MCBS Analysis Plan

MCBS Analysis Plan, page 1of 7

ANALYSIS P LAN
MEDICARE CURRENT BENEFICIARY SURVEY
Overview ............................................................................................................................. 1
Uses of MCBS information ................................................................................................ 2
Tracking performance ..................................................................................................... 2
Baseline cost estimates.................................................................................................... 3
Analysis of supplementary health insurance................................................................... 4
Analysis of alternatives to fee- for-service Medicare ...................................................... 4
Monitoring and evaluating the effects of program changes............................................ 4
Access to care.............................................................................................................. 5
Correlates of health care ............................................................................................. 5
Institutional and long term care facilities, and other living arrangements...................... 5
Variations in Medicare need and use among subgroups of enrollees ............................. 6
Information about the nation’s health dollar................................................................... 7
Summary: Shaping the future health care system for the elderly ....................................... 7

OVERVIEW
As health care costs mount and America’s population ages, the Medicare program
becomes an increasingly pivotal issue in national politics.
Certainly this is the case regarding the Federal budget. Much is made of the impending
insolvency of the Part A trust fund, but there is a more immediate budget issue in the
form of general revenue going to Part B. In Federal fiscal year (FY) 1970, 0.5 percent of
Federal on-budget receipts was spent on SMI – about a tenth of the 5.4-percent on-budget
deficit that year. In FY 2002 SMI outlays consumed 5.4 percent of receipts, almost a third
of the 18.3-percent deficit. And in FY 2007, projections show SMI outlays to be 5.5
percent of receipts, almost three quarters of the 7.3-percent deficit.

MCBS Analysis Plan, page 2of 7

Medicare is also a pivotal issue in terms of the number of people touched by the program.
The size of the elderly population in the United States is growing, in number and in its
share of the total population. Over the past half century, the elderly U.S. population
(those aged 65 years and older) increased in numbers, from 12.3 million people in 1950
to 35.0 million in the year 2000. During that same time period, elderly people increased
as a proportion of the population, from 8.2 percent of all Americans to 12.4 percent.
Government projections of the population growth put the number of elderly Americans at
82.0 million in the year 2050 — 20.3 percent of the total population. And for each elderly
person affected by the Medicare program there are likely to be several children concerned
about their parent’s health and financial well-being.
From a public policy perspective, then, it is essential to understand how and how well
Medicare is meeting the needs of its population. It is equally important to be able to
speculate reliably on the economic consequences of changes to the program, both for the
Federal government and for Medicare beneficiaries.
The Medicare Current Beneficiary Survey (MCBS) is the ideal survey to provide this
information. It is specifically designed to produce the data needed to assess the
feasibility, impact, and costs of proposals to change Medicare, and to monitor the impact
of broader health care changes on Medicare beneficiaries. It provides a unique analytical
data set for use by analysts who seek to understand how Medicare interacts with the
social, economic, and demographic life factors of its beneficiaries. Unlike other health
services research surveys, the MCBS is focused on the Medicare population, resulting in
data that can be used to detect and to simulate broader changes than can be done with
smaller data sets.
The Medicare Current Beneficiary Survey is also an ideal vehicle to monitor the impact
of health reforms over time. Being continuously in the field with a stable instrument, it
provides data to examine the effects of health care reform through pre- and post-reform
comparisons. MCBS users can detect changes in health insurance coverage, beneficiary
satisfaction with care, health status, use of services and public and private expenditures
for health care very rapidly. The analytical power of the data is enhanced by a routine
oversample of disabled enrollees and the oldest old – groups of particular policy interest.

USES OF MCBS INFORMATION
Tracking performance
As an indication of its usefulness for program evaluation, CMS is using MCBS data to
measure five of the Performance Goals in its FY2004 Government Performance and
Results Act (GPRA) Annual Performance Plan. These goals involve:
•

Influenza and pneumococcal vaccination,

•

Use of mammography services,

•

Effectiveness of dissemination of Medicare information to beneficiaries,

MCBS Analysis Plan, page 3of 7

•

Beneficiary understanding of basic features of the Medicare program, and

•

Awareness of opportunities to enroll in Medicare Savings Programs.

Baseline cost estimates
The CMS Office of the Actuary is regularly required to make cost estimates for proposed
changes in the Medicare program. These estimates are prepared for CMS leadership, for
the Department, OMB, and the White House; and for the Congress. Accurate cost
estimates depend upon current and reliable data on the patterns of health service use and
expense of the Medicare population. Person-based use and expense data, which the
MCBS provides, are particularly valuable in estimating costs. They permit analytic
differentiation between persons with and without certain characteristics, such as
secondary insurance coverage, or across variables that are graduated or scaled, like
income.
Typically it is difficult or impossible to produce cost estimates for Medicare program
expansions solely from Medicare administrative data. Proposals usually affect
subpopulations of enrollees along income lines, supplementary insurance characteristics,
and the like – information not collected by the program. A prime example was provided
by the debate over the cost of prescription drug coverage under the Medicare
Catastrophic Coverage Act (1988) .In order to obtain baseline data on the current cost of
drugs for the Medicare population, the Office of the Actuary had to rely on external data
sources that were outdated and not completely suited for the task. This resulted in an
inability to produce a generally accepted set of cost estimates for the drug benefit until
after the legislation was enacted. Without the MCBS, similar problems can be expected
in obtaining baseline data for the cost of program restructuring or expansions. The MCBS
addresses this problem by providing a complete picture of the use of all health services
by the Medicare population, most particularly those services not covered under Medicare
for which administrative records are not available.
A recurring question that arises in estimating costs of program expansions is the size of
"induced demand." This factor, also called the “insurance effect,” reflects the increase in
use and expenditure that occurs when Medicare coverage is extended to a previously
uncovered service. Induction can occur in the sho rt term and in the long term, and there is
considerable uncertainty over its size in the Medicare population. By providing utilization
rates for services not covered by Medicare, distributed by secondary insurance coverage
status and health status, the MCBS assists estimation of the magnitude of the induction
factor.
After the fact, the longitudinal nature of the MCBS provides the capability to compare
cost estimates to actual experience. This provides a feedback mechanism to validate and
improve cost estimation techniques. Empirical verification of the cost associated with
induced demand should greatly support the actuaries' estimation and incorporation of this
phenomenon in subsequent cost estimates.

MCBS Analysis Plan, page 4of 7

Analysis of supplementary health insurance
A primary information need when evaluating changes to Medicare is the extent and type
of supplementary health insurance coverage among the Medicare population. Program
changes intended to promote more cost-effective choices by Medicare beneficiaries, if
they are to achieve their ends, depend on accurate knowledge of the extent of
supplemental health insurance coverage, and its effect on point-of-service beneficiary
liabilities, deductibles and coinsurance.
The MCBS continuously collects detailed information on the number and type of
additional policies owned by beneficiaries. It also documents any shifts in coverage by
health insurance policies. Information is available on supplemental insurance
expenditures for all type of services--including those, which Medicare does not cover,
such as prescription drugs and long term nursing care.
In addition to aiding in estimates of the cost of potential new benefits, information on the
distribution of secondary insurance coverage of services facilitates analysis of who is
likely to be affected by proposed changes. The MCBS collects a wide variety of
person-level socioeconomic and demographic data, allowing an evaluation of which subpopulations will be most helped by Medicare program expansions.
Analysis of alternatives to fee-for-service Medicare
The relatively rapid rise (and collapse) of the Medicare+Choice (M+C) program has
focused attention on how managed care and other alternatives to fee-for-service (FFS)
benefits work or could work in the Medicare program. The MCBS has been used—and
continues to be used—to compare experience of enrollees in M+C and FFS, especially as
utilization patterns of M+C enrollees are not recorded in the administrative data systems.
The ability to oversample in geographic areas heavily served by M+C pla ns was used for
several years to boost the number of survey participants enrolled in those plans, providing
information about the differences in life factors, program knowledge, and program
satisfaction in the M+C and FFS populations.
Monitoring and evaluating the effects of program changes
The Medicare Current Beneficiary Survey is the best vehicle to monitor and evaluate the
impact of Medicare changes over time. Because it is continuously in the field with a
stable panel of beneficiaries, it can be used to examine the effects of Medicare program
changes through pre- and post-reform comparisons. Using the MCBS, analysts can detect
changes in secondary health insurance coverage, health status, satisfaction with care, use
of Medicare services and out-of packet expenditures for health care. The analytical power
of the data is enhanced by a routine oversample of disabled enrollees and of the oldest
old, and by periodic oversamples of other populations of policy interest.

MCBS Analysis Plan, page 5of 7

Access to care
An example of the value of the MCBS for monitoring and evaluating program changes is
the issue of access to care. The reconciliation acts of 1989 and 1990 changed the way in
which Medicare pays physicians. In the face of widespread concern that physician
payment reform might have the unintended consequence of decreasing the Medicare
population's access to physician care, Congress mandated that CMS and PPRC monitor
access and prepare annual reports to Congress on their findings. Both CMS and PPRC
relied on the MCBS as the primary data source in their analyses of the impact of
physician payment reform on access to physician care.
Correlates of health care
Health services research has demonstrated that health and health care use do not operate
in a vacuum. There are intricate and little-understood connections between health and
health care and social, economic, and demographic life factors. MCBS data have been
used to examine the effects of:
Income and wealth on access and health status,
Access to nursing homes, access to transitional care, and the relationship of housing
characteristics, residential patterns, and social context (within the family, among friends
and neighbors) on use and access to health services,
Other public insurance, such as the Qualified Medicare Beneficiary (QMB) program, on
use of services (a special supplement was added to the MCBS to evaluate the extent to
which the benefit is being used, and to determine the best means to reach eligible
persons), and
Medicaid spend-down on use patterns for both community and institutionalized
individuals.
Institutional and long term care facilities, and other living arrangements
The past decade has been witness to rapid growth of retirement communities and other
forms of living arrangements between the tradition single-family dwelling and the
traditional nursing home. What is not well understood is how people in those various
settings perceive and need Medicare. Through the MCBS, information is available on the
array of services available to enrollees who live in different settings, and on the
characteristics of those enrollees. The MCBS can be used to identify patterns that predict
institutionalization and compare the characteristics that lead to this over time, including
living arrangements, age and degree of disability or dependence as measured by ADLs
and IADLs.
MCBS data are being used to study long term home health care and transitional (post
acute) care and long-term care rendered by providers such as hospitals and rehabilitation

MCBS Analysis Plan, page 6of 7

facilities. The MCBS provides detailed information on the financing of all forms of long
term care and sources of payment for these services including third party payments and
out of pocket or family expenditures.
Analysis of the balance between institutional and noninstitutional services, availability,
cost and living arrangements can be examined by comparing information collected by
both the community and the facility components. Extensive information is collected on
home health- care in the community component, thus facilitating comparisons of the two
groups (institutionalized and non- institutionalized) by functional status, selected
demographic characteristics and patterns and cost of utilization over time.
Variations in Medicare need and use among subgroups of enrollees
Many older studies of the Medicare population focused upon the elderly as a
homogenous group, primarily because of sample size concerns. The MCBS is a rich data
base focused on the elderly and the disabled, which allows analysis of a variety of
subgroups that are of particular interest to policy makers, such as the dually eligible
enrollees, near-poor elderly enrollees, disabled enrollees, the oldest old, people living
alone, the working elderly and people living in rural areas. It is an ideal data set for use in
policy studies that increasingly focus on subsets of the Medicare population.
The MCBS collects detailed information on persons who have dual, or crossover,
coverage for Medicaid and Medicare. Most proposals for health care system reform pay
close attention to this group of low- income elderly persons. The data collected permit
detailed comparisons of differences in demographic and socioeconomic characteristics
for dual-eligible persons compared to those with Medicare-only entitlement, as well as
comparisons of use rates and expenditure differences for these two populations. This
information improves our understanding of how the dually-eligible enrollees contribute to
the expenditure growth of both programs. It also provides an information base for
evaluating the person and expenditure impacts of changes to either Medicare or Medicaid
that have feedback or transfer payment effects on the other program.
The MCBS collects detailed data on income and assets, allowing analysis of the near
poor elderly who are of particular policy interest. This type of analysis is of particular
interest at the present, as Congress debates legislation offering subsidies for drug
insurance to low- income enrollees: do they use prescription drugs in different patterns?
Do they have other forms of drug insurance?
MCBS will provide a rich data source for the analysis of the health care use and
expenditure patterns of non-elderly Medicare beneficiaries. Crosscutting person level
data are available about this subgroup, including demographic characteristics, initial
reasons for eligibility, utilization patterns and access to health care. Analysts can examine
the effects of any health care reform initiatives on disabled beneficiaries' financial
liability and access to care.

MCBS Analysis Plan, page 7of 7

The continuous nature of the survey shows trends in the type and distribution of services
by the Medicare population over time. For example, analysts can track the effects of the
aging of the elderly population on use of services and health expenses. Because of its
structure, the MCBS can be used to separate cohort effects from secular effects in
Medicare trends.
Information about the nation’s health dollar
CMS is responsible for maintaining the U.S. National Health Accounts. These accounts
comprise estimates of the nation's total annual expenditures for health care, by type of
service and channel of payment. Figures are estimated not only for past years, but for
future years as well. The National Health Accounts are the official federal government
estimates of health care spending; as such, they are the essential information for
understanding the current system of financing health care.
Capture all personal health care expenditures for the Medicare population, the MCBS
provides a complete picture of more than 40 percent of the nation's health expenditures.
Used in connection with other data, this information will strengthen the accuracy of the
National Health Accounts, making them a better tracking tool for the financial effects of
health care reform on the total health system. This is an important contribution to
improving the data that is widely used to form the nation's picture of the directions,
progress and problems of its health sector.

SUMMARY: SHAPING THE FUTURE HEALTH CARE SYSTEM FOR THE
ELDERLY

The MCBS is a continuous information system to monitor emerging structural changes in
the U. S. health care financing and delivery system. Linking survey collected information
with the CMS administrative files further expands the power of the data to provide not
only a continuous service use record, but to allow prospective estimates to be made.
The MCBS provides a rich information base about trends and changes in the types of care
received, the settings for care, and more general changes in the patterns of medical
practice. This information will allow us to identify and report on significant trends and
changes as they are occurring. It will also provide detailed information on beneficiaries'
health status and history, their ability to perform activities of daily living, their insurance
and financial position, and their family circumstances and support systems. This
information will be essential to evaluating the financial, social, administrative, and
political feasibility of proposals intended to expand current alternatives in order to meet
the needs of the growing elderly population of the U.S.