Medicaid Managed Care: Serving the Disabled Challenges State Programs
(Chapter Report, 07/31/96, GAO/HEHS-96-136).
Pursuant to a congressional request, GAO examined state efforts to
include disabled Medicaid beneficiaries in prepaid managed care
programs, focusing on the: (1) extent to which states are implementing
Medicaid prepaid managed care programs for disabled beneficiaries; and
(2) steps that have been taken to safeguard the interests of the
stakeholder groups.
GAO found that: (1) few states have significant experience with prepaid
care programs for disabled Medicare beneficiaries; (2) 13 states have
submitted proposals to enroll their disabled beneficiaries in prepaid
care, and 12 of these states intend to make this enrollment mandatory;
(3) half of the states with prepaid enrollment programs rely on disabled
individuals to inform them of their dissatisfaction with their plans;
(4) states that mandate the enrollment of their disabled beneficiaries
in prepaid managed care plans develop more appropriate quality assurance
mechanisms; (5) these states build safeguards into health care programs
by incorporating adequate planning and consensus-building, and tailoring
programs to meet the specific needs of disabled Medicare beneficiaries;
(6) it is difficult for states to maintain a financially sound managed
care system for disabled individuals; (7) some disabled groups have
substantially higher medical costs than other groups; (8) individuals
with high-cost disabilities find it more difficult to enroll in managed
health care programs and experience limited health care services; (9) a
few states are adopting an approach, to limit the amount of profit a
plan can earn or lose; (10) this approach lessens the incentive to limit
service to high-cost beneficiaries; and (11) careful attention should be
given when including the disabled population in prepaid health care
plans, given their complex health care needs and states' limited
experience with serving disabled individuals in prepaid settings.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: HEHS-96-136
TITLE: Medicaid Managed Care: Serving the Disabled Challenges
State Programs
DATE: 07/31/96
SUBJECT: Managed health care
State-administered programs
Medical economic analysis
Health resources utilization
Health maintenance organizations
Health services administration
Quality assurance
Medical services rates
Health insurance cost control
Handicapped persons
IDENTIFIER: AFDC
Aid to Families with Dependent Children Program
AIDS
Supplemental Security Income Program
Medicaid Program
Medicare Program
Arizona
Delaware
Oregon
Tennessee
Utah
Virginia
Massachusetts
Ohio
Wisconsin
District of Columbia
California
Colorado
Florida
Maryland
Michigan
New Jersey
Pennsylvania
Ohio Accessing Better Care Program
Wisconsin I-Care Program
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to the Chairman and Ranking Minority Member, Subcommittee on
Medicaid and Health Care for Low-Income Families, Committee on
Finance, U.S. Senate
July 1996
MEDICAID MANAGED CARE - SERVING
THE DISABLED CHALLENGES STATE
PROGRAMS
GAO/HEHS-96-136
Medicaid Managed Care for the Disabled
(101389)
Abbreviations
=============================================================== ABBREV
AFDC - Aid to Families With Dependent Children
AIDS - acquired immunodeficiency syndrome
CMA - Community Medical Alliance
HCFA - Health Care Financing Administration
HHS - Department of Health and Human Services
HIV - human immunodeficiency virus
HMO - health maintenance organization
ICF/MR - intermediate care facility for the mentally retarded
IMD - institution for mental diseases
NF - nursing facility
SSI - Supplemental Security Income
Letter
=============================================================== LETTER
B-271532
July 31, 1996
The Honorable John H. Chafee
Chairman
The Honorable Bob Graham
Ranking Minority Member
Subcommittee on Medicaid and Health Care
for Low-Income Families
Committee on Finance
United States Senate
This report, prepared at your request, describes state efforts to
include disabled Medicaid beneficiaries in prepaid managed care
programs.
We are sending copies to the Secretary of Health and Human Services;
the Administrator, Health Care Financing Administration; and state
Medicaid directors. We will also make copies of this report
available to others on request.
Please contact me at (202) 512-7114 if you or your staff have any
questions. Major contributors to the report are listed in appendix
I.
William J. Scanlon
Director, Health Financing and Systems Issues
EXECUTIVE SUMMARY
============================================================ Chapter 0
PURPOSE
---------------------------------------------------------- Chapter 0:1
Prepaid managed care plans, which deliver medical services for a
fixed (or "capitated") per-person fee, are an increasingly common
part of Medicaid, the nation's largest health care program for the
poor. With their emphasis on primary care, restricted access to
specialists, and control of services, prepaid plans are seen as a way
to help control spiraling Medicaid costs, which totaled $159 billion
in fiscal year 1995. Thus far, states have extended prepaid care
largely to low-income families--about 30 million individuals--but to
few of the additional 6 million Medicaid beneficiaries who are
mentally or physically disabled. Managed care's emphasis on primary
care and control of service use differs from the care needs of
disabled beneficiaries--many of whom need extensive services and
access to highly specialized providers, which in some cases are
essential to prevent death or further disability. However, because
over one-third of all Medicaid payments go for their care, greater
attention is being focused on whether disabled individuals can be
integrated successfully into managed care.
These efforts affect three key stakeholder groups: disabled
beneficiaries, who include a small number of very vulnerable
individuals who may be less able than others to effectively advocate
on their own behalf for access to needed services; the prepaid care
plans, which are concerned about the amount of financial risk
involved in treating people with extensive medical needs; and the
states and federal government, which run Medicaid. The Chairman and
Ranking Minority Member of the Subcommittee on Medicaid and Health
Care for Low-Income Families, Senate Committee on Finance, asked GAO
to examine (1) the extent to which states are implementing Medicaid
prepaid managed care programs for disabled beneficiaries and (2) the
steps that have been taken to safeguard the interests of all three
stakeholder groups. GAO's review of safeguards focused on two
areas--efforts to ensure quality of care and strategies for setting
rates and sharing financial risk.
BACKGROUND
---------------------------------------------------------- Chapter 0:2
Medicaid is funded jointly by the states and the federal government
and operated mainly by the states. It provided health care coverage
for 40 million people in fiscal year 1995, about one in seven of whom
was disabled. Some categories of mildly disabled individuals have
health care costs that closely mirror those of the general
population, but others, such as those with cystic fibrosis or
end-stage acquired immunodeficiency syndrome (AIDS), have costs that
are much higher.
Medicaid has traditionally been a fee-for-service program, meaning
that doctors, hospitals, and other providers are paid based on the
number and type of services they provide. States have relatively
wide latitude in structuring Medicaid programs, including making
prepaid care available to those who wish to enroll. But states must
obtain federal approval to require prepaid plan enrollment or to
restrict individuals to specific plans. This approval is designed to
help ensure that everyone who is eligible has access to care.
RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3
Serving disabled beneficiaries through Medicaid managed care poses
complex, new challenges to the states. To date, few states have
significant experience with prepaid care for disabled Medicaid
beneficiaries, many of whom have chronic conditions that require
ongoing and costly specialty care. Of the six states that require
some or all of their disabled population to enroll in prepaid care,
only one program is more than 3 years old. Eleven others have
voluntary programs enrolling a small percentage of disabled
beneficiaries. However, because of continued concern about cost
containment, 13 more states have submitted proposals to enroll
disabled beneficiaries in prepaid care, with 12 of them intending to
make enrollment mandatory.
One of the challenges for states is developing both the service
networks and the necessary assurances that the health care needs of
disabled beneficiaries are being met appropriately. However, about
half of the states tend to rely on mechanisms such as the freedom of
disabled individuals to disenroll from or switch prepaid plans or on
their access to the states' and plans' complaint and grievance
systems to help ensure quality of care. While analyses of patterns
of disenrollment or complaints can provide meaningful information, in
the aggregate they may not be sufficient to detect systemic
deficiencies in care for disabled beneficiaries. In contrast, states
that either mandate enrollment or provide small, voluntary programs
focused exclusively on disabled beneficiaries tend to be furthest
along in developing assurances that appropriate, quality care is
available to them. Examples of such actions include requiring health
plans to designate advocates to help coordinate the services disabled
beneficiaries receive and to provide access to specialists
specifically trained in care for disabled individuals.
A second challenge for states is developing and administering a
managed care system for disabled beneficiaries that is financially
sound. However, few states have ventured beyond current rate-setting
approaches that base capitation rates on average costs for large
segments of statewide Medicaid populations, such as families with
children or disabled individuals. However, within the disabled
population some groups, such as quadriplegics or AIDS patients, have
substantially higher medical costs (sometimes more than $50,000 in a
given year), while others, such as mildly disabled individuals, may
have few or no additional costs beyond those of the general
population. The ability to identify individuals with high-cost
disabilities could lead managed care plans to try to avoid enrolling
them or to encourage them to disenroll by limiting services
inappropriately. The development of more appropriate rate-setting
approaches that link rates more closely to individuals' likely costs
is still in the experimental stage. States are further along in
adopting methods to share the risk of losses experienced by plans
that enroll a relatively large number of high-cost individuals. For
example, reinsurance programs are the most common form of sharing
such losses. A few states are adopting an approach called a "risk
corridor," which limits the amount of profit that health care plans
can earn as well as the amount of loss they could face, thereby
reducing the incentive to inappropriately limit services or to avoid
enrolling high-cost individuals.
PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4
FEW STATES HAVE SIGNIFICANT
EXPERIENCE
-------------------------------------------------------- Chapter 0:4.1
Arizona, Delaware, Oregon, Tennessee, Utah, and Virginia are the only
states requiring some or all of their disabled beneficiaries to
participate in prepaid care programs. These states enroll disabled
beneficiaries in prepaid managed care plans that also cover other
types of Medicaid recipients. Arizona's program, established in 1982
and currently enrolling more than 70,000 disabled, is the only
mandatory program more than 3 years old. In contrast, Massachusetts,
Ohio, Wisconsin, and the District of Columbia have small-scale
voluntary programs solely for disabled individuals, none of which
serves more than 3,000 beneficiaries. Seven other states
(California, Colorado, Florida, Maryland, Michigan, New Jersey, and
Pennsylvania), as well as Massachusetts, allow disabled beneficiaries
to enroll voluntarily in plans open to other Medicaid beneficiaries.
In these states, less than 20 percent of the disabled population have
chosen to enroll.
One problem identified thus far in states with federal approval to
restrict beneficiaries' freedom to change providers has been
coordinating enrollment for the estimated one-third of disabled
individuals who are "dually eligible" for health care under Medicaid
and Medicare. Medicare law guarantees these individuals more freedom
in switching providers than they have under Medicaid managed care
programs, which require prepaid plan enrollment. The few states
wrestling directly with this issue have taken varying approaches,
ranging from adjusting their programs to conform with Medicare
requirements to seeking waivers of Medicare law that would allow
requirements closer to Medicaid's.
SIGNIFICANT EFFORTS NEEDED
TO ENSURE QUALITY
-------------------------------------------------------- Chapter 0:4.2
States that rely on monitoring the services prepaid care plans
provide to the average enrollee may find that these efforts do not
provide enough specificity for assessing care received by disabled
enrollees. For example, problems in care provided to a very
vulnerable disabled category, such as quadriplegics, might escape
general view because few if any cases of quadriplegia would generally
appear in random samples across the entire population served by a
health care plan. Most states recognize a need to specifically
monitor managed care for disabled enrollees and plan to do so as they
expand their programs.
Important aspects of states' quality assurance activities can fall
into two main categories: (1) building safeguards into the programs
through adequate planning and consensus-building and (2) tailoring
various aspects of the program (such as enrollment and monitoring) to
meet the specific needs of disabled beneficiaries. To date, most of
the efforts have been made by several states with mandatory
participation by disabled individuals or by states with programs
targeted exclusively to disabled beneficiaries on a voluntary basis.
The following are examples:
-- Oregon's Medicaid staff met weekly with health plans, advocates
for disabled individuals, and others for more than a year before
the program was implemented.
-- Wisconsin requires the health plan serving participants in its
targeted prepaid care program, which serves only disabled
beneficiaries, to have a Medicaid advocate on staff who is
knowledgeable about disabilities. Wisconsin also mandates that
case managers conduct needs assessments within 55 days of
enrollment in the plan.
-- Massachusetts allows specialists to act as primary care
providers and uses a health needs assessment that assists
enrollment staff in working with beneficiaries to select a plan.
Information about the services provided to disabled patients is
essential for effective monitoring. Since services are no longer
paid for on a fee-for-service basis, however, the reimbursement
process no longer produces this information. Developing
comprehensive, consistent data on services provided under prepaid
care takes time and effort. To date, only Arizona has substantial
experience in doing so. The effort, which can be expensive and
time-consuming, can permit states to identify areas in which service
utilization rates are overly low or high. It can also allow states
to track movement of high-cost individuals among health plans, a step
that could help spot service delivery problems.
EXPERIMENTATION IS UNDER WAY
IN RATE-SETTING AND
RISK-SHARING
-------------------------------------------------------- Chapter 0:4.3
Prepaid care capitation rates are normally based on average costs for
broad categories of beneficiaries, such as all disabled people in a
state. However, some categories of disabled individuals have very
high costs, while others have relatively low costs. Paying the same
rate for groups with different health care needs increases the risk
that plans will seek to enroll only the healthier, less expensive
individuals. If plans feel financial pressure from treating
high-cost cases, they may also seek to limit inappropriately the
services these individuals receive. Three states (Massachusetts,
Missouri, and Ohio) are experimenting with ways to set rates for
disabled enrollees that more accurately reflect their varying needs
for care. For example, Ohio is exploring an approach that varies the
prepaid rate based on prior medical costs, with medical plans
receiving more money for people with demonstrated higher needs.
Most states that include disabled beneficiaries in prepaid care, and
especially those with mandated enrollment, provide some form of
"safety net" for plans that experience losses related to treating
high-cost cases. The most common form is called
"reinsurance"--essentially an insurance policy that plans can buy.
Reinsurance is directed only at losses. Five states (District of
Columbia, Massachusetts, Ohio, Utah, and Wisconsin) have implemented
another type of arrangement, called a "risk corridor," that not only
shares losses between the plan and the state but also restricts how
much of its capitation payments the plan can retain after paying for
enrollees' health care needs. In Massachusetts, for example, plans
serving those who are severely disabled must return to the state any
profit that exceeds 10 percent of the capitation payments they
received. Under a risk corridor, a plan's incentive to limit
services inappropriately and thereby increase the amount it may
retain is reduced because such amounts are limited to a maximum.
RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5
GAO is not making recommendations in this report.
AGENCY COMMENTS
---------------------------------------------------------- Chapter 0:6
GAO provided a draft of this report to the Administrator, Health Care
Financing Administration (HCFA), and to Medicaid officials from the
17 states in its study. In addition, GAO requested comments from
several independent experts in the fields of Medicaid and prepaid
care for people with disabilities. HCFA had no comments, while
comments from states and researchers were primarily technical or
clarifying and were incorporated as appropriate. Officials from one
state commented that the draft seemed to question the suitability of
prepaid managed care for people with disabilities. GAO believes,
rather, that careful attention to program design and implementation
is needed when including this vulnerable population in prepaid care,
given their complex health care needs and the limited experience to
date with serving them in prepaid settings.
BACKGROUND
============================================================ Chapter 1
Medicaid, a joint federal-state health financing program for the
poor, provided health care coverage for more than 40 million people
in fiscal year 1995. Medicaid expenditures--about $159 billion in
fiscal year
1995\1 --have more than tripled in the past 10 years. Under current
projections, they will double again within 8 years.
To help constrain rising costs, a number of states are making
increased use of prepaid managed care in their Medicaid programs.
Under this approach, a medical plan such as a health maintenance
organization (HMO) agrees to make a specified set of medical benefits
available in exchange for a prepaid amount of money per person. This
approach is considered less expensive than the traditional
fee-for-service approach because it eliminates the incentive to
provide unnecessary or overly expensive services in order to maximize
revenues.
Thus far, most states have focused their Medicaid managed care
efforts on programs for low-income families, which accounted for
about 73 percent of Medicaid beneficiaries in fiscal year 1994.
However, states are now directing more attention to using managed
care for another group of Medicaid beneficiaries--those who are
disabled. These individuals constitute about 15 percent of all
Medicaid beneficiaries, but because many of them have a heavy need
for specialized medical services, they account for over one-third of
all Medicaid expenditures.
For a number of reasons, such as their ongoing dependence on
specialized care and the wide diversity of types and severity of
conditions, bringing disabled people into managed care presents
challenges that differ from covering many other segments of the
population. Dealing with these challenges involves ensuring that
adequate mechanisms are in place to safeguard the interests of all
three major stakeholder groups: the disabled beneficiaries, who are
concerned about adequate access to quality care; the managed care
plans, which are concerned about not assuming inappropriate or
excessive financial risk; and the states and federal government,
which are concerned about protecting the interests of both
beneficiaries and taxpayers. For the most part, this is new
territory: Most states have little or no experience in adjusting
their managed care programs to meet these specialized needs.
--------------------
\1 Amounts include both health services and administration for
federal fiscal year 1995, which ended Sept. 30, 1995.
DISABLED PEOPLE ARE ONE OF
SEVERAL GROUPS ELIGIBLE FOR
MEDICAID
---------------------------------------------------------- Chapter 1:1
The range of services provided under Medicaid varies from state to
state. Established in 1965 as title XIX of the Social Security Act
(42 U.S.C. 1396-1396s), Medicaid programs are required under federal
law to provide eligible beneficiaries with certain primary, acute,
and long-term care benefits. Examples include physician services,
hospital care, laboratory services, preventive care for children, and
nursing facility care. At their option, states\2 may also elect to
provide coverage for an array of other services, such as prescription
drugs, medical equipment, eyeglasses, dental care, and ancillary
services such as physical and speech therapy. Medicaid is
administered at the state level, with the Health Care Financing
Administration (HCFA) within the Department of Health and Human
Services (HHS) providing oversight and coordination at the federal
level.
Those eligible for Medicaid come primarily from two cash assistance
programs: Aid to Families With Dependent Children (AFDC) and
Supplemental Security Income (SSI). AFDC is the primary route by
which children and their families become eligible, while elderly,
blind, and disabled individuals become eligible primarily through
SSI. Coverage expansions since 1984 have also increased the number
of beneficiaries not linked to cash assistance payments. These
include people who are eligible for Medicare, low-income children and
pregnant women who are not receiving AFDC, and several mandatory and
optional coverage groups among disabled and elderly individuals.
About 6 million disabled individuals were covered by Medicaid in
fiscal year 1994. To qualify for SSI--and therefore for Medicaid in
most
states\3 --beneficiaries must meet certain program criteria for
disability and for maximum allowable financial resources.
Eligibility criteria center on an individual's ability to function in
daily life and the existence of a disabling diagnosis or condition.\4
In 1996, the maximum allowable income was $470 per month for an
individual and $705 for a couple.\5 States have the option of
extending eligibility to people who receive state payments that
supplement SSI benefits;\6 to some people whose incomes are above SSI
levels but who are sufficiently disabled to need institutional care;
and, with federal approval, to some people who are at risk of needing
institutional care. For 1996, the federally specified maximum income
level for an "SSI-related" individual was $1,410 per month.
More than half of all disabled people receiving SSI as of December
1994 were eligible on the basis of a mental disability. Such
disabilities included mental retardation, autism, schizophrenia,
paranoia, and, under certain circumstances, substance abuse.\7 For
those who were eligible on the basis of physical disabilities, the
main categories were diseases of the nervous system, sense organs,\8
musculoskeletal and connective tissues, or circulatory system.
Specific conditions in these categories included blindness, muscular
dystrophy, cerebral palsy, Parkinson's syndrome, brain tumors,
rheumatoid arthritis, osteoporosis, and chronic heart disease.
Children constitute about 22 percent of disabled SSI recipients.
Among them, mental retardation is the leading cause for eligibility.
Since a 1990 U.S. Supreme Court decision,\9 SSI disability criteria
for those 18 and younger have been based on developmental delays and
limitations in ability to engage in age-appropriate activities.
--------------------
\2 Besides the 50 states, 6 other entities have programs: the
District of Columbia, American Samoa, Guam, the Northern Mariana
Islands, Puerto Rico, and the Virgin Islands. For this report, we
refer to all 56 as "states."
\3 Eleven states elect to retain the more restrictive Medicaid
eligibility criteria that were in place for blind, disabled, and
elderly beneficiaries before SSI was established in 1972. These
states may use more restrictive definitions of disability or more
restrictive financial eligibility criteria than SSI.
\4 Specifically, a disabled person is one who is unable to engage in
any substantial gainful activity because of a medically determined
physical or mental impairment that is expected to result in death or
that has lasted (or can be expected to last) at least 12 months.
Eligibility for children is based on developmental delays and
functional impairment.
\5 Maximum financial assets exclusive of a home, automobile, burial
space(s), and personal effects were $2,000 for an individual and
$3,000 for a couple.
\6 In some cases, states are required to provide supplemental
payments. In February 1994, less than 0.1 percent of SSI recipients
qualified for these required payments.
\7 Drug or alcohol addiction by itself does not qualify an individual
for SSI benefits. Rather, individuals must be disabled (as defined
by SSI law) with addiction as a factor contributing to the
disability.
\8 Because blindness is included among disabilities of the sense
organs, throughout this report we will refer to blind and disabled
beneficiaries collectively as disabled.
\9 Sullivan v. Zebley, 493 U.S. 521 (1990).
DISABLED PEOPLE ACCOUNT FOR
OVER ONE-THIRD OF MEDICAID
EXPENDITURES
---------------------------------------------------------- Chapter 1:2
Many disabled Medicaid beneficiaries have a level of medical need
that is atypical of the general population. A disabled person's
degree of disability can range from mild to very severe. At the more
severe levels, individuals may be technology-dependent, requiring
medical devices to compensate for loss of a vital body function.
Many of them also require ongoing nursing care to avert death or
further disability.
Because of their atypical medical needs, disabled individuals have
medical costs that are generally higher than those of the typical
Medicaid beneficiary. In fiscal year 1994, disabled individuals were
about 15 percent of the Medicaid population and accounted for 39
percent of Medicaid expenditures, including long-term care. Table
1.1 shows an Urban Institute analysis of how expenditures for
different services in 1993 were distributed on a per-person basis.
Average total expenditures for disabled ($7,956) and elderly
individuals ($9,293) were each more than three times the spending for
other children and adults.
Table 1.1
Average Medicaid Per-Person Expenditures
by Beneficiary Group, 1993
Physic
ian, Long-
Inpati lab, X Outpat Other\ term All
Beneficiary group ent ray ient a Total care\b services
---------------------- ------ ------ ------ ------ ------ ------ --------
Disabled $2,072 $443 $773 $1,183 $4,471 $3,485 $7,956
Elderly 541 139 155 793 2,385\ 6,907 9,293\c
c,d
Other adults 805 381 304 313 2,041\ 27 2,067\c
c,d
Other children 452 159 165 203 1,116\ 74 1,191\c
c,d
--------------------------------------------------------------------------------
Note: Some data on this table are estimates made by the Urban
Institute to correct problems in data reported to HCFA by states.
\a Includes prescription drugs, case management, family planning,
dental, children's preventive services, vision, and other
practitioner care (such as therapy).
\b Long-term care includes institutional care, inpatient mental
health care, and home health services.
\c Totals do not add because of rounding.
\d Totals do not sum from the listed services because they include
payments to Medicare and prepaid health plans that cannot be assigned
to specific services.
Source: Urban Institute calculations based on HCFA data.
MANAGED CARE IS GROWING
THROUGHOUT THE MEDICAID PROGRAM
---------------------------------------------------------- Chapter 1:3
In general terms, managed care refers to a range of health care
models that use primary care practitioners to control and coordinate
the delivery of services. The best-known options are prepaid (or
"capitated") models that involve payment of a set monthly amount per
enrollee (the capitation fee) to provide or arrange for a specified
set of services.\10 Faced with rising Medicaid expenditures--the
fastest-growing portion of most state budgets--many states have begun
to incorporate managed care into their service delivery approach.
Managed care is seen as a way to help control these costs because it
discourages providers from providing unnecessary services and directs
beneficiaries to obtain care in the most cost-effective settings (for
example, obtaining primary care at a clinic rather than a hospital
emergency room).
Managed care is also seen as a way to better ensure that Medicaid
beneficiaries have access to quality care. In theory, managed care
improves access and quality by linking individual beneficiaries with
a single provider responsible for coordinating their health care
needs. Our earlier review of these efforts found that the capitated
managed care programs were succeeding, at least to some degree, in
providing the kinds of benefits for which they had been designed. We
found access to care was slightly better than in traditional
fee-for-services programs and quality was about equal between the
two.\11
In June 1995, almost 15 percent of all those who received Medicaid
services were enrolled in prepaid managed care plans.\12 Their
numbers, while still small in relationship to the total number of
beneficiaries, are growing swiftly. For example, from June 1993 to
June 1995 enrollment in capitated plans more than doubled, from 2.1
million to 5.3 million.
To date, most states have largely targeted their managed care
programs--particularly those that require enrollment in prepaid
plans--to children and adults who qualify for Medicaid through AFDC
or other programs, not toward elderly and disabled individuals.
However, states are increasingly including--or planning to
include--disabled and aged populations as well.
--------------------
\10 Other managed-care options include primary care case management
models, which are similar to traditional fee-for-service arrangements
except that providers generally receive a per capita case management
fee to coordinate the care for enrolled patients in addition to
reimbursement for each service they deliver.
\11 Medicaid: States Turn to Managed Care to Improve Access and
Control Costs (GAO/HRD-93-46, Mar. 17, 1993).
\12 The percentage enrolled would be somewhat lower if calculated on
the number of people eligible to receive services rather than those
who actually did. Data on the number of eligible people in fiscal
year 1995 were not available.
FEDERAL REQUIREMENTS GOVERN
STATE USE OF MANAGED CARE IN
MEDICAID
---------------------------------------------------------- Chapter 1:4
A key feature of state Medicaid programs historically has been
beneficiaries' freedom to choose from among participating providers.
While this freedom helped protect quality because beneficiaries who
felt the care they received was inadequate or inferior could simply
change providers, it did not guarantee that providers would be
available to treat them. Managed care approaches attempt to
guarantee access to a provider but often in exchange for some
limitation on beneficiaries' freedom of choice. The extent to which
state Medicaid managed care programs restrict beneficiaries' choice
of providers determines, in part, whether states will need to seek
approval from HCFA to waive one or more provisions of Medicaid law.
States have three options for using capitated managed care plans in
their Medicaid programs--one that requires no waiver of Medicaid
statute and two that do. Since the late 1960s, states have had the
option--with no need for a waiver--to contract with prepaid managed
care plans to deliver health care services to Medicaid beneficiaries,
provided certain conditions are met. One is voluntary enrollment:
Beneficiaries must ordinarily be permitted to disenroll at any time
and return to the Medicaid fee-for-service program.\13 Other
conditions relate to such matters as the kinds of plans that can
participate: They must be federally qualified or state-certified
HMOs,\14 have a mix of enrollment that is no more than 75 percent
Medicaid/Medicare enrollees, and engage in a range of quality
assurance activities.
The other options for using managed care exist under waiver
authorities granted to the Secretary of HHS. These authorities allow
the Secretary to waive certain statutory requirements--including the
beneficiaries' freedom to choose from among participating
providers--so that a state can develop alternative methods of service
delivery or reimbursement. These waivers are of two general
types--program\15 and demonstration.\16 Table 1.2 compares various
characteristics of the two types of waivers. Although managed care
approaches and mandated enrollment (that is, a program that requires
Medicaid participants to select among managed care approaches) can be
authorized under either type of waiver, waiving the federal
regulations concerning the types and enrollment mix of prepaid
organizations can only be done under a demonstration waiver.
Table 1.2
Comparison of Managed Care Flexibility
Under Program and Demonstration Waivers
Program waivers Demonstration waivers
---------------------------------- ----------------------------------
General characteristics
----------------------------------------------------------------------
Allows for waiver of a limited set Allows for waiver of nearly any
of Medicaid requirements provision in Medicaid law
Approval is generally based on Approval is based on the
meeting certain established discretion of the Secretary of HHS
conditions
Waivers can be renewed for 2-to 5- Generally not renewable\a
year periods
Generally used to establish More recently used to establish
primary care case management broad changes in Medicaid programs
programs and home and community-
based service programs
Characteristics pertaining to prepaid managed care
----------------------------------------------------------------------
Prepaid plans must still meet Prepaid plans may enroll Medicaid
federal requirement for 25% or patients exclusively
more private enrollment
Full range of mandatory services Benefit package may be modified\b
must be offered
Prepaid plan enrollment "lock-in" Prepaid plan enrollment "lock-in"
limited to may be extended to 12 months
1 month\c
No restrictions on access to Access to family planning
family planning providers providers may be restricted
----------------------------------------------------------------------
\a The Congress has authorized renewal of some demonstration waivers.
\b To date, only Oregon has been permitted to modify the benefits
package for traditional Medicaid beneficiaries. Other states have
been permitted to offer a modified package only to those newly
eligible for Medicaid coverage under the demonstration.
\c Lock-in is 6 months for prepaid plans meeting certain federal
requirements.
The use of prepaid managed care to provide health care for disabled
beneficiaries is also affected by the statutory requirements of other
programs besides Medicaid. Specifically, because many disabled
Medicaid beneficiaries are simultaneously eligible for one or more
other federal programs--most notably Medicare\17 --state prepaid
programs must accommodate requirements of these other programs. The
Medicare statute, in particular, contains a number of provisions that
cannot be waived and that directly affect basic features of Medicaid
prepaid care. For example, the Medicare statute requires
participating health plans to have an enrollment mix with no more
than 50 percent publicly insured enrollees, in contrast to Medicaid's
allowance for up to 75 percent publicly insured members.
--------------------
\13 An exception exists for federally qualified HMOs and certain
other federally designated organizations. After a 1-month trial
period has passed, states may restrict an enrollee's ability to
disenroll for 5 months.
\14 A limited set of other organizations also qualify, such as
certain types of federally designated community health centers.
\15 Program waivers are of several types, all of which are authorized
under section 1915 of the Social Security Act.
\16 Demonstration waivers are also known as section 1115 waivers,
after the portion of the Social Security Act that authorizes them.
\17 Medicare, authorized by title XVIII of the Social Security Act,
is a federal health insurance program that covers most people aged 65
or older, all people who receive Social Security disability benefits
for 24 months or more, and most people who suffer from kidney
failure. Medicare consists of two parts: part A, which covers
inpatient hospital, skilled nursing care, home health, and hospice
services, and part B, which covers physician and a wide range of
other services, including physical therapy.
APPLYING MANAGED CARE TO
DISABLED BENEFICIARIES POSES
ADDITIONAL CHALLENGES
---------------------------------------------------------- Chapter 1:5
Interest in using prepaid managed care programs for disabled Medicaid
beneficiaries has prompted concerns about whether this approach is
suitable to meet the needs of disabled beneficiaries. One positive
viewpoint is that disabled individuals have much to gain from managed
care because of its guarantee of access to a primary care
practitioner and its potential for coordinating an array of available
services. Improved access may particularly benefit segments of the
disabled population that have historically been unable to locate
practitioners willing to serve them. However, because prepaid plans
typically emphasize primary care, limit access to specialty care, and
carefully control the utilization of services as ways to control
costs and manage care, they are potentially disadvantageous to
certain disabled beneficiaries because of their need for extensive
services and access to a range of highly specialized providers. For
example, compared with nondisabled children in the general
population, disabled children use twice as many physician visits and
prescribed medications and five times as many other services, such as
physical therapy. Among Medicaid children, the average per-person
health care costs in 1992 were seven times higher for disabled than
for nondisabled children. Other estimates place the per-person cost
for moderately disabled individuals at two to three times the cost
for nondisabled individuals.
The "medical necessity" standards within many prepaid plans are one
example of the potential problems that disabled individuals may face,
according to advocacy groups. While not unique to prepaid care,
these standards often call for "substantial improvement" or
"restoration of function" as conditions for recommending therapies or
certain types of medical equipment. However, many disabled people
have conditions that preclude making substantial improvement or
restoring functions. Advocacy groups worry that medical necessity
standards may restrict disabled people from receiving therapy or
equipment when they need it basically to maintain their existing
level of functioning or to substitute for lost functioning. Advocacy
groups have also raised concerns about the potential for managed care
plans to disrupt the network of providers that disabled persons have
assembled over time.
Another concern that has been raised about using managed care for
disabled people is the potential effect on what could be called the
"perverse incentives" inherent in a prepaid managed care approach.\18
While incentives in a fee-for-service system may encourage a provider
to deliver too many services, prepaid programs may encourage health
plans to deliver fewer, or less expensive, services than enrollees
need, such as using a physical therapist skilled in sports medicine
rather than in specific disabilities such as spinal abnormalities.
--------------------
\18 Under prepaid managed care, plans are at financial risk--that is,
they must cover losses if the cost of providing services to enrollees
exceeds the amount received in capitation fees. Conversely, they may
keep the excess if the cost of providing services is less than the
amount received in capitation.
OBJECTIVES, SCOPE, AND
METHODOLOGY
---------------------------------------------------------- Chapter 1:6
The Chairman and Ranking Minority Member of the Subcommittee on
Medicaid and Health Care for Low-Income Families of the Senate
Committee on Finance asked us to examine (1) the extent to which
states are implementing prepaid Medicaid managed care for disabled
beneficiaries and (2) what steps states have taken to safeguard the
interests of the three major stakeholder groups--disabled
beneficiaries, prepaid health care plans, and the government--with a
focus on quality assurance and rate-setting mechanisms. On the basis
of discussions with subcommittee staff, we focused our review on the
delivery of primary and acute medical services. We also focused our
work on prepaid managed care programs--thus excluding those types of
managed care that are not risk based--because prepayment has the
potential to result in underservice to enrolled members.
To identify states with Medicaid managed care programs for disabled
beneficiaries, we reviewed HCFA documentation and interviewed
national Medicaid experts, including officials at organizations such
as the National Academy for State Health Policy and the Medicaid
Working Group. From the 17 states identified as having Medicaid
prepaid managed care programs for their disabled population,\19 we
obtained information on a wide range of topics, including
quality-monitoring activities and rate-setting methodologies. We
interviewed officials in these states to obtain their views on
problems they had encountered serving disabled individuals in prepaid
managed care plans and ways they had gone about solving them.
On the basis of what we learned about the states as a whole, we
selected three states--Arizona, Oregon, and Massachusetts--for
additional study. Arizona and Oregon have relatively long-standing
programs that provide a degree of operational experience generally
not present in other states. Massachusetts has administered for
almost 5 years a targeted program for severely incapacitated adults
that has served as a model for other state experiments. Our work in
these three states included interviewing Medicaid and other state
officials, selected providers, and advocacy groups. We obtained and
analyzed data provided by the three states, and where they were
available, we reviewed existing federal, state, and independent
studies of the programs.
During our review, we also interviewed other researchers and
knowledgeable officials and reviewed available studies of managed
care programs for disabled persons. We performed our work for this
study between November 1995 and May 1996 in accordance with generally
accepted government auditing standards.
--------------------
\19 The 17 states are Arizona, California, Colorado, Delaware, the
District of Columbia, Florida, Maryland, Massachusetts, Michigan, New
Jersey, Ohio, Oregon, Pennsylvania, Tennessee, Utah, Virginia, and
Wisconsin.
STATES ARE MOVING TOWARD MANAGED
CARE FOR DISABLED MEDICAID
RECIPIENTS
============================================================ Chapter 2
Of the 17 states that enrolled some portion of their disabled
Medicaid beneficiaries in prepaid managed care plans, enrollment
ranged from less than 1 percent to all of a state's disabled
beneficiaries. Six states have programs that are mandatory for some
or all disabled beneficiaries. Of the remaining 11 states, 3 operate
small-scale, voluntary programs focused specifically on disabled
beneficiaries; 7 allow disabled beneficiaries to participate
voluntarily wherever prepaid plans for the general Medicaid
population are available; and 1 does both.
Thirteen states exclude one or both of their more vulnerable disabled
populations--those in institutional care and those receiving home and
community-based long-term care--from prepaid plan enrollment.
However, under certain circumstances, all 17 states include one or
more groups of "dually eligible" beneficiaries, who are
simultaneously eligible for Medicaid and another federally funded
program, such as Medicare. Coordinating enrollment and other
requirements for this dually eligible group is difficult, according
to state officials.
In addition to the 17 states currently enrolling disabled
beneficiaries, more states have plans under way to include them in
prepaid managed care. Specifically, 10 additional states have waiver
proposals either approved or pending; three of these are for pilot
programs to gain experience with the approach.
17 STATES HAVE MEDICAID PREPAID
MANAGED CARE PROGRAMS FOR
DISABLED BENEFICIARIES
---------------------------------------------------------- Chapter 2:1
As of February 1996, 17 states have implemented prepaid managed care
programs for disabled Medicaid beneficiaries (see fig. 2.1). Six of
17 mandate prepaid plan enrollment for most or all of their disabled
Medicaid beneficiaries. Three states designed small-scale programs
specifically for disabled individuals in which participation is
voluntary. In seven states, disabled beneficiaries may voluntarily
enroll wherever prepaid health care plans are available for the
general Medicaid population. The remaining state operates both a
small-scale program for disabled individuals and allows beneficiaries
to enroll wherever plans are available.
Figure 2.1: States With
Prepaid Managed Care Plans for
Disabled Medicaid
Beneficiaries, February 1996
(See figure in printed
edition.)
Note: Washington, D.C. (not pictured), has implemented a targeted
voluntary program.
Table 2.1 shows the available comparative data on enrollment for
these programs. Limitations in reporting formats preclude
comparisons for two states.\20 For the five mandatory programs with
available data--Arizona, Oregon, Tennessee, Utah, and
Virginia--participation ranged from 15.2 to 100 percent of all
disabled Medicaid beneficiaries. Participation by eligible
beneficiaries in the voluntary programs targeted exclusively to
disabled individuals ranged from less than 1 percent to almost 11
percent, and participation in the remaining voluntary programs ranged
from 3 to 20 percent.
Table 2.1
Enrollment of Disabled Beneficiaries in
17 State Medicaid Prepaid Managed Care
Programs, February 1996
Percen
Total tage Year
enroll enroll enroll
Total ed in ed in ment
disabl prepai prepai by
ed d d disabl
eligib progra progra ed
State les m m began
-------------------------------------- ------ ------ ------ ------
Mandatory programs
----------------------------------------------------------------------
Arizona 64,456 56,775 88.0\a 1982
Delaware 12,198 N/A N/A 1996
Oregon\b 39,906 28,423 71.2 1995
Tennessee 138,93 138,93 100.0 1994
1 1
Utah\c 17,155 8,158\ 47.6 1982
d
Virginia 91,082 13,817 15.2 1995
\d
Voluntary programs targeted only to disabled individuals
----------------------------------------------------------------------
District of Columbia 3,200\ 8 0.25\f 1996
e
Ohio 36,000 294 0.82\h 1995
\e,g
Wisconsin 22,041 2,404 10.9 1994
\e,i
Voluntary programs for the general Medicaid population
----------------------------------------------------------------------
California 770,06 28,262 3.7 1972
7 \j
Colorado 45,042 8,842 19.6 1974
Florida N/A N/A N/A 1981
Maryland 83,350 10,496 12.6 1975
Michigan 234,51 42,373 18.1 1972
7
New Jersey 143,79 4,226 2.9 1983
3
Pennsylvania 247,90 50,443 20.4 1972
2
Voluntary program targeted to disabled individuals and voluntary
program for the
general Medicaid popul
----------------------------------------------------------------------
Massachusetts 164,36 7,935 4.8 1992
6
----------------------------------------------------------------------
Note: N/A means the state does not distinguish in enrollment and/or
eligibility reports the categories of SSI and related beneficiaries
that include aged and disabled.
\a Medicaid eligible individuals not enrolled in a prepaid plan are
Native Americans who live on reservations and who elected to receive
care from an Indian Health Service facility.
\b Oregon allows disabled beneficiaries, under certain conditions, to
receive services in managed or nonmanaged fee-for-service settings.
\c In 1995, the Utah program became mandatory in urban areas only.
Enrollment of disabled beneficiaries in the urban areas was phased in
and should be completed by July 1996.
\d Enrollment figures include both mandatory and voluntary
participants.
\e Numbers reflect those eligible to participate in the targeted
programs. See table 2.2 for more detail about which disabled
beneficiaries may enroll in each program.
\f Enrollment began in February 1996. As of March 1996, 180 children
were enrolled.
\g Program is limited to three counties.
\h Enrollment began in one county in May 1995, another in June 1995,
and the remaining county in September 1995. March enrollment for the
three counties totaled 355.
\i Program is limited to one county and enrollment is capped at
3,000, making current enrollment 80 percent of capacity.
\j Enrollment figures are somewhat understated because data from one
county do not distinguish between enrollment in prepaid and primary
care case management providers.
Sources: State enrollment and eligibility reports for February 1996.
Table 2.2 describes some basic features of the four state programs
designed for disabled beneficiaries. Two states--Ohio and
Wisconsin--began these specialized programs in selected urban
communities to gain experience before expanding their programs
statewide. The District of Columbia's program is available to all
eligible disabled children who live in the District. The fourth
state--Massachusetts--administers a prepaid program to care for
severely disabled beneficiaries but also allows disabled
beneficiaries statewide to enroll in prepaid plans.\21
Table 2.2
Description of Four State Prepaid
Programs Designed Specifically for
Disabled Beneficiaries
State Program name and description
------------------ ------------------------------------------------------------
District of The Managed Care System for Disabled Special Needs Children
Columbia is designed to reduce barriers to care faced by disabled
children and their families. Begun in February 1996, the
program contracts with a single nonprofit managed care plan
to serve disabled people 22 and younger and is designed to
move them, whenever possible, from institutions into
community settings. Each enrolled child is assigned a
primary care practitioner--usually his or her current
provider--and a case manager who develops an individualized
plan of care through a face-to-face assessment and helps
coordinate needed medical and social services, including
transportation and home adaptation. Enrollment will be
phased in over 6 months with the goal of serving a total of
about 3,000 within 2 years. Participation is voluntary and
enrollees may change primary care practitioners at any time.
One goal of the program is to build a comprehensive database
profiling each enrollee to determine if caring for children
with complex medical needs can be improved through managed
care.
Massachusetts Massachusetts contracts with three prepaid plans that focus
on care of people with severe physical disabilities or end-
stage AIDS. Combined enrollment in any given month is about
300. The program began in 1992 with a single health plan--
now called the Community Medical Alliance (CMA)--which
coupled primary care with enhanced home visits and case
management. Severely disabled enrollees are screened for
each of the following criteria: permanent triplegia or
quadriplegia; a need for personal care or other equivalent
assistance to maintain independent living; and one of
several specified diagnoses, such as spinal cord injury,
cerebral palsy, or end-stage muscular dystrophy. AIDS
enrollees must meet the clinical criteria of end-stage AIDS.
At CMA medical care is provided by a clinical team of
physicians and nurse practitioners, with each patient
assigned to a nurse practitioner. Care is provided in the
most appropriate setting--most often the patient's home--as
an alternative to specialty and hospital care.
Ohio The Accessing Better Care program began enrolling physically
disabled and chronically ill beneficiaries under age 65 in
three metropolitan areas in 1995. It offers a flexible
benefits package that includes home and community-based care
as alternatives to institutional care whenever possible and
currently enrolls about 300 of the 36,000 eligible
beneficiaries. Each area has its own prepaid health care
plan, which is a partnership between an HMO and an academic
medical center. Care is delivered by an interdisciplinary
care team led by a social worker or nurse case manager.
Individual care plans for each enrollee are developed from
initial assessments. Specialists play active roles on the
care teams.
Wisconsin Designed for disabled beneficiaries over the age of 15, I-
Care began operation in 1994 and is a joint venture between
a rehabilitation center and an HMO. It currently operates in
one metropolitan county and will eventually serve 3,000
beneficiaries. Care coordinators assess initial medical and
social needs through an in-home visit, develop an individual
care plan jointly with providers and social workers, and
assist beneficiaries and their families in selecting and
accessing providers. Care coordinators are nurses or social
workers knowledgeable about disabilities. All prepaid plan
staff are trained in working with people with disabilities.
Beneficiaries may select from clinic sites throughout the
city. Whenever possible, enrollees' existing primary care
practitioners are invited to join the provider panel.
Wisconsin hopes to expand the program to two additional
counties in 1997.
--------------------------------------------------------------------------------
Six states--Arizona, Delaware, Oregon, Tennessee, Utah, and
Virginia--mandate prepaid plan enrollment for some or all of their
disabled beneficiaries.
-- Arizona, Delaware, Oregon, and Tennessee mandate prepaid
enrollment under demonstration waivers for all Medicaid
beneficiaries. Oregon allows beneficiaries, in concert with
their social service case workers, to select fee-for-service
care (either managed--called primary care case management--or
nonmanaged) when prepaid care does not best meet their health
care needs. As a result, enrollment of disabled beneficiaries
in Oregon is about 71 percent, compared with 100 percent in
Tennessee.\22 In Arizona, Native Americans who live on
reservations may elect to receive health care from either a
prepaid plan or Indian Health Service facilities.
-- Utah and Virginia mandate prepaid enrollment in selected areas
under the program waivers they received from HCFA.
The extent to which these six states with mandatory enrollment adapt
their managed care programs specifically for disabled beneficiaries
is further discussed in chapters 3 and 4.
--------------------
\20 These two states--Delaware and Florida--do not disaggregate SSI
and related categories to distinguish among aged, blind, and disabled
beneficiaries.
\21 Two of the four states also have an approved or pending
demonstration waiver that would require some or all disabled
beneficiaries to participate in prepaid managed care. The District's
program, which is an approved demonstration, is voluntary.
\22 Demonstration waivers allow states the flexibility to determine
health care delivery systems for specified beneficiaries, including
giving some a range of managed care alternatives while keeping others
in fee-for-service care. Consequently, a state with "mandatory"
prepaid enrollment may have certain beneficiaries in other settings.
For example, Oregon designated primary care case management as an
acceptable managed care alternative.
MOST STATES EXCLUDE THE MORE
VULNERABLE FROM MANAGED CARE
---------------------------------------------------------- Chapter 2:2
Medicaid covers care for two types of more severely disabled
individuals--those in institutional care\23 and those receiving home
and community-based long-term care.\24 People meeting these criteria
are at least partially unable to care for themselves because of an
injury, illness, or other disabling condition. The range of services
they need extends beyond primary and acute medical care to include
assistance with everyday activities, such as dressing and using the
bathroom, that the individual cannot do independently because of his
or her disability. Such services include personal care attendants,
homemaker services, adult day care, and respite for family
caregivers.
Thirteen of the 17 states exclude one or both of these vulnerable
populations (see table 2.3). More specifically, 12 exclude disabled
beneficiaries who reside in institutions, and 10 exclude those
receiving home and community-based long-term care. Under
demonstration waivers, two states--Arizona and Tennessee--mandate
prepaid plan enrollment of these populations for their primary and
acute care needs but provide long-term care under separate
arrangements. Specifically, long-term care in Tennessee remains
fee-for-service and in Arizona is coordinated by a single
contractor--typically a state or county agency--for each county. In
Oregon, residents of institutions for the mentally retarded and the
mentally ill are not enrolled in prepaid plans, while nursing home
residents are.
Table 2.3
Extent to Which 17 States Include
Severely Disabled Beneficiaries in
Medicaid Prepaid Care Programs, February
1996
Home and community-
Institutional based services
populations participants
State included?\a included?\b
---------------------- ---------------------- ----------------------
Mandatory programs
----------------------------------------------------------------------
Arizona NF, ICF/MR, IMD Yes
Delaware No No
Oregon NF Yes
Tennessee NF, ICF/MR, IMD Yes
Utah No Yes
Virginia No No
Voluntary programs targeted only to disabled individual
----------------------------------------------------------------------
District of Columbia NF, ICF/MR No
Ohio No No
Wisconsin No No
Voluntary programs for the general Medicaid population
----------------------------------------------------------------------
California No No
Colorado NF Yes
Florida No No
Maryland No No
Michigan No No
New Jersey No No
Pennsylvania No Yes
Voluntary program targeted to disabled individuals a program for the
genera population
----------------------------------------------------------------------
Massachusetts No Yes
----------------------------------------------------------------------
\a Institutionalized beneficiaries include residents of nursing
facilities (NF), intermediate care facilities for the mentally
retarded (ICF/MR), and institutions for mental diseases (IMD).
\b Home and community-based services programs provide a broad range
of services to beneficiaries who, in the absence of such services,
would require care in Medicaid-covered institutions. Beneficiaries
these programs serve include disabled people who might need care in a
nursing facility and those who are developmentally disabled or
mentally retarded who might need care in an ICF/MR.
Of the 17 states, only the District of Columbia includes long-term
care in the set of services covered by capitation payments to health
plans.\25 However, a few such programs have existed on a small scale
since the 1980s, and HCFA is currently reviewing Colorado's request
to implement a pilot program in one county. Integrating primary and
acute care into a single prepaid contract with long-term care
presents certain challenges. Among them are the lack of generally
accepted standards regarding the use of various long-term care
services; prepaid plans' lack of experience providing long-term care;
the potential for the demise of existing community-based providers
with experience in delivering such care; and the difficulty in
establishing adequate rates for the combined set of services.
Concerns about integrating the two types of care include the
potential for medically based prepaid plans to emphasize medical
technology or institutional care over the social and supportive
services that many beneficiaries prefer. In addition, integration
raises concerns about who should perform care needs assessments and
case management services--state or prepaid plan staff--given the lack
of recognized standards for appropriate long-term care and the fact
that in such integrated arrangements a single provider is responsible
for major portions of an individual's life needs.
--------------------
\23 Institutional care in Medicaid refers to care delivered in
nursing facilities, intermediate care facilities for the mentally
retarded (called ICF/MR), and, more limitedly, in institutions for
those with mental diseases.
\24 Home and community-based care is made available, with HCFA
approval, to Medicaid beneficiaries who, in the absence of such
services, would be likely to require care in a Medicaid-covered
institution.
\25 Long-term care is included because the District's program is
designed to help disabled children make the transition from
institutional to community-based care. Arizona's long-term care
program is primarily operated by state and county agencies that
subcontract to separate providers for long-term care and for primary
and acute care services, although the Medicaid program does not
require them to do so.
ENROLLMENT OF DUALLY ELIGIBLE
INDIVIDUALS CREATES CHALLENGES
---------------------------------------------------------- Chapter 2:3
Another consideration for states with Medicaid managed care
initiatives is whether to include beneficiaries who are also eligible
for medical services or supplies through another federal program.
For Medicaid beneficiaries, these programs fall into two
categories--Medicare and title V and related school-based programs.
Medicare is a federal health insurance program that covers, among
others, all people who have received Social Security disability
benefits for 24 months or longer. Medicare and Medicaid provide
essential and complementary services to dually eligible
beneficiaries. For example, Medicare is the primary provider of
inpatient and physician care, while Medicaid generally provides
prescription drugs. Some estimate that about one-third of disabled
Medicaid beneficiaries nationally are also covered by Medicare, but
proportions will vary from state to state. For example, Oregon
officials estimate that 45 percent of disabled beneficiaries are also
covered by Medicare.
Title V of the Social Security Act authorizes state programs to
improve the health of mothers and children, including children with
special health needs. These programs, which are limited in scope and
vary among states, provide and promote state and community-based
systems of services for children with special health needs and
typically serve children from low- and moderate-income families.
Such programs arrange for initial assessments, service plans,
outpatient specialty physician services, and therapies and care
coordination for children with various chronic conditions. Disabled
children may receive various therapies and assistive
equipment--speech therapy and wheelchairs, for example--funded from
title V as well as from schools that must assure children access to
certain medical services that allow them to participate in school.
In many cases, these services and equipment are also covered by
Medicaid.
The often conflicting or overlapping requirements of Medicaid and
other programs, particularly Medicare, have been cited as a barrier
to including dually eligible beneficiaries in mandatory prepaid
managed care programs. In general, state officials cited the
inflexibility of Medicare rules as a deterrent to developing a
Medicaid prepaid program that includes those dually eligible for
Medicare. Including those eligible for other programs gives rise to
the need for negotiations and extensive coordination between the
Medicaid staff, HCFA, and representatives from plans and other
agencies serving those beneficiaries. Table 2.4 provides examples of
barriers states encounter when attempting to include dually eligible
beneficiaries.
Table 2.4
Examples of Barriers States Face in
Including Dually Eligible Beneficiaries
in Medicaid Prepaid Managed Care
Programs
Program Barrier
------------------ ------------------------------------------------------------
Medicare Plans may be unwilling to participate in Medicaid prepaid
programs if the addition of dually eligible beneficiaries
threatens to raise their percentage of publicly funded
enrollees above 50 percent, which would disqualify them from
Medicare participation.
Medicare rules regarding surrogate decisionmakers--those
allowed to make decisions for people not able to make their
own--are more restrictive than those of Medicaid, thereby
complicating prepaid plan enrollment of dually eligible
individuals and affecting who may aid them in selecting a
plan.
If individuals want to join the same prepaid plan for both
their Medicare-and Medicaid-covered services, timing
differences between the two programs may require them to
remain in Medicare fee-for-service care for up to 2 months
after they have enrolled in the Medicaid prepaid plan.
Title V programs These programs for children with special health care needs
are typically administered in states by departments of
health, which are often separate from Medicaid agencies.
Including these children requires significant interagency
coordination.
School-based Conflicts surrounding medical equipment such as wheelchairs
programs or devices to help overcome communicational impairments
arise when children receive services through prepaid plans
and through school-based programs. In some cases, schools
and plans cannot easily agree on whether the equipment is
"medically" or "educationally" necessary. In others, schools
or plans restrict the use of the equipment to either the
classroom or the home, potentially leading to the need to
duplicate services.
--------------------------------------------------------------------------------
The nature and extent of coordination barriers between Medicare and
Medicaid vary depending on the extent to which states require prepaid
plan enrollment and the extent to which Medicare prepaid plans are
available. Coordination issues are lessened when dually eligible
beneficiaries remain in Medicare fee-for-service care and join a
Medicaid prepaid plan for services not covered by Medicare. But
coordination issues increase in states where beneficiaries are
required to enroll in Medicaid prepaid plans and are to be "locked
in" for specified periods. Chief among these are Medicare's
requirement that beneficiaries are free to choose a prepaid plan or
to use fee-for-service care and, when in a prepaid plan, are allowed
to disenroll at will. As a result, in states restructuring their
Medicaid programs under waivers, the potential benefits of
coordinated care may elude Medicaid beneficiaries dually eligible for
Medicare who may not be enrolled in a single managed care plan for
both sets of services or have a single primary care provider--which
undermines one goal of a prepaid program.
Although many of the state and federal officials we interviewed
described coordination of these programs as a difficult process, most
states with prepaid programs for disabled Medicaid beneficiaries are
extending enrollment to the dually eligible. In all 17 states, one
or both of these groups of dually eligible individuals may elect to
enroll in prepaid plans (see table 2.5). For beneficiaries with
Medicare eligibility, 12 states open their Medicaid prepaid programs
to participation and 5 do not. Three of the 12 states reported
allowing enrollment only if Medicare services were obtained in a
fee-for-service setting. Another six states reported that
coordination between the two programs was not a major issue because
the state had no or very few Medicare prepaid plans available to
beneficiaries.
Table 2.5
Extent to Which 17 States Include Dually
Eligible Beneficiaries in Medicaid
Prepaid Care Programs, February 1996
Medicare populations Title V populations
State included? included?
---------------------- ---------------------- ----------------------
Mandatory programs
----------------------------------------------------------------------
Arizona Yes Yes
Delaware No Yes
Oregon Yes Yes
Tennessee Yes Yes
Utah Yes Yes
Virginia No Yes
Voluntary programs targeted only to disabled indiv
----------------------------------------------------------------------
District of Columbia No Yes
Ohio No Yes
Wisconsin Yes Yes
Voluntary programs for the general Medicaid popul
----------------------------------------------------------------------
California Yes Yes
Colorado Yes Yes
Florida Yes No
Maryland Yes Yes
Michigan Yes No
New Jersey Yes Yes
Pennsylvania Yes Yes
Voluntary program targeted to disabled individuals a voluntary program
for Medicaid population
----------------------------------------------------------------------
Massachusetts No Yes
----------------------------------------------------------------------
MORE STATES MOVING TOWARD
INCLUDING DISABLED
BENEFICIARIES
---------------------------------------------------------- Chapter 2:4
Current signs point to increasing movement in the direction of
prepaid managed care for disabled beneficiaries, including greater
reliance on managed care programs in which their participation is
mandatory. Table 2.6 lists additional states with approved and
pending demonstration waivers that include disabled beneficiaries.
Table 2.6
Status of Additional Approved and
Pending Demonstration Waivers Including
Disabled Beneficiaries as of February
1996
Date Date
State submitted approved Prepaid enrollment Current status
------------ ------------ ------------ ------------------ ------------------
Approved and implemented for groups other than disabled individuals
--------------------------------------------------------------------------------
Minnesota July 1994 Apr. 1995 Four mandatory Will enroll
pilot programs for disabled
disabled individuals in
individuals 1997
Oklahoma Jan. 1995 Oct. 1995 Mandatory (urban Will enroll
providers will be disabled
prepaid) individuals in
1997
Vermont Feb. 1995 July 1995 Mandatory Will enroll
disabled
individuals in
1997
Approved, pending implementation
--------------------------------------------------------------------------------
Kentucky June 1995 Oct. 1995 Mandatory Implementation
planning under way
Massachusett Apr. 1994 Apr. 1995 Mandatory Awaiting state
s legislative
approval
Ohio Mar. 1994 Jan. 1995 Mandatory Will enroll
disabled at a
future
undesignated date
Pending
--------------------------------------------------------------------------------
Alabama July 1995 Mandatory--a one- Under HCFA review
county pilot
Illinois Sept. 1994 Mandatory (urban Under HCFA review
providers will be
prepaid)
Louisiana Jan. 1995 Mandatory Financing plan
disapproved
Missouri June 1994 Voluntary pilot Under HCFA review
program for
disabled
New York Mar. 1995 Mandatory--to Under HCFA review
include disabled
in year 2
Texas Sept. 1995 Mandatory with Under HCFA review
managed fee-for-
service option
Utah July 1995 Mandatory if Under HCFA review
income lower than
100 percent of
federal poverty
level
--------------------------------------------------------------------------------
In addition to Arizona, Delaware, Oregon, and Tennessee, which
currently mandate prepaid plan enrollment for virtually all their
disabled beneficiaries, six other states have received approval to
require Medicaid beneficiaries, including those who are disabled, to
enroll in prepaid managed care plans.\26 All but two of the six are
statewide programs, and four of the six will enroll disabled
beneficiaries 1 or more years after enrolling other beneficiaries.
Six of the seven states seeking to include disabled beneficiaries in
their prepaid programs have sought approval to require prepaid plan
enrollment by most or all of their Medicaid beneficiaries, including
those who are disabled. Of the seven, three--including Utah, which
currently mandates enrollment in three urban areas--have sought
approval for statewide mandatory programs. Hawaii, which currently
mandates enrollment for low-income families under a demonstration
waiver approved in 1993, intends to seek approval to include disabled
beneficiaries in the near future. Maryland and New Jersey are
currently discussing their proposals with HCFA officials.
--------------------
\26 One of these six--Ohio--is currently experimenting with a
voluntary targeted prepaid care program for disabled beneficiaries in
three counties.
QUALITY ASSURANCE EFFORTS FOCUSED
ON CARE FOR DISABLED BENEFICIARIES
ARE CONCENTRATED IN A HANDFUL OF
STATES
============================================================ Chapter 3
Enrolling disabled Medicaid beneficiaries in prepaid managed care
heightens the need for states to ensure the quality of care provided.
Prepaid managed care, especially when participation is mandatory,
diminishes beneficiaries' ability to "vote with their feet" by
changing plans or physicians when they are unhappy with their care.
Given disabled people's often extensive need for care, states need an
adequate set of mechanisms both to address fears and uncertainties
about receiving care in a managed setting and to ensure that health
plans are meeting their commitments.
Important aspects of states' quality assurance activities can fall
into two main categories: (1) building safeguards into the programs
through adequate planning and consensus-building and (2) tailoring
various aspects of the program (such as enrollment and monitoring) to
meet the specific needs of disabled individuals. To date, most of
the efforts have been made by five states with mandatory
participation by disabled beneficiaries (Arizona, Delaware, Oregon,
Utah, and Virginia) or by four states with programs targeted
exclusively for disabled beneficiaries on a voluntary basis
(Massachusetts, Ohio, Wisconsin, and the District of Columbia).
Among other things, their initiatives include requiring plans to
designate advocates or case managers for disabled beneficiaries and
to include access to specific types of specialty providers, and
developing encounter data and quality-of-care standards for
evaluating their managed care programs for disabled beneficiaries.
The remaining states (mainly those with voluntary programs for the
general Medicaid population) are primarily relying on broadly scoped
monitoring actions that may not be sufficient to detect problems with
care provided to specialized groups such as disabled individuals.
ADEQUATE PLANNING AND
CONSENSUS-BUILDING
---------------------------------------------------------- Chapter 3:1
Adequate planning and consensus among all the affected
parties--health plans, disabled beneficiaries (and their advocates),
and state officials--are critical for the development of and
transition to an effective Medicaid managed care program, according
to officials in the nine states furthest along in tailoring their
programs. In particular, they stressed the need to involve
beneficiaries and advocates in planning and program design. In two
of the three states we visited (Massachusetts and Oregon), advocates
and state officials who work with disabled beneficiaries cited the
use of consensus meetings, which often involved health plan
management and medical staff, as key to the smooth transition to
managed care.
Oregon's experience is an example of how these consensus meetings
worked. For more than a year before bringing disabled beneficiaries
into managed care, Oregon's Medicaid staff held weekly meetings with
health plan representatives, beneficiary representatives, and state
social service agencies (from whom most disabled residents received
case management services). These meetings covered such topics as
building a common set of definitions for terms like "case management"
and "case workers"--terms each group routinely used with different
meanings. The need to arrive at such definitions was not unique to
Oregon: An official in another state said coordination meetings were
needed to define "disabled" because health plans anticipated diabetic
or asthmatic enrollees, not quadriplegics or other individuals with
medically complex needs.
Officials in the three states also noted the importance of ongoing
meetings among stakeholders to address issues as they arose. They
said that once the programs had been implemented, Medicaid staff met
routinely with health plan management, medical directors, and
advocacy and social service agency representatives to discuss such
issues as rates, data reporting, and matters related to health care.
These groups sometimes formed subcommittees to study specific problem
areas. For example, in Oregon and Arizona the Medicaid and health
plan medical directors have subcommittees to develop practice
guidelines and study issues concerning disabled children. To date
the state and health plan medical directors in Oregon have adopted
practice guidelines for preventive care, cerebral palsy, spina
bifida, and cleft palate. Guidelines for cystic fibrosis, Down's
syndrome, pediatric asthma, and sickle cell disease are being
developed. One of Arizona's subcommittees evaluates new treatments
and technologies; it granted approval for the use of certain
prescription drug treatments for cystic fibrosis and multiple
sclerosis.
TAILORING PROGRAMS SPECIFICALLY
TO MEET THE NEEDS OF DISABLED
BENEFICIARIES
---------------------------------------------------------- Chapter 3:2
The same nine states have taken action in a number of other ways to
better adapt aspects of their managed care programs to address the
concerns of disabled beneficiaries. These actions include addressing
concerns about disabled individuals' ability to continue seeing
established caregivers, helping disabled individuals and their
families decide which plan to select, providing access to a range of
available services, and monitoring the quality of services provided.
RESOLVING CONCERNS ABOUT
MEDICAL NECESSITY
-------------------------------------------------------- Chapter 3:2.1
Among the most important issues regarding access to services that
need resolution is how the concept of "medical necessity" will be
applied in prepaid care situations involving disabled enrollees.
Definitions vary widely in their sensitivity to the needs of disabled
individuals: Some include the need for improvement or restoration of
function within a specified number of treatments or time period
(often 60 days), while others include consideration of preventing the
progression of adverse health conditions or the cost-effectiveness of
the treatment.
The strict application of a narrow definition of medical necessity
can conflict with disabled enrollees' needs, particularly in the case
of services that offer little hope for improvement but can help to
maintain existing quality of life. For example, people with
neuromuscular disabilities may need physical therapy to prevent
deterioration and reduce discomfort even when restoration or
functional improvement is not possible. Conversely, state officials
also pointed out that, in applying the concept of medical necessity,
health plans' flexibility can provide an opportunity for them to
supply services over and above those available in the fee-for-service
program. For example, in one state a child received a
technologically enhanced bed (not covered under the state's
fee-for-service program) because health plan officials decided the
bed was likely to reduce hospitalizations for pressure sores and
infections.
The three states we visited address concerns about medical necessity
primarily through the appeal process, giving the medical director of
the Medicaid program authority to overturn health plan decisions
regarding what is medically necessary for an individual recipient.
However, advocates for disabled individuals said reliance on the
complaint and grievance process puts an undue burden on beneficiaries
because (1) the process requires a significant amount of
self-advocacy on the part of beneficiaries who may not be capable of
it and (2) the process can be extremely time-consuming.
Some states are beginning to include a definition of medical
necessity in health plan contracts and to supplement this definition
with guidance on or monitoring of its application. For example,
Arizona approached the issue by including the concept of
"habilitation" (the extent to which treatment helps to maintain a
recipient's current ability to function) in its monitoring of health
plan services. Similarly, Oregon issued guidance for health plans to
use in approving various therapies and equipment emphasizing such
nonmedical outcomes as enhancement of independent living.
ADDRESSING CONCERNS THROUGH
ENROLLMENT AND RELATED
PROGRAMS
-------------------------------------------------------- Chapter 3:2.2
Among states that offer prepaid managed care to disabled
beneficiaries, only those with mandatory enrollment have significant
percentages of their populations participating. The low
participation in other states may reflect, among other things,
concerns of disabled individuals about relying on a prepaid care
system. When prepaid plan enrollment can be required of
beneficiaries, state decisions about enrollment--such as who will
enroll recipients, what sorts of education programs will be involved,
and how beneficiaries will be assigned to a health plan if they do
not choose one--become more prominent among the concerns of disabled
beneficiaries, according to advocates in the states we visited.
States we contacted reported using various approaches to enrollment,
assignments, and exceptions to remain in the fee-for-service system.
CLIENT ENROLLMENT
------------------------------------------------------ Chapter 3:2.2.1
Some states view the process of enrolling beneficiaries in Medicaid
managed care programs as an important opportunity to educate and
counsel beneficiaries--sometimes individually--about both managed
care and the need to choose from among participating health plans.
The three states we visited generally applied many of the steps they
use for other beneficiaries when they enrolled disabled
beneficiaries.\27 Other steps included the following:
-- Oregon sends disabled beneficiaries (1) a booklet that the state
developmental disabilities council created to educate
beneficiaries about managed care and (2) a chart comparing the
features of available health plans. The booklet contains
worksheets to help beneficiaries identify their health care
needs and detail their existing provider network so that they
can better select an appropriate health plan.
-- In Massachusetts advocates were concerned that managed care
might disrupt the existing provider networks from which disabled
beneficiaries receive care.\28 State staff and advocates adopted
a health needs assessment that enrollment staff use to help
beneficiaries select existing or other appropriate providers.
The state also adopted a more flexible approach allowing
specialists to serve as primary care providers for their
disabled patients and allowing beneficiaries to enroll with
providers outside their geographic service areas.
Although each state makes educational materials available, state
staff familiar with cognitively impaired beneficiaries in two states
reported that written materials were seldom helpful to this
population. Instead, for these beneficiaries, education largely
occurs through the one-on-one relationships between case workers and
beneficiaries and their families.
To ensure that needed equipment and supplies are provided without
interruption while a beneficiary's enrollment is processed and a
primary care provider is selected, Oregon also developed a process to
inform the selected plan in advance about an individual's health care
needs. "Continuity of care referral" forms alert prepaid plans to
life-sustaining, ongoing treatment needs of individuals enrolling.
Social service agency case workers, who enroll disabled beneficiaries
in the prepaid plans they select, complete a form for each individual
with life-sustaining needs--such as oxygen supplies--and forward it
directly to the plan's care coordinator. Delaware and Virginia also
require plans to either maintain existing plans of care or develop
transition plans for people with ongoing care needs.
--------------------
\27 Examples of such steps include sending beneficiaries materials
informing them of their rights and responsibilities under managed
care, sending summary information about participating health plans,
and making materials prepared by health plans available for
beneficiary consideration. In these three states, the materials sent
to beneficiaries contained lists of participating clinics and
hospitals but not individual primary care practitioners (except in
one instance) or specialty care practitioners. However, in all three
states, beneficiaries could contact state enrollment staff or
participating health plans to obtain primary care practitioner lists.
\28 Massachusetts' program waiver requires all
beneficiaries--including disabled beneficiaries--to select either a
prepaid plan or a primary care case manager. In October 1995, HCFA
approved state officials' request to begin assigning beneficiaries to
a prepaid plan.
EXCEPTIONS
------------------------------------------------------ Chapter 3:2.2.2
States with mandatory prepaid plan enrollment face decisions about
whether, or if, to allow exceptions for certain beneficiaries to
receive fee-for-service care. Delaware, Tennessee, Utah, and
Virginia--four of the six states with mandatory programs--essentially
do not allow exceptions, though individuals may receive care in a
fee-for-service setting for a short time while eligibility and
enrollment forms are processed. In Arizona, Native Americans living
on reservations may elect not to enroll in prepaid care. Oregon
allows case workers, in concert with beneficiaries and their
families, to decide whether prepaid managed care is the best delivery
system. Most exceptions involve an ongoing relationship with a
practitioner who does not participate in any of the managed care
networks, while others are for situations in which the beneficiary is
involved in an ongoing treatment regime or when changing
practitioners could seriously harm the individual.\29 Some exceptions
are short-lived, delaying enrollment by up to 1 year; others may be
permanent. Oregon Medicaid officials monitor exceptions granted by
case workers to determine, among other things, whether any trends
develop.
--------------------
\29 Oregon also grants exceptions for religious or Indian heritage
reasons.
ASSIGNMENT
------------------------------------------------------ Chapter 3:2.2.3
States with mandatory enrollment programs for broader segments of the
population, such as low-income families, often develop systems that
automatically assign beneficiaries to a health plan if they do not
select one for themselves. This assignment is usually based on
geographic proximity. The three states we visited each took a
different approach to assigning disabled enrollees. Oregon decided
against assigning disabled beneficiaries, relying instead on each
social services case worker selecting a plan in consultation with
beneficiaries or their families. In Arizona, severely disabled
recipients who receive both acute and long-term care do not have a
choice among providers because only one contract is awarded in each
county. Less severely disabled beneficiaries are assigned to plans,
when they do not choose from among those available, based on results
of the state's competitive bidding process, with the lowest-cost
plans receiving proportionately more assignments. In Massachusetts,
nurse review panels analyze beneficiaries' claims histories to
determine the health needs of those who do not choose health plans so
that assigned plans are more likely to have the capacity to meet
their needs. Enrollment staff try to contact beneficiaries after
assignment to ask about their satisfaction and help them select
another provider if the relationship is unsatisfactory.
ACTIVE MANAGEMENT OF A
DISABLED BENEFICIARY'S CARE
-------------------------------------------------------- Chapter 3:2.3
Most of the nine states furthest along in tailoring their programs
have also taken steps to ensure that disabled beneficiaries receive
coordination of services through a process called "case management."
The four small-scale programs aimed exclusively at disabled
individuals have included specific requirements for case management
services. Examples of such requirements are (1) coordination of a
range of needed services, such as transportation, community support
services, and primary and specialty care; (2) development of
individual plans of care that ensure continuity and coordination of
care among various clinical and nonclinical settings; and (3)
stipulation that case managers be social workers or nurses with
specific training in working with disabled people. In Wisconsin's
program, if case managers do not perform a needs assessment within 55
days of enrollment, the beneficiary is automatically disenrolled from
the program.
Among the mandatory programs, Arizona and Delaware have also taken
steps regarding case management. Arizona's program for the more
severely disabled requires contractors to provide case management
services, although no such requirement exists for health plans
serving the less disabled. Arizona requires case managers to perform
a needs assessment within 15 days and complete a plan of care within
30. Failure to meet these requirements can result in a financial
penalty. For the severely disabled, Arizona also established maximum
ratios of 1 case manager to 40 community-based enrollees and 1 case
manager to 120 institutionalized enrollees. Delaware requires plans
to provide case managers for disabled children. These case managers
visit children in their homes to assess the children's needs in
concert with their families.
REQUIRING PLANS TO PROVIDE
STAFF ADVOCATES OR ACCESS TO
SPECIFIC SPECIALTIES
-------------------------------------------------------- Chapter 3:2.4
Several other actions are similar to case management in that they are
directed at ensuring that disabled enrollees receive appropriate
care. One of these actions is a requirement for a "designated
advocate." Oregon and Wisconsin require health plans to have
designated contact staff available for disabled enrollees and their
families. These staff, called "exceptional needs care coordinators"
in Oregon and "Medicaid advocates" in Wisconsin's targeted program,
function as advocates for enrolled beneficiaries and must meet
specific requirements for experience or training in working with
those who are disabled. Oregon also created a state-level ombudsman
to serve as a contact point for disabled beneficiaries and to help
coordinate the activities of the plan-based care coordinators.
Advocates for disabled individuals in Oregon said the coordinators
perform a vital role in educating health plans on appropriate care
for disabled enrollees.
Another way in which states can help ensure access to appropriate
care is to require health plans to provide enrollees access to
specific specialty services. For example, Utah requires prepaid
plans that serve disabled children to provide timely access to
pediatric subspecialty consultation and care, and rehabilitative
services from professionals with pediatric training. For selected
disabling conditions such as spina bifida and cerebral palsy,
children must have timely access to coordinated multispecialty
clinics for their disorder. Delaware requires health plans to
consider disabled enrollees' requests for specialists to serve as
primary care practitioners, including requests for specific
specialists not affiliated with the prepaid plan. Denials of these
requests may be appealed to the Medicaid agency. As of April 1996,
two requests for pediatric specialists had been granted, both of
which were for specialists not affiliated with the plan. Prepaid
plan response to such requests is included in periodic state
monitoring.
MONITORING PLANS FOR
COMPLIANCE WITH CONTRACTUAL
REQUIREMENTS
-------------------------------------------------------- Chapter 3:2.5
Another category of quality assurance mechanisms is the compliance
monitoring normally performed for all Medicaid prepaid care plans.
This monitoring helps to assure the state that health plans are
delivering the health services they are paid for and doing so in
accordance with state and federal requirements. However, most of
this monitoring activity is not specifically targeted to any
eligibility group. Without some form of adjustment (a step some
states are beginning to take), this monitoring will have limited
effectiveness in systematically identifying problems that disabled
beneficiaries may be having with their care.
Federal regulations promulgated by HCFA are the basis for much of the
monitoring activity. They require, for example, that prepaid plans
allow enrollees, to the extent possible, to choose their health
practitioners and maintain a program that allows enrollees to voice
complaints and provides for speedy resolution. States may establish
performance measures to determine compliance with federal access
standards. For example, as measures of access to care, states may
set standards for time frames for linking enrollees with primary care
practitioners, waiting times for scheduled appointments, enrollee
travel time to a provider, and capacity ratios of providers to
enrollees. The following are additional examples of HCFA
requirements for access and quality:\30
-- Health plans must offer enrollees health services comparable to
those available for non-enrolled beneficiaries in the same
locale.
-- Health plans may not discriminate against enrollees on the basis
of their health status or need for health services.
-- Health plans may not terminate enrollment because of an adverse
change in the enrollees' health.
-- Emergency services must be available 24 hours a day, 7 days a
week.
-- The state must provide for annual external reviews conducted by
an independent reviewer.
-- Health plans must maintain an internal quality assurance
program.
States typically monitor compliance with these and other state and
federal requirements through periodic (usually annual) site visits
and reviews of health plan policies and procedures. Some data, such
as disenrollments and complaints and grievances data, are collected
and reviewed quarterly. In addition to these requirements, some
states survey enrollees periodically to determine their level of
satisfaction with the care received from participating plans.
Monitoring activities specifically related to disabled enrollees were
limited. Of the 17 states with prepaid programs that include
disabled Medicaid beneficiaries, 9 reported no specific monitoring
efforts designed to assess quality of and access to care for this
population.\31 For example, only those states with programs targeted
specifically to disabled beneficiaries analyzed complaints and
grievances by eligibility category to learn the views of disabled
enrollees. Also, two states we visited reported using the results of
their monitoring efforts to apply incentives and sanctions to
influence health plan behavior. However, neither state had
encountered treatment of disabled enrollees that would warrant the
use of sanctions.
Among those nine states without specific monitoring efforts for
disabled enrollees, there was acknowledgment that more needed to be
done. Officials in eight of these states said more focused efforts
would be needed as more disabled beneficiaries enrolled in prepaid
plans. For example, these states currently rely heavily on disabled
recipients' freedom to disenroll from or transfer among prepaid plans
and the existence of a complaint and grievance program. However,
disabled beneficiaries may choose to disenroll rather than complain
about the care they receive and, even if they complain, their
concerns may be masked by a low overall complaint rate for all
eligibility categories unless complaints are analyzed by eligibility
group. Thus, without more focused effort, such measures will not
reveal systemic problems in care for disabled enrollees.
Current monitoring programs do have the potential to provide more
information about care delivered to disabled beneficiaries, as the
following examples indicate:
-- States could extend their current efforts to assess specific
aspects of health care delivery to disabled enrollees. Some
states that do not assess care for disabled individuals do
conduct assessments of maternal and child care. States that
conduct reviews of prenatal, well child, or asthma care could
require studies of care for specific disabling conditions
present among the plan's enrollees. For example, a
Massachusetts prepaid plan, on its own initiative, studied the
management of pressure sores, a common cause of hospitalization,
as an area of quality improvement. One result was the
development of a variety of methods, including new screening
protocols for earlier intervention and an accelerated schedule
for wheelchair seating evaluations, to further improve care.
-- Consumer surveys could include questions about eligibility
status, and samples could be designed to ensure that sufficient
numbers of disabled beneficiaries were included. States could
also request disenrollment or utilization data reported by
eligibility category to allow comparisons with other eligibility
groups or across health plans. States might, as one reported,
interview individuals requesting disenrollment to gather more
in-depth information about the care received. In 1996, Virginia
will conduct a survey of all disabled beneficiaries who
disenrolled during the year.
-- States have great flexibility in deciding how to structure
required external reviews, which represent an opportunity for
closer scrutiny of issues facing disabled individuals. Oregon,
the District of Columbia, and Virginia are seeking proposals
from external professional review contractors for studies
specifically designed to measure the quality of care for
disabled enrollees.
The steps taken to monitor plans once they are up and running need
not be limited to modifying existing Medicaid oversight requirements.
Targeted quality-of-care studies and quality improvement goals have
been instituted by one or more of the eight states.
--------------------
\30 Other federal requirements pertain to health maintenance plan
organization and administration, data systems, financial solvency,
marketing, member services, and utilization review. These
requirements are specified in sec. 1903(m) of the Social Security
Act and 42 C.F.R. 434.
\31 One of these nine (Delaware) has taken specific steps to allay
concerns of disabled beneficiaries about relying on prepaid plans and
to help ensure access to appropriate providers, some of which have
been discussed in this chapter. However, as of April 1996, the state
had no specific measures to assess care received by disabled
enrollees.
TARGETED QUALITY-OF-CARE
STUDIES
------------------------------------------------------ Chapter 3:2.5.1
Arizona, the District of Columbia, Ohio, and Virginia will begin in
1996 to conduct additional quality-of-care studies focused
specifically on care for disabled enrollees. Arizona's studies will
include outcome measures, such as the frequency and reasons for
hospitalizations and emergency room visits, rates and changes in
pressure ulcers, and changes in functional abilities. Since 1990,
Arizona staff have also visited a random sample of developmentally
disabled beneficiaries in their homes to determine satisfaction with
services and progress in fulfilling individualized plans of care
developed by their health plans. The District of Columbia, Ohio, and
Virginia are collecting encounter data from health plans and will
evaluate care beginning in 1996 or 1997. A fourth
state--Wisconsin--is scheduled to release by December 1996 an
evaluation of its program conducted by independent researchers.
QUALITY IMPROVEMENT GOALS
------------------------------------------------------ Chapter 3:2.5.2
Massachusetts uses quality improvement goals and contractor selection
specifications to build health plan capacity to meet the needs of
Medicaid beneficiaries. Each year, the state identifies quality
improvement goals for all health plans and requires each plan to
select additional goals. Twice annually, Medicaid staff meet with
health plans to review progress in meeting stated goals. For each
contracting cycle, the state identifies capabilities it expects
successful contractors to possess. These goals and specifications
have included developing capacity to serve disabled individuals. For
example, one health plan elected to develop and implement a program
for enrollees with human immunodeficiency virus (HIV) or AIDS to
provide case management and access to specialists trained in
infectious diseases. Selection criteria for 1995 required prepaid
plans to demonstrate how they provided reasonable access to services
for enrollees with physical and communicational disabilities as
measured, in part, by enrollee satisfaction.
PROGRAMS CAN BE STRENGTHENED BY
ANALYSIS OF ENCOUNTER DATA
---------------------------------------------------------- Chapter 3:3
Under a fee-for-service approach to Medicaid, states have ready
access to data on services performed because they reimburse providers
for those services. These data--called claims data in
fee-for-service systems and encounter data in prepaid managed
care--consist of such information as the patient's identity, type of
service, date of delivery, diagnosis, and provider. In a prepaid
care setting, states do not need such data for reimbursement
purposes. Many plans have--and use--this information, but unless
states specifically request it, the information can largely disappear
from view.
This information can play an important role in quality assurance,
estimations of future service use, research, and program planning.
It can also play an important role in rate-setting, the subject of
the next chapter. However, state experience to date shows that a
substantial investment of time and effort is needed to assemble a
workable encounter database, although the potential applications
appear to make the effort worthwhile.
ENCOUNTER DATA HAVE HAD
LIMITED USE TO DATE
-------------------------------------------------------- Chapter 3:3.1
When Arizona, Oregon, and Tennessee received approval to implement
statewide Medicaid mandatory managed care programs, HCFA required
them to collect and validate encounter data, mainly for use in
independent evaluations of the programs. These states, which have
had significant experience in collecting such data, all had
difficulty obtaining information of sufficient quality and
comprehensiveness to use in quality assurance reviews. The problems
were numerous: The data were not readily available, health plans
used a variety of data systems, and definitions varied from plan to
plan.
-- Arizona has had by far the most experience in collecting and
using this information for quality assurance purposes. However,
the state spent over 10 years and $30 million getting to the
point that the Medicaid department could use encounter data for
quality analysis.
-- Oregon and Tennessee are experiencing collection and validation
problems similar to those Arizona experienced initially. In
each of these states, staff spent considerable time editing the
data, working with health plans to overcome problems, working to
resolve significant data reject and coding problems, and
implementing validation strategies. In Oregon and Tennessee,
relatively complete and usable data elements were not available
until almost 2 years after enrollment began.\32
State use of encounter data in quality reviews is also limited to
some extent by the lack of a recognized standard for what level of
care is considered appropriate for people with disabilities. In
addition, quality measures for chronic and disabling conditions are
just now being developed. Current federal and privately funded
research and development in the field of quality analysis will
provide states with more definitive criteria to use in their
analyses.
--------------------
\32 Some other states are also collecting encounter data but are not
attempting to use them in this way as yet. Florida and Maryland
reported collecting encounter data, but neither reported using such
data, at present, in studies of care for disabled enrollees.
California collects encounter data in one county. Delaware,
Pennsylvania, Utah, and Virginia have either just begun or will
within the next year collect such data.
ENCOUNTER DATA ANALYSIS
SHOWS POTENTIAL IN QUALITY
CONTROL APPLICATIONS
-------------------------------------------------------- Chapter 3:3.2
While assembling adequate databases is difficult and expensive, the
effort can yield substantial results in terms of the ability to
monitor programs. The types of studies that could be conducted using
person-level encounter data include tracking patterns of services by
health plan or eligibility group, identifying providers serving
special needs populations, and tracking the movement of high-cost
patients among health plans. Encounter data could also be analyzed
to reveal patterns of under- or overutilization. Although linking
such patterns to quality of care in all cases is limited by the lack
of recognized standards, patterns of service use can reveal access
problems. For example, Arizona officials analyzed encounter data and
found very low use of dental services among all beneficiaries. The
access problem was resolved when state officials removed the
requirement that beneficiaries receive a referral from their primary
care provider before obtaining dental care.
Encounter data for Oregon's disabled enrollees are just becoming
available for analysis.\33 As a result, no studies are yet under way.
However, state officials listed the following as possible uses for
encounter data: comparing utilization to adopted practice guidelines
to assess the extent to which they had been implemented; identifying
providers serving special needs populations; identifying and tracking
high-cost enrollees; identifying areas of underservice for selected
services; identifying gaps in follow-up care or preventive care for
selected enrollees; and analyzing enrollment to detect adverse
selection by selected diagnoses.
We also identified other innovative uses of data systems for more
limited quality or access reviews:
-- Arizona monitors case management for certain disabled Medicaid
beneficiaries through on-line systems. The state provides case
managers with the terminals and software with which they record
the individuals' plan of care and progress in meeting stated
goals. This information is then transmitted to the Medicaid
department for immediatiate review.
-- The contracting health plan in the District of Columbia also
plans to use an on-line system for its own and the District's
monitoring of care for enrollees. For case managers providing
24-hour medical access to beneficiaries and their families, the
system gives access to care plans, service authorizations, and
even scanned-in photographs of the children.
-- In Massachusetts, the Medicaid department compared managed care
aggregated utilization data with fee-for-service claims data to
determine whether mental health services were underutilized.
The state took immediate action to work with prepaid plans that
needed to correct utilization problems.
--------------------
\33 Enrollment by disabled beneficiaries was phased in over 9 months
beginning in February 1995. As a result, the collection and
availability of encounter data for disabled beneficiaries lag behind
data for previously enrolled groups of beneficiaries.
RISK-ADJUSTED RATES AND
RISK-SHARING CAN HELP REDUCE
INCENTIVES TO UNDERSERVE DISABLED
BENEFICIARIES
============================================================ Chapter 4
Adequate quality-of-care safeguards provide some protection against
the potential risks of prepaid managed care. Paying health plans a
capitation rate in advance to provide enrollees a set of services
creates an incentive to improve efficiency by eliminating unnecessary
services. However, it simultaneously creates certain risks. First
is a risk of underservice, because plans can profit by reducing the
number or quality of beneficial services. Second is the risk that
when the same capitation rate is paid for enrollees with different
health care needs, plans will seek to enroll the healthier, less
expensive individuals. These risks may be greater when plans feel
financial pressure from actual or potential losses from serving
enrollees with extreme needs.
States are examining ways to reduce these incentives and pressures in
prepaid care plans that have a disproportionate share of
beneficiaries with high-cost medical needs, such as severely disabled
people. States' efforts have been of three main types:
-- Using risk-adjusted capitation rates to more closely match the
reimbursement rates with anticipated costs of treating
individual recipients.
-- Sharing financial risk by providing retrospective adjustments
(called "reinsurance") to reimburse plans for losses resulting
from very high-cost individuals or disproportionate numbers of
enrollees with above-average costs.
-- Establishing funding agreements with "risk corridors" that
reimburse plans for a portion of losses but also require plans
to return part of the profits exceeding a specified level.
For the 17 states we contacted with managed care programs for
disabled beneficiaries, most state activity to date has centered on
reinsurance. Initiatives to establish risk-adjusted rates for
disabled enrollees or to set up risk corridors in funding agreements
are fewer in number and have much shorter track records.
Risk-adjusted rates--currently implemented in only two states--are
seen as potentially beneficial by many states but also as
administratively difficult to develop and maintain. As the only
mechanism that specifically limits health plan profits, risk
corridors appear to have the greatest potential for reducing plans'
incentives to underserve or to enroll only the healthier
beneficiaries. To date, five states have taken steps to build risk
corridors into their payments to plans.
TRADITIONAL RATE-SETTING
APPROACH DOES NOT ADDRESS
NEGATIVE INCENTIVES AND
PRESSURES
---------------------------------------------------------- Chapter 4:1
In setting capitation rates, states make an effort to account for
differences in expected costs for broad categories of beneficiaries.
To do this, they frequently divide the eligible population into
subgroups, or cells, of individuals with similar characteristics. Of
the 17 states we contacted, 16 established rate cells according to
Medicaid eligibility category, such as all disabled people or all
children in AFDC-eligible families, with some adjustment for age or
the geographic area in which the beneficiaries reside.\34
Setting capitation rates in this way meets HCFA requirements and
provides appropriate payments to plans as long as each plan's
enrollment mix of beneficiaries with complex health care needs is
comparable with the mix of the population used to set the rates.\35
The money saved serving enrollees with lower-than-average costs pays
the cost of serving enrollees with higher-than-average costs.
However, plans may not enroll disabled people with health care needs
comparable with those included in setting the rates. While some
disabled enrollees may require little medical treatment, others may
have disabilities, such as quadriplegia, that require extensive
treatment. The identifiability of such groups and the high costs
associated with their care heighten the incentives for health care
plans to avoid enrolling such individuals.
In most state programs, the rate-setting methods do not take into
account the cost variation associated with different types of
disabling conditions. Researchers have identified significant
variation in medical costs within different subcategories of
conditions. For example, using 1992 fee-for-service claims data
divided along clinical diagnoses, researchers found average annual
costs ranging from nothing (for the 5 percent of the disabled
population that had no medical claims during the year) to $35,000 per
year in one state for an individual diagnosed with quadriplegia.
Similarly Oregon found tremendous variation in 1993 health care costs
among its 199 highest-cost children. The 6-month group average was
$21,472, but amounts varied from a high of $410,420 to a low of
$5,014. In 1995, Oregon's average 6-month capitation rate for
disabled individuals in the cell that includes these children was
$3,023.\36
--------------------
\34 For example, Oregon's 1996 (January through September) average
rate for categorically eligible children in families under the
federal poverty level is $126.15 per month, while its average rate
for blind or disabled individuals who have no Medicare coverage is
$521.81 per month.
\35 Richard Kronick, Zhiyuan Zhou, and Tony Dreyfus, "Making Risk
Adjustment Work for Everyone," Inquiry, Vol. 32 (Spring 1995).
\36 Because (1) Oregon lacked a systematic approach to case
management for these children and (2) plans have limited flexibility
in providing low-cost in-home care, the most "medically fragile" of
these children were taken out of prepaid care and are being served on
a fee-for-service basis with case management by state staff.
HIGH-COST CASES STRENGTHEN
PRESSURE TO SEEK HEALTHIER
ENROLLEES
-------------------------------------------------------- Chapter 4:1.1
With such a broad range of costs within the category of disabled
enrollees, a health plan being paid on the basis of average costs may
make profits or experience losses unrelated to its ability to provide
high-quality health care services efficiently. Instead, these
profits or losses may be a function of how many high-cost cases it
does or does not enroll. A health plan with a disproportionate
number of high-cost cases that result in unanticipated losses is said
to be experiencing "adverse selection," while a plan with few
high-cost cases is said to be experiencing "favorable selection."
The greater the difference between the high- and low-cost recipients
in each cell, the greater the pressure on plans to avoid enrolling
high-cost recipients or to underserve the high-cost beneficiaries who
do enroll. Favorable selection may happen unintentionally in that,
as research suggests, some people--often those with few health care
needs--may be more prone to select prepaid care when given the
option. But, plans can also avoid enrolling high-cost members by
using a variety of methods that may be difficult for states to
detect.
-- Manipulating the panel of providers. Health plans can avoid
high-cost recipients by dropping providers that attract
high-cost patients. For example, a former health plan official
told us that the health plan she worked for identified a
specific provider who was responsible in large part for the
plan's attracting a significant number of enrollees with AIDS--a
condition that frequently requires extensive and expensive
treatment, especially in its later stages. This plan dropped
the provider from its panel in favor of an AIDS treatment clinic
and saw its AIDS caseload decrease. The decision to drop the
provider may have been for other reasons, in that by adding the
clinic and dropping the individual provider, the health plan may
have improved its capacity to treat people with AIDS and the
quality of care they would receive. The outcome demonstrates,
however, recipients' attachment to specific providers and health
plans' ability to (1) identify specific providers as magnets for
high-cost recipients and (2) reduce the cost to treat these
recipients by dropping or replacing certain providers.
-- Limiting access to information about specialty providers.
Health plans can also make it difficult for prospective
enrollees to find out which specialty providers are available
through the plan. In one state we studied, the state and health
plans initially resisted distributing a handbook produced by an
advocacy group designed to help disabled Medicaid recipients
select a health plan that could best meet his or her particular
needs. The booklet contained a worksheet for individuals to
detail their specialty care requirements. The state and health
plans were concerned about the possibility of adverse selection
and felt that, without this information, the high-cost cases
would be more evenly distributed among the various health plans.
-- Using marketing efforts to discourage enrollment. Some states
have allowed plans to conduct direct marketing as a way of
enrolling beneficiaries in managed care. However, through
direct marketing, health plans can also attempt to deliberately
influence the distribution of high-cost enrollees. For example,
they may seek information on a person's health status or
discourage--or not aggressively market to--those likely to have
more expensive needs. Consequently, several of the states in
our review prohibited or severely limited the amount and content
of marketing by health plans.
-- Remaining silent about new treatment approaches. The wide gap
between the relatively healthy and the sick within a rate cell
also discourages the dissemination of information about health
plans that have found innovative and successful ways to treat
enrollees with difficult conditions. One health plan official
told us that when the plan develops innovative and successful
ways to treat the chronically ill, it does not advertise this
fact because the resulting increase in enrollment of chronically
ill individuals could be financially devastating. This health
plan had success in managing asthma, and as word of its success
spread, the number of asthmatics enrolling in the plan increased
dramatically. This increase had such a negative financial
impact on the health plan that it asked the state to cap its
enrollment to prevent additional high-cost recipients from
enrolling. When the incentive of health plans to develop
innovative treatments decreases, disabled individuals are
adversely affected in that they may miss out on new and
effective treatments.
Determining whether a health plan is facing adverse selection goes
beyond reviewing the plan's financial statements to see if there is a
profit or loss. Health plans that do poorly managing care may lose
money and blame it on adverse selection even though they may in fact
be the beneficiary of favorable selection. Alternatively, plans
actually experiencing adverse selection may limit services to such an
extent they are still able to show a profit. Medicaid officials told
us it is not uncommon for all participating health plans to describe
themselves as victims of adverse selection, an impossible situation.
However, these states had limited ability to verify or refute such
claims with any certainty.
STATES COULD EXPERIENCE
ADVERSE SELECTION AND LOSE
MONEY WITH MANAGED CARE
-------------------------------------------------------- Chapter 4:1.2
Health plans are not the only players in the Medicaid managed care
marketplace that can face adverse selection and financial risks.
When prepaid managed care plan enrollment is voluntary (as it is in
11 of the 17 states now using prepaid care for some or all of their
disabled beneficiaries), the state may experience adverse selection.
Specifically, where participation is voluntary, beneficiaries with
relatively few health care needs (who may have few, if any, existing
relationships with specialists) may choose prepaid care, while
beneficiaries needing more expensive care (who may have long-standing
relationships with specific providers) may choose to remain in
fee-for-service care. When enrollment is mandatory but exceptions
are allowed, a state may similarly face adverse selection.
Enrollment patterns in which the users of the most expensive medical
services are in fee-for-service care and the relatively healthy in
prepaid managed care are not problematic if the rate the state pays
the health plans is adequately adjusted for the health status of the
enrollees. However, in many cases, the rates paid to health plans
are based on the average cost of providing care to an entire
eligibility category and may not appropriately account for those that
do not elect prepaid care. Consequently, the state pays the full
cost of treating the expensive beneficiaries through fee-for-service
care and too high a rate for the lower-cost health plan members.
This problem may be compounded in that it is likely that future
capitation rates would be based on the costs of serving those
remaining in fee-for-service care--individuals who are likely to be
less healthy and consequently more costly.
Just as it is difficult to tell if a health plan is experiencing
adverse selection, it is very difficult to determine whether a state
is experiencing adverse selection. An Oregon Medicaid official
suspects that the state's enrollment exemption process for disabled
individuals, which allows case workers to determine if prepaid
managed care is appropriate for individual beneficiaries, may be
resulting in adverse selection for the state.
RISK ADJUSTMENT AND REINSURANCE
HAVE SOME IMPACT ON INCENTIVES
FOR FAVORABLE SELECTION OR
UNDERSERVICE
---------------------------------------------------------- Chapter 4:2
To address the concerns associated with adverse and favorable
selection, some states are beginning to experiment with risk-adjusted
methods for setting capitation rates. Risk adjustment is an attempt
to match the rates paid to health plans with the expected costs of
providing appropriate services to individual recipients. It
essentially groups beneficiaries according to expected future expense
and narrows the gap between the highest- and lowest-cost individuals
in any given rate cell. This reduces the payoff for selecting only
the healthiest recipients and provides better assurance that the
state is not paying too much for individuals who are relatively
healthy or too little for individuals who need such complex and
expensive care that health plans are at best unwilling to attract and
at worst unwilling or unable to accommodate them.
However, the actual application of risk-adjustment methods to the
development of capitation rates for disabled Medicaid beneficiaries
is very limited.\37 To date, only two states (Massachusetts and Ohio)
have implemented any risk-adjustment methods, and only one other
state (Missouri) has active plans to do so. Other states told us
that risk adjustment was too administratively difficult to implement
and that they looked to reinsurance to protect plans that experience
adverse selection. Reinsurance does not, however, affect plans'
incentive to seek favorable selection.
--------------------
\37 To date, research on risk-adjustment mechanisms has concentrated
more extensively on the Medicare population. We examined
risk-adjustment mechanisms among the Medicare population in two
reports, Medicare: Changes to HMO Rate Setting Method Are Needed to
Reduce Program Costs (GAO/HEHS-94-119, Sept. 2, 1994) and Medicare
Managed Care: Growing Enrollment Adds Urgency to Fixing HMO Payment
Problem (GAO/HEHS-96-21, Nov. 8, 1995).
RISK ADJUSTMENT IS LARGELY
UNTESTED FOR DISABLED
ENROLLEES
-------------------------------------------------------- Chapter 4:2.1
The three states experimenting with risk-adjusted rates have based
their adjustments on a beneficiary's prior utilization of medical
services or a beneficiary's clinical diagnosis. Researchers point
out that such measures may better predict future costs since disabled
individuals, compared with the population as a whole, have a higher
percentage of their health care costs related to chronic (recurring
or consistent) conditions than to acute (random) conditions. Still,
for risk-adjustment methods to be useful, attention must be paid to
whether the predictive measures are sufficiently reliable and
administratively feasible to collect.
RISK ADJUSTMENT USING
PRIOR UTILIZATION RATES
------------------------------------------------------ Chapter 4:2.1.1
Utilization-based risk adjustment attempts to predict a person's
future health care costs based on a measure of prior use, such as the
costs of services or the number of hospital days used in a previous
period. For example, a health plan could be paid a
higher-than-average amount if the person spent several days in the
hospital in the last year or a lower-than-average rate if the person
spent no time in the hospital or did not visit the doctor in the last
year.
To set capitation rates for its disabled population, Ohio is moving
forward with a pilot project that uses a beneficiary's prior
utilization (measured in dollars) in the fee-for-service system.
This program, called Accessing Better Care, uses eight rate cells for
the disabled population. Seven of the cells are based on prior
expenditures, and the eighth is for newly eligible beneficiaries.
Monthly capitation rates range from $165 (for beneficiaries with
prior annual costs of $1,000 or less) to $4,501 (for beneficiaries
with prior annual costs of $50,000 or more). Figure 4.1 shows how
Ohio's disabled beneficiaries are distributed among the seven
prior-expenditure categories.\38 More than half of all disabled
beneficiaries are in the lowest-cost cell.
Figure 4.1: Ohio's
Risk-Adjusted Capitation Rates
and the Percentage of Disabled
Population at Each Rate
(See figure in printed
edition.)
--------------------
\38 The distribution is an average from 1991 and 1992 fee-for-service
data.
RISK ADJUSTMENT USING
CLINICAL DIAGNOSIS
-------------------------------------------------------- Chapter 4:2.2
Another approach predicts future health care costs using
beneficiaries' individual clinical diagnoses. Various methods to
identify or classify diagnoses can be used, employing inpatient data,
outpatient data, or both. Some methods rely on an individual's
primary diagnosis, and others incorporate measures of the severity of
the primary diagnosis as well as the existence of secondary diagnoses
of conditions that may aggravate the individual's health status.
Massachusetts is one of two states working with a diagnosis-based
approach. It has identified a few disabling conditions that warrant
higher rates and, as a result, has created a three-celled,
diagnosis-based risk-adjustment system for its disabled Medicaid
population. Severely disabled beneficiaries and those with end-stage
AIDS (in both cases meeting clinical criteria detailed by the state
and receiving an enhanced benefits package) have capitation rates of
about $1,500 and $4,400 per month, respectively, compared with about
$500 per month for beneficiaries in the general disabled category.
Missouri is currently developing a diagnosis-based methodology as
part of a prepaid care program for disabled Medicaid beneficiaries
scheduled to begin in 1997. Their methodology adjusts the capitation
rate paid for an individual recipient according to both the type of
diagnosis and its severity. The state identified approximately 400
distinct diagnoses and computed a rate-adjustment factor for each.
The rate for an individual is the sum of the adjustment factors for
each of the individual's diagnoses. The methodology accounts for
multiple diagnoses: the capitation rate for an individual with
diagnoses of muscular dystrophy and mild mental retardation would be
higher than the rate for a mildly retarded individual with no
additional disabling diagnoses. Table 4.1 shows an initial estimate
of the resulting base rate along with a sample of the diagnoses
identified and the associated capitation adjustment.
Table 4.1
Missouri Sample Diagnostic Categories
and Rate Adjustments
Adjust
Sample diagnostic categories ment
-------------------------------------------------------------- ------
Hemophilia, other clotting factors $1,495
Cystic fibrosis, respiratory failure 1,108
Quadriplegia 517
Muscular dystrophy/paraplegia 263
Mild and moderate mental retardation 74
----------------------------------------------------------------------
Note: Rate adjustments are additive. For example, the rate for an
individual with muscular dystrophy and mild mental retardation would
be $480 per month ($143 for the base rate plus $263 for muscular
dystrophy and $74 for mild mental retardation).
Source: Missouri Division of Medical Services.
RISK-ADJUSTMENT METHODS
MUST BE ACCURATE, FREE
FROM MANIPULATION, AND
WORKABLE
------------------------------------------------------ Chapter 4:2.2.1
Although risk-adjustment mechanisms are designed primarily to prevent
adverse and favorable selection, implementing a risk-adjustment
scheme involves a number of other considerations. Risk-adjustment
methodologies must not only be reasonable predictors of future health
care costs, they must also be relatively insulated from manipulation
by health plans or providers, and they must be feasible in terms of
administrative and data requirements.
To prevent adverse or favorable selection, a risk-adjustment
mechanism must be able to predict health care costs. Researchers
have demonstrated that prior utilization and diagnosis-based
methodologies can both have predictive power superior to that of
rates based on eligibility category. Some prior utilization models
are able to explain nearly 40 percent of the variation in health care
costs for disabled individuals, and diagnosis-based models have been
able to explain about 25 percent. However, even the best predictors
of health care costs explain less than half the variation in costs of
providing care. Plans then still have an incentive to avoid the
higher-cost members of a rate cell.
The basis selected for risk adjustment can affect the behavior of
health plans.
-- With utilization models that use cost as a measure, health plans
have less of an incentive to hold down costs because less
efficient health plans may be rewarded with higher capitation
rates as participants are characterized as high use.
Conversely, a plan that manages its enrollees' care very
efficiently may be penalized with lower capitation rates in the
future. Moreover, a prior utilization method based on the
number of hospital admissions affects health plan behavior in a
different way than one based on the number of days an individual
spends in the hospital. In both cases, the health plan could
manipulate the measure affecting future rates without
necessarily losing much in terms of efficiency.\39
-- With diagnosis-based risk-adjustment methods, "upcoding" exists
in which providers and plans record the most severe
diagnosis--the diagnosis associated with the greatest capitation
adjustment--of those available for an individual's symptoms.
The use of multiple diagnoses as factors in the rate-setting
methodology creates a situation in which providers could record
unwarranted diagnoses to raise future capitation rates.
To be useful, risk-adjustment methodologies must also have feasible
administrative and data requirements. Measures of health status,
collected through surveys, may help predict the need for future
health care but may be too administratively burdensome to be
practical. Risk-adjustment methods based on information about the
use of services are more practical. However, while such information
was routinely available in fee-for-service claims, states need a new
source, such as encounter data, to classify individuals enrolling in
health plans.
The lack of fee-for-service data is not a problem limited to states
that adjust their capitation rates by risk. Any state that moves
most or all of its Medicaid population into managed care will find
that prior rate-setting methods based on averaging fee-for-service
claims will be unsuitable. While rates calculated using older
fee-for-service data might be trended forward using any of a variety
of factors, over time such trending may cause rates to be
unrepresentative of the health care services being used. To address
these difficulties, some states are using or evaluating
individual-level encounter data as a basis from which they can
generate capitation rates in the future.
--------------------
\39 To account for the effects of adverse selection, some states are
considering making retrospective adjustments to capitation rates
based on utilization. Such an adjustment can ease the effects of a
large number of high-cost cases, but it may also strengthen the
incentive for health plans to act inefficiently. With a
retrospective adjustment, plans would be rewarded for their
inefficient behavior in the current year, rather than having to wait
for higher rates in the future.
REINSURANCE RELIEVES
FINANCIAL PRESSURE ON PLANS
BUT NOT NEGATIVE INCENTIVES
-------------------------------------------------------- Chapter 4:2.3
State officials we contacted recognized the benefits of prospectively
risk-adjusting capitation rates but--with the exceptions of Ohio,
Massachusetts, and Missouri--were reluctant to do so in their
programs because they felt it was too difficult administratively.
Instead, officials rely on reinsurance to decrease the pressure on
health plans serving high-cost individuals. With reinsurance, the
reinsurer (sometimes the state) protects health plans against adverse
selection or unexpectedly high-cost cases. To obtain coverage, the
plan pays a reinsurance premium.
Reinsurance programs come in many forms. Most programs involve a
reinsurance threshold, or deductible, with health plans being
responsible for all the costs of serving a group or an individual up
to that amount. Once the threshold is met, the state shares the cost
of treating the group or individual with the health plan. Table 4.2
shows the range of reinsurance options Oregon offers in its current
health plan contracts.
Table 4.2
Oregon's Reinsurance Levels and Rates,
1995-97
Annual deductible Percentage of Reinsurance
per person before liability over the premium as a
state will deductible to be percentage of
participate paid by the plan capitation
------------------ ------------------ ------------------
$10,000 5 29.9
15,000 10 19.7
30,000 20 9.3
50,000 20 4.1
----------------------------------------------------------
Source: Oregon Office of Medical Assistance Programs.
State Medicaid officials told us that, by protecting health plans
from extraordinary costs, reinsurance also helps build health plan
capacity. For example, new health plans and plans with small
enrollments need time to absorb spikes in service costs and the cash
flow fluctuations inherent in prepaid managed care. With reinsurance
available, these plans can participate and compete in the programs.
While reinsurance relieves some pressure on health plans faced with
expensive cases, it does not remove the negative incentives discussed
earlier. Plans still may benefit from enrolling the healthiest
eligibles or from underserving the high-cost cases that do enroll.
Reinsurance compensates plans only after they lose money on a case or
on all their enrollees. While reinsurance may relieve some pressure
on plans facing losses, it may not affect the incentives plans create
with individual providers to limit services. These incentives set
out in the provider's contract may not automatically adjust when the
costs of a provider's patient reach the reinsurance threshold.
In some areas, Medicaid managed care reinsurance may not be readily
available in the private market and may not be available at all for
small health plans. Consequently, state Medicaid agencies become de
facto insurance companies with the associated risk and resource
requirements. As reinsurers, states face the challenge of setting
appropriate reinsurance premiums--inappropriate premiums could lead
either to plans paying too much, and thus increasing the pressure to
underserve, or plans paying too little, which leaves the state in a
money-losing position.
RISK CORRIDORS HAVE GREATEST
IMPACT ON NEGATIVE FINANCIAL
INCENTIVES
---------------------------------------------------------- Chapter 4:3
Five states--the District of Columbia, Massachusetts, Ohio, Utah, and
Wisconsin--are building risk corridors into their contracts to help
mitigate the potentially negative incentives affecting health plans'
treatment of disabled enrollees.\40 Unlike reinsurance, risk
corridors work in two directions, sharing both losses and profits
with health plans below and above preestablished ratios.
As the only mechanism that specifically limits health plan profits,
risk corridors have the greatest impact on incentives facing health
plans to either reach for the lowest-cost recipients in any given
rate cell or to underserve the high-cost enrollees they cannot avoid.
The point at which profit and loss sharing begins--the width of the
risk corridor--varies from state to state, as does the degree to
which profits and losses are shared. Table 4.3 shows the risk
corridor arrangement Massachusetts has in its current contract with a
plan that provides prepaid care for the severely disabled.
Table 4.3
Massachusetts Risk Corridors for Plan
Providing Prepaid Care to the Severely
Disabled
Situation at end of contract
period Outcome
---------------------------------- ----------------------------------
Plan has medical expenditures The difference above 10% reverts
totaling to the state
more than 10% below capitation
payments
Plan has medical expenditures Plan keeps 40% of the difference;
between 60% reverts to the state
0 and 10% below capitation
payments
Plan has medical expenditures State pays 50% of the difference
between
0 and 10% above capitation
payments
Plan has medical expenditures State pays 75% of the difference
totaling
more than 10% above capitation
payments
----------------------------------------------------------------------
Source: Massachusetts Executive Office of Health and Human Services.
By reducing the potential for profits, the state is affecting
implicit health plan calculations regarding the costs and benefits of
restricting services. When $1 saved from restricting service
translates to $1 of profit, a health plan may be willing to risk
losing enrollees who are dissatisfied with health plan service. With
risk corridors, however, $1 saved may only translate to 30 or 40
cents in profit, reducing the benefit side of the equation. Because
health plans understand how risk corridor arrangements operate before
entering into Medicaid prepaid care agreements, corridors also have
the unique feature of being a retrospective adjustment with a
prospective impact. Risk corridors and their profit limits may
affect health plan risk arrangements established with individual
providers in a way that reinsurance does not. In their provider
contracts, plans may limit the incentives to reduce services when
their profits will be limited.
--------------------
\40 In addition to these five states, Tennessee has limited profits
for certain of its managed care organizations to 10 percent but plans
to discontinue the practice in December 1996.
OBSERVATIONS, CONCLUSIONS, AND
COMMENTS
============================================================ Chapter 5
Enrolling disabled beneficiaries in prepaid managed care is a growing
trend in Medicaid. Moreover, because much of the proposed expansion
is directed toward mandatory managed care, the future expansion of
prepaid care for disabled Medicaid beneficiaries appears likely to be
even more sweeping in its effect. Thus far, two-thirds of the states
providing prepaid care for disabled beneficiaries offer it on a
voluntary basis. By contrast, 12 of the 13 states with newly
approved or pending Medicaid managed care waivers intend to mandate
participation by disabled beneficiaries.
The implications of this shift toward mandatory programs are
substantial. Prepaid care has operated in both the public and
private arenas as a system based on averages. For example,
populationwide averages drive the expectations of what services
should be provided and how much they will be used. Likewise, prepaid
rates are calculated on average costs, and quality has been
monitored, in part, using aggregated average utilization rates. To
adequately safeguard the interests of disabled beneficiaries,
however, state programs must recognize that these beneficiaries are
quite distinct from the general Medicaid population. Not only are
their health needs greater than those of the general population, but
included among them are a small number of highly vulnerable
individuals whose needs are extensive and critical to the prevention
of death or further disability. Not addressing these differences
heightens the risk that prepaid care plans will try to hold down
their costs by (1) discouraging enrollment from high-cost segments of
the disabled population or (2) inadequately serving those high-cost
beneficiaries they cannot avoid.
Thus far, actions at the state level do not reflect a widespread
acknowledgment of the changes in approach that should occur when
applying managed care to disabled beneficiaries rather than the
general population. In most states, the level of effort to
anticipate and accommodate the needs of the various stakeholder
groups (disabled individuals and their advocates, the health care
plans, and the government) in their current programs has been limited
largely because participation in these programs has been voluntary.
The efforts have tended to be most extensive in those few states that
have already put mandatory or targeted programs in place.
No clear blueprint has yet emerged for how to incorporate disabled
beneficiaries into Medicaid managed care plans. The limited efforts
to date have not been in place long enough to allow definitive
conclusions about how effective they are. At this relatively early
stage, however, several key areas are emerging that merit
consideration by all parties seeking to develop effective prepaid
programs. These key areas, and examples of state actions to address
them, are illustrated in figure 5.1.
Figure 5.1: Key Approaches for
Including Disabled
Beneficiaries in Medicaid
Managed Care and Examples of
State Initiatives
(See figure in printed
edition.)
To date, few states have significant, long-term experience with
programs that mandate enrollment by their disabled population. Even
fairly extensive experience with voluntary programs may not fully
prepare health plans and state officials if, as research suggests,
those who select prepaid care in voluntary situations tend to be
healthier than those who do not. A state may find it useful to
develop and operate a targeted or relatively small-scale program
before moving to any large-scale effort to mandate the enrollment of
disabled beneficiaries. Small-scale programs would allow health
plans, beneficiaries, and state staff to gain experience with meeting
the diverse and complex needs of disabled individuals in a prepaid
setting.
For states that elect to move immediately into a large-scale program,
the areas shown in figure 5.1 are even more critical. Adequate
preparation, consensus-building, and program safeguards assume
greater significance when substantial numbers of people are being
added, particularly if their ability to change plans readily is
limited.
Understanding the various approaches currently being tried will
provide states with a good starting point for planning their own
efforts. Making prepaid managed care work for disabled individuals
will be achieved only through the combined and continuing efforts of
states, health plans, and beneficiaries and their advocates.
AGENCY AND OTHER COMMENTS
-------------------------------------------------------- Chapter 5:0.1
We provided a draft of this report to the Administrator, HCFA. The
draft report was reviewed by officials in HCFA's Office of Managed
Care, Office of Research and Demonstrations, and the Medicaid Bureau.
HCFA officials had no technical or other comments on the report
draft. In addition, we provided relevant sections of the draft
report to Medicaid staff from the 17 states in our report. All but
one state responded with comments, generally agreeing with the
accuracy of the information. Officials in Arizona commented that the
draft report seemed to suggest prepaid managed care is not suitable
for people with disabilities. We believe, instead, that given the
limited state and health plan experience with serving disabled
individuals in prepaid care and the medical complexity of their
health care needs, careful attention is required in designing,
implementing, and monitoring programs for this population.
In addition to requesting comments from HCFA and state agencies, we
provided the draft report to several independent researchers from the
National Academy for State Health Policy, the Medicaid Working Group,
MEDSTAT, and Fox Health Policy Consultants. These researchers
generally agreed with the accuracy and comprehensiveness of our
presentation of the issues and programs. We incorporated technical
and clarifying comments from states and external researchers as
appropriate.
MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix I
Kathryn G. Allen, Assistant Director, (202) 512-7059
Cheryl A. Williams, Evaluator-in-Charge, (503) 235-8451
Deborah A. Signer
Stanley G. Stenerson
Mark D. Ulanowicz
*** End of document. ***