly faster rates of growth than do other parts.

Additionally, the Trustees assume that the growth in Medicare payment rates will reduce the volume and intensity growth of these services by 0.1 percent per year relative to the assumption from the factors model. The Trustees' assumption is based on the work of the 2010-2011 and 2016-2017 Medicare Technical Review Panels, both of which concluded that there would likely be a small net negative impact on volume and intensity growth due to reduced incentives to develop new technologies, provider exits, and the impact of greater bundling of services for payment purposes./77,/78

For new technology that leads to new services, Medicare would pay lower fees than would otherwise be the case, and providers would be less likely to adopt new services and innovations, thereby lowering the demand for, and intensity of, the medical care provided. Regarding provider exits, as fee-for-service fees declined relative to those assumed for private health insurance plans, facilities of marginal profitability would likely exit the Medicare market, reducing capacity and volume. This change could also cause a more bifurcated health system in which only providers that could operate profitably under Medicare would offer services to Medicare beneficiaries, with a tendency to provide only the more basic services not associated with new medical technologies. Finally, the innovations being tested for the Medicare program, such as bundled payments or accountable care organizations, could reduce incentives to adopt new cost-increasing technologies and increase incentives to adopt new cost-decreasing technologies for those participating in these programs and/or could contribute to greater efforts to avoid services of limited or no value within the service bundle.

Reflecting all of these considerations, the year-by-year long-range cost growth rate assumption for these HI and SMI Part B services starts at 3.6 percent in 2046, or GDP plus 0percent, and gradually declines to 3.4 percent by 2096, or GDP minus 0.3 percent.

(ii) Physician services

Payment rate updates are 0.75 percent per year for those qualified physicians assumed to be participating in advanced alternative payment models (advanced APMs) and 0.25 percent for those assumed to be participating in the merit-based incentive payment system (MIPS) in the long range. The year-by-year cost growth rates for physician payments are assumed to decline from 3.2 percent in 2046, or GDP minus 0.4 percent, to 2.8 percent in 2096, or GDP minus 0.9 percent.

(iii) Certain SMI Part B services that are updated annually by the CPI increase less the increase in productivity.

Such services include durable medical equipment (DME) that is not subject to competitive bidding,/79 care at ambulatory surgical centers, ambulance services, and medical supplies, which are updated by the CPI and reduced by the 10-year moving average increase in economy-wide productivity. For these services, the Trustees initially assume that the rate of per beneficiary volume and intensity growth is equivalent to that derived for the overall health sector using the factors model. This volume and intensity growth is assumed to be reduced by 0.1 percent per year, as described above. The volume and intensity assumption is combined with the long-range CPI assumption (2.4 percent) minus the productivity factor (1.0 percent) to produce a long-range growth assumption for these SMI Part B services. The corresponding year-by-year cost growth rates gradually decline from 2.8 percent in 2046, or GDP minus 0.8 percent, to 2.6 percent in 2096, or GDP minus 1.1 percent.

(iv) All other Medicare services, for which payments are established based on market processes, such as prescription drugs provided through Part D and the remaining Part B services.

The Trustees assume that per beneficiary outlays for these other Part B services, which constitute about 36 percent of total Part B expenditures in 2031, and for all Part D services grow at the same rate as the overall health sector as determined from the factors model. The services are assumed to grow similarly because their payment updates are determined by market forces, such as the competitive-bidding process for Medicare Part D. The year-by-year cost growth rates gradually decline from 4.3 percent in 2046, or GDP plus 0.7 percent, to 4.1 percent by 2096, or GDP plus 0.4 percent.

In addition, these long-range cost growth rates must be modified to reflect demographic impacts. Beginning with the 2020 report, these impacts reflect the changing distribution of Medicare enrollment by age, sex, and the beneficiary's proximity to death, which is referred to as a time-to-death (TTD) adjustment. The TTD adjustment reflects the fact that the closer an individual is to death, the higher his or her health care spending is. Thus, as mortality rates improve and a smaller portion of the Medicare population is likely to die at any given age, the effect of individuals getting older and spending more on health care is offset somewhat./80

After combining the rates of growth from the four long-range assumptions, the weighted average cost growth rate for Part B is 3.8 percent, or GDP plus 0.2 percent in 2046, declining to 3.7 percent, or GDP plus 0 percent by 2096. When Parts A, B, and D are combined, the weighted average cost growth rate is 3.8 percent in 2046, or GDP plus 0.2 percent, declining to 3.7 percent, or GDP plus 0 percent by 2096.

As in the past, the Trustees have established detailed growth rate assumptions for the initial 10 years of the projection period by individual type of service (for example, inpatient hospital care and physician services), reflecting recent trends and the impact of all applicable statutory provisions. For each of Parts A, B, and D, the assumed growth rates for years 11 through 25 of the projection period are set by interpolating between the rate at the end of the short-range period and the rate at the start of the final 50 years of the long-range period described above. The 2016-2017 Medicare Technical Review Panel concluded that both the current length of the transition period and the current approach to the transition are reasonable, and they recommended that the Trustees continue to use the same approach to transitions between short-range and long-range projections for both HI and SMI./81

* * *

The report is posted at: https://www.cms.gov/files/document/2022-medicare-trustees-report.pdf

(Continues with Part 10 of 12)

TARGETED NEWS SERVICE (founded 2004) features non-partisan 'edited journalism' news briefs and information for news organizations, public policy groups and individuals; as well as 'gathered' public policy information, including news releases, reports, speeches. For more information contact MYRON STRUCK, editor, [email protected], Springfield, Virginia; 703/304-1897; https://targetednews.com