Arizona Medicaid: Competition Among Managed Care Plans Lowers Program
Costs (Letter Report, 10/04/95, GAO/HEHS-96-2).
Pursuant to a congressional request, GAO reviewed Arizona's Medicaid
program, focusing on: (1) how the program contains costs; (2) how health
plan competition contributes to its cost containment success; (3) the
effect of cost containment on beneficiary access to care; and (4)
lessons learned from Arizona's cost containment success that could apply
to other states' Medicaid programs.
GAO found that: (1) Arizona's Medicaid program mandates managed care
enrollment and pays health plans a capitated fee for each beneficiary
served; (2) although the program initially experienced problems, it now
successfully contains health care costs and provides access to
mainstream medical care; (3) while other states' Medicare costs have
risen, Arizona's capitation rates have steadily declined; (4) although
the cost to administer Arizona's Medicaid program is relatively high,
the program saves both the federal and state government millions of
dollars annually in acute care costs; (5) the program has successfully
contained costs by developing a competitive Medicaid health care market;
(6) the program's emphasis on cost containment and competitive billing
has not adversely affected beneficiary access to appropriate care; and
(7) other states could learn from Arizona's cost containment success by
deviating from constraining managed care regulations, developing and
using market forces, implementing controls to protect beneficiaries from
inadequate care, investing in data collection and analysis capabilities,
and revamping existing Medicaid programs.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: HEHS-96-2
TITLE: Arizona Medicaid: Competition Among Managed Care Plans
Lowers Program Costs
DATE: 10/04/95
SUBJECT: Health care cost control
Medicaid programs
State-administered programs
Health maintenance organizations
Beneficiaries
Health care services
Medical economic analysis
Competitive procurement
Health services administration
IDENTIFIER: Arizona
Arizona Health Care Cost Containment System
Maricopa County (AZ)
Pima County (AZ)
**************************************************************************
* 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. *
* *
* A printed copy of this report may be obtained from the GAO Document *
* Distribution Facility by calling (202) 512-6000, by faxing your *
* request to (301) 258-4066, or by writing to P.O. Box 6015, *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time. *
**************************************************************************
Cover
================================================================ COVER
Report to the Chairman, Committee on Commerce, House of
Representatives
October 1995
ARIZONA MEDICAID - COMPETITION
AMONG MANAGED CARE PLANS LOWERS
PROGRAM COSTS
GAO/HEHS-96-2
Arizona Medicaid
(101325)
Abbreviations
=============================================================== ABBREV
AFDC - Aid to Families With Dependent Children
AHCCCS - Arizona Health Care Cost Containment System
ALTCS - Arizona Long Term Care System
HCFA - Health Care Financing Administration
HHS - Department of Health and Human Services
HMO - health maintenance organization
OMD - Office of the Medical Director
PCCM - primary care case management
PMMIS - prepaid medical management information system
RFP - request for proposal
SOBRA - Sixth Omnibus Budget Reconciliation Act
SSI - Supplemental Security Income
Letter
=============================================================== LETTER
B-261607
October 4, 1995
The Honorable Thomas J. Bliley, Jr.
Chairman, Committee on Commerce
House of Representatives
Dear Mr. Chairman:
Seeking to control escalating health care costs and improve access to
the Medicaid program, states are adopting various managed care
delivery systems. More than 13 years ago, Arizona was the first
state to obtain approval from the Health Care Financing
Administration (HCFA) to develop and implement a mandatory statewide
Medicaid managed care system. Arizona's Medicaid program today
benefits significantly from health plans competing with each other to
win contracts to provide health care to Medicaid beneficiaries in
both the state's urban and rural areas.
Because of Arizona's experience in implementing a statewide managed
care program, you asked us to review Arizona's Medicaid program and,
specifically, to discuss (1) the program's cost containment
experience, (2) the role of health plan competition in the program's
cost containment success, (3) the effect of cost containment on
beneficiary access to appropriate care, and (4) lessons about
Arizona's cost containment success that could apply to other states'
Medicaid programs.
To do this work, we interviewed federal and state Medicaid officials
and officials of managed care plans operating in Arizona. We
reviewed documents relating to the 1994 competitive contract process,
Arizona Health Care Cost Containment System's (AHCCCS) quality
management policies for health plans, and many studies of Arizona's
Medicaid program. We obtained and analyzed cost information for 1991
to 1995 and plan profit data from 1986 to 1994. We also interviewed
local advocacy groups representing the Medicaid population, such as
attorneys in state and county legal aid offices and private nonprofit
groups, and reviewed beneficiary satisfaction studies. Our review
addresses Medicaid's acute care expenditures and services only. We
conducted our work from September 1994 through August 1995 and
followed generally accepted government auditing standards.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
While many states are converting their traditional fee-for-service
Medicaid programs to managed care delivery systems, Arizona's
Medicaid program offers valuable insights--especially in fostering
competition and monitoring plan performance. Since 1982, Arizona has
operated a statewide Medicaid program that mandates enrollment in
managed care and pays health plans a capitated fee for each
beneficiary served. Although the program had problems in its early
years, such as the contract termination of the program administrator
and the state's takeover of the administration, it has succeeded both
in containing health care costs and providing beneficiaries access to
mainstream medical care.
Arizona's recent cost containment record is noteworthy. According to
one estimate, Arizona's Medicaid program saved the federal government
$37 million and the state $15 million in acute care costs during
fiscal year 1991.\1 While other states' per capita costs for Medicaid
have continued to grow, Arizona's capitation rates declined by 11
percent in 1994. Reviews have also shown that, since its inception,
the per capita growth rate of Arizona's program has been less than
the national per capita growth rate for states with traditional
Medicaid programs. Although the amounts that Arizona spends to
administer its program are higher than what other states spend, these
additional expenditures more than pay for themselves in net program
savings.
Arizona's program succeeded in containing costs by developing a
competitive Medicaid health care market. Health plans that submit
capitation rates higher than their competitors' bids risk not winning
Medicaid contracts. In the latest cycle, seven plans bid
unsuccessfully, including two that had previously had contracts.
Health plans compete to serve Arizona's Medicaid population because
doing so can be profitable. In 1994, health plans earned an
aggregate of $56 million in profits, or 6.7 percent of gross income.
Altogether, 95 bids, double the number from the previous contract
cycle, were submitted for the 42 contracts awarded.
Arizona's emphasis on cost control does not appear to have hindered
beneficiaries' access to appropriate care. For one thing, the
competitive bidding process assigns more points to access and quality
factors than capitation rates. In addition, the program specifies
standards that plans must meet for the number and types of providers
in each contract's geographical location, requires plans to routinely
provide data documenting a plan's stability and levels of care
provided, and requires plans to conduct various studies measuring
patient outcomes. According to HCFA, Arizona is among the states
taking the lead in developing systems that collect encounter data and
more accurately measure patient outcomes. Equally important,
beneficiaries and their advocates have expressed overall satisfaction
with Arizona's program.
Although each state's Medicaid program is different, other states
that are considering implementing or are currently operating a
managed care program can benefit from Arizona's experience. Our work
suggests that key conditions for containing Medicaid costs without
compromising beneficiaries' access to appropriate medical care
include
freedom from certain federal managed care regulations,
development and use of market forces,
controls to protect beneficiaries from inadequate care, and
investment in data collection and analysis capabilities.
Arizona's experience also demonstrates that a successful program
requires substantial preparation and development. States need a
transition period to make the dramatic shift from third-party payer
in a fee-for-service system to health plan overseer monitoring costs,
access, and quality of care.
--------------------
\1 This is the latest year for which a comprehensive study exists of
Arizona's cost containment performance compared with similar states
with traditional Medicaid programs.
BACKGROUND
------------------------------------------------------------ Letter :2
Almost all states pay for the bulk of their Medicaid beneficiaries'
health care through a fee-for-service system, which simply reimburses
providers for their services. To control service utilization and
therefore costs, states are increasingly adopting managed care--which
coordinates beneficiaries' care--for the acute care portion of their
programs. Managed care can range from capitated models that pay
organizations a set monthly fee for all services to primary care case
management models (PCCM), which pay for each service rendered but
also pay certain primary care physicians a small amount to coordinate
their patients' care. Capitated models are considered the strongest
form of managed care because they give providers the greatest
incentive to control utilization; PCCMs are considered the weakest.\2
Most states with managed care programs use a PCCM model alone or in
conjunction with some care paid on a capitated basis.
Arizona's Medicaid program is different. The state never
participated in the traditional fee-for-service Medicaid program,
implementing instead in 1982 a program that mandates managed care for
all beneficiaries and pays all of its health plans a fixed amount per
person to provide all covered services. The program, AHCCCS
(pronounced "access"), features a competitive contract award process
to purchase health care services from private- or county-operated
health care plans. The capitation rate is set after a bidding
process in which health plans submit separate bids for each county
they wish to serve. (App. I summarizes the competitive contract
process.)
To help ensure that AHCCCS beneficiaries have access to appropriate
medical care, health plan contracts stipulate specific provider
networks, ensuring provider availability in both urban and rural
locations. AHCCCS monitors the medical and financial performance of
contracting plans. (App. II summarizes oversight activities.)
For a state to mandate enrollment of its Medicaid population in a
statewide managed care program, it must obtain an 1115 demonstration
waiver from the federal government.\3 In 1982 Arizona was the first
state to have a waiver approved for this purpose. Initially, AHCCCS
had waivers from federal regulations requiring that a full scope of
services be provided to enrollees. Over the years, however, AHCCCS
has added the full range of Medicaid services, including family
planning, behavioral health, and long-term care.\4
As the program has evolved, however, it has had difficulties,
including a failed attempt to contract for administrative services in
its start-up phase, health plans with financial difficulties, and
impediments in setting up data systems. We have reviewed the program
throughout its history and made recommendations that HCFA and the
state have implemented.\5
(App. III details AHCCCS' history.)
AHCCCS' acute care program now serves over 432,000 beneficiaries in
15 counties.\6 About three-fourths of the beneficiaries live in
Arizona's two urban counties, encompassing the cities of Phoenix and
Tucson. The remaining beneficiaries live in Arizona's 13 rural
counties. Several of the rural counties are so remote they are
categorized as "frontier."
--------------------
\2 Capitated payment systems challenge health plans to establish
cost-effective health care systems and search for new ways to achieve
savings and manage utilization. Health plans are at risk for the
cost of services provided to an enrollee and must absorb the loss if
these costs exceed the monthly capitation payment. Health plans'
cost control methods include emphasizing cost-effective preventive
medicine; identifying inefficient patterns of obtaining health care,
such as going to the emergency room for primary care; contracting
with a third party to handle segments of care, such as home health or
pharmaceuticals; and bargaining with providers to obtain substantial
discounts from usual fees.
\3 Named for section 1115(a) of the Social Security Act, these
waivers enable states to conduct demonstrations that operate
differently from conventional fee-for-service Medicaid programs.
Since 1992, 11 states have obtained such waivers from HCFA.
\4 In 1989 AHCCCS began the Arizona Long Term Care System (ALTCS)
that offers acute care, nursing home care, and home and
community-based services to the elderly and physically and
developmentally disabled. Currently, ALTCS has about 20,000
enrollees. Since 1990 AHCCCS has expanded mental health care
coverage to people enrolled in both its ALTCS and acute care
programs.
\5 Medicaid: States Turn to Managed Care to Improve Access and
Control Costs (GAO/HRD-93-46, Mar. 17, 1993); Medicaid: Lessons
Learned From Arizona's Prepaid Program (GAO/HRD-87-14, Mar. 6,
1987); Arizona Medicaid: Nondisclosure of Ownership Information by
Health Plans (GAO/HRD-86-10, Nov. 10, 1985); and The Health Care
Financing Administration's Monitoring of the Arizona Health Care Cost
Containment System (GAO/HRD testimony, June 15, 1984).
\6 Eligible AHCCCS beneficiaries include the federally mandated
groups--Aid to Families With Dependent Children (AFDC) recipients,
Supplemental Security Income (SSI) recipients, and women and children
covered under the 1987 Sixth Omnibus Budget Reconciliation Act
(SOBRA). Medicaid- eligible American Indians can choose to receive
services from the Indian Health Service or, off-reservation, from an
AHCCCS health plan.
AHCCCS PRODUCES NOTEWORTHY COST
SAVINGS RECORD
------------------------------------------------------------ Letter :3
Arizona's success in encouraging competition among managed care
health plans has saved the state and the federal government money.
AHCCCS has lowered the 1995 per member capitation rates paid to
health plans from rates paid a year earlier. In the last several
years, it has also contained the rate of growth in per capita
Medicaid expenditures, resulting in per capita costs that are lower
than those in traditional, but otherwise comparable, Medicaid
programs. While AHCCCS' program management requires high
administrative expenditures of the state, the large savings in
medical costs still result in a net savings.
AHCCCS CAPITATION RATES
DECREASED BY 11 PERCENT IN
1995
---------------------------------------------------------- Letter :3.1
AHCCCS capitation rates fell, on average, by about 11 percent from
1994 to 1995, which is remarkable for at least three reasons. (Fig.
1 shows the annual change in capitation rates by beneficiary category
for each of the last 4 years.) First, average spending per Medicaid
beneficiary was projected to rise nationally during the same period.
Second, the cost of serving beneficiaries in the AHCCCS program was
already below the cost that would have been incurred in a traditional
Medicaid program.\7 Third, health plans serving AHCCCS beneficiaries
are now exposed to greater risk than under previous contracts for
expenses generated by beneficiaries with extensive health care
needs.\8 Typically, the plans' exposure to greater financial risk
would have tended to make the contracts less attractive to health
plans and, all else equal, would have led to higher--not
lower--capitation rates.
Figure 1: Annual Percentage
Change in AHCCCS Capitation
Rates, by Eligibility Category
(See figure in printed
edition.)
--------------------
\7 This is the finding of a 1994 Laguna Research Associates study
(see footnote 9).
\8 Effective October 1, 1994, AHCCCS changed its reinsurance and
deferred liability policies. Reinsurance is a benefit that AHCCCS
provides to contracted health plans to reduce their financial risk
for services rendered to a member whose liability reaches a certain
amount, known as a deductible. Reinsurance is now limited to
inpatient services only (ambulatory claims were eliminated), and the
acute care services threshold was increased. In 1994 deferred
liability was eliminated. Deferred liability allowed health plans to
be reimbursed on a fee-for-service basis for care provided to newly
enrolled beneficiaries who met special conditions, such as being
hospitalized at the time of enrollment, receiving active
chemotherapy, or enrolling in the program in the last 2 weeks of a
high-risk pregnancy. AHCCCS made these changes to shift more risk to
the health plans and encourage them to manage care efficiently.
STUDY SHOWED AHCCCS
CONTROLLED COSTS BETTER THAN
TRADITIONAL PROGRAMS
---------------------------------------------------------- Letter :3.2
AHCCCS slowed the growth rate in Medicaid expenditures compared with
what would have occurred had Arizona served the same beneficiaries
using a traditional Medicaid fee-for-service program,\9 according to
a 1994 Laguna Research Associates study conducted for HCFA.
Specifically, AHCCCS' annual per capita growth rate from 1983 through
1991 (the most recent year for which these cost comparison figures
are available\10 ) for the combined AFDC and SSI beneficiary
population was 6.8 percent versus 9.9 percent for a traditional
Medicaid program. (See fig. 2 for a comparison of AHCCCS costs with
traditional Medicaid costs.)
Figure 2: Per Capita Cost of
AHCCCS Compared With
Traditional Program, Fiscal
Years 1983-91
(See figure in printed
edition.)
The biggest slowdown in AHCCCS growth rates occurred for the
SSI-eligible beneficiary population after 1987. From the late 1980s
until 1991, costs for Arizona's AFDC-eligible beneficiaries were
lower but rose at about the same rate as per capita costs in
traditional Medicaid programs. However, during the same period,
costs under AHCCCS grew much more slowly for SSI-eligible
beneficiaries than the Laguna Research study estimated they would
have grown under a traditional program. (See figs. 3, 4, and 5.)
For example, per capita costs for disabled beneficiaries grew by 5.4
percent between 1987 and 1991 compared with the 17.3 percent that
Laguna estimated they otherwise would have grown. For aged
beneficiaries under AHCCCS, annual costs grew by 9.5 percent,
compared with the benchmark of 17.2 percent.
Figure 3: Cost of Serving an
AFDC-Eligible Recipient in
AHCCCS Compared With Estimated
Cost in a Traditional Program,
Fiscal Years 1983-91
(See figure in printed
edition.)
Figure 4: Cost of Serving an
SSI Disabled Recipient in
AHCCCS Compared With Estimated
Cost in a Traditional Program,
Fiscal Years 1983-91
(See figure in printed
edition.)
Figure 5: Cost of Serving an
SSI Aged Recipient in AHCCCS
Compared With Estimated Cost in
a Traditional Program, Fiscal
Years 1983-91
(See figure in printed
edition.)
In addition, Laguna Research estimated that AHCCCS spent less per
person in 1991 than would have been spent under a traditional
program. Overall, Arizona spent 81 percent of what a traditional
program would have spent, although this percentage varied by Medicaid
eligibility populations. For example, spending for disabled
beneficiaries was only 65 percent of spending for those in a
traditional program, whereas spending for blind beneficiaries was
estimated to have been the same. (See fig. 6 for AHCCCS per capita
spending compared with traditional programs by beneficiary category.)
Figure 6: AHCCCS Capitation
Rates as a Percent of Estimated
per Capita Costs in a
Traditional Program, by
Eligibility Category, Fiscal
Year 1991
(See figure in printed
edition.)
--------------------
\9 Managed Medicaid Cost Savings: The Arizona Experience, Laguna
Research Associates (San Francisco: 1994) (study limited to AFDC and
SSI recipients).
\10 Laguna Research Associates has submitted a draft report to HCFA
that updates this comparison to fiscal year 1993. The data in this
draft report show a continuation of this trend: that is, AHCCCS per
capita costs grew more slowly in 1992 and 1993 than they would have
under a traditional program.
NET SAVINGS GENERATED
DESPITE SLIGHTLY HIGHER
ADMINISTRATIVE COSTS
---------------------------------------------------------- Letter :3.3
Arizona's program demonstrates that adequately funding oversight
activities more than pays for itself in net program savings. AHCCCS
administrative costs are higher than the amount usually spent
administering a traditional Medicaid program. In 1994 Arizona's
administrative costs amounted to about 7 percent of what the state
spent on medical services; most Medicaid programs spend about 4 to 5
percent on administration. Although we did not evaluate the
appropriateness of AHCCCS' higher administrative costs, the Laguna
Research report concluded that running an AHCCCS-like program is
necessarily more expensive than running a traditional Medicaid
program. Even with its higher administrative costs, the AHCCCS
program generates net savings. For example, the report estimated
that in 1991 AHCCCS generated medical savings of $70.7 million,
required additional administrative costs of $19.2 million, and thus
produced a net savings of $51.5 million for that year. Because the
federal government covers about two-thirds of Arizona's Medicaid
costs, it saved roughly $37 million.\11
--------------------
\11 The federal government pays for a certain share of each state's
Medicaid medical expenses, called the federal medical assistance
percentage. This varies by state: as of fiscal year 1993 in
Arizona, this percentage was 65.89. In addition, the federal
government pays 50 percent of most administrative expenses for all
states, although it pays a higher percentage for certain
administrative categories.
COMPETITIVE BIDDING CONTROLS
COSTS, WHILE PROFITS SPUR PLAN
PARTICIPATION
------------------------------------------------------------ Letter :4
By using competitive bidding to award Medicaid managed care
contracts, Arizona has harnessed market forces to help contain health
costs. Plans are not assured of winning a contract; consequently,
they have an incentive to submit the lowest bid for which they can
provide the required beneficiary services and still earn a profit.
Health plans are attracted to the program because they can earn
profits from the AHCCCS contracts. Arizona and the federal
government both benefit because the health plans--facing a fixed
capitation rate--have an incentive to practice cost-conscious
medicine and control costs.
COMPETITIVE PROCESS
ENCOURAGES LOW BIDS
---------------------------------------------------------- Letter :4.1
Health plans that submit capitation rates higher than their
competitors' risk exclusion from Arizona's Medicaid market. AHCCCS
awards a limited number of contracts in each county: a maximum of 10
in Maricopa County--where Phoenix is located; 5 in Pima County--where
Tucson is located; and 2 in each of the 13 remaining, more sparsely
populated counties.
With each bidding cycle, the state has improved its ability to
evaluate bids because it collects encounter, cost, and profitability
data from each plan. In the latest contract cycle, seven bidders
failed to win any acute care contracts, including two health plans
that previously served AHCCCS beneficiaries. These two plans lost
their acute care contracts because they failed to fully recognize the
competitive pressures.
AHCCCS collects utilization, cost, and profit data from the health
plans that serve its Medicaid beneficiaries. Independent actuaries
use these data to estimate the cost of serving beneficiaries and
establish reasonable bid ranges for capitated payments. Bid range
information is not shared with the health plans, but the utilization
data are. An AHCCCS official stated that health plans face less risk
when they have reliable utilization data on which to base their bids,
a factor that also encourages plan participation.
Existing plans know that they face competition not only from other
plans but also from first-time bidders. In the latest cycle, two
commercial insurers, subsidiaries of Cigna and Blue Cross/Blue
Shield, succeeded in obtaining contracts, one for the first time.
First-time bidders are not necessarily disadvantaged, compared with
current providers, in the contract award process. Prior performance
can favorably or unfavorably affect the scores of current health
plans' bids. Since first-time bidders have no performance record,
their scores are unadjusted. In short, existing AHCCCS health plans
with good performance receive some advantage in the contract award
process, but first-time bidders do not receive lower scores simply
because they have not had an AHCCCS contract previously.
AHCCCS may permit health plans to submit both an initial and a final
capitation bid, but the award process favors plans that submit low
bids both times. AHCCCS evaluates the initial bids and provides
feedback to the plans on their cost projections. However, AHCCCS
does not reveal specifics on either the state's own cost estimates or
competitors' bids. After receiving feedback, health plans have an
opportunity to submit a second--and final--capitation bid. However,
both bids are considered in the contract award process. That is, a
plan that has submitted a high initial bid will be at a disadvantage
compared with a competitor that submitted a low initial bid, even if
their final bids are identical.
Another incentive for low bids is tied to the market shares of
contract winners. Although all Medicaid beneficiaries may specify
which available plan they will join, only about 56 percent actively
pick a plan. The remainder are assigned to a plan by AHCCCS, as are
all fully state-funded beneficiaries. Although it considers factors
other than cost, AHCCCS assigns more beneficiaries to lower cost
plans.
POTENTIAL PROFITS, EVEN IN
RURAL AREAS, ENCOURAGES
HEALTH PLAN PARTICIPATION
---------------------------------------------------------- Letter :4.2
AHCCCS' 1994 contract bidding process resulted in an unprecedented
level of competition that helped achieve cost savings. Twenty-one
health plans submitted 95 bids to provide Medicaid services in
Arizona's 15 counties for the 1995-1997 contract period. This was
more than double the number of bids (44) submitted in 1992, the start
of the previous contract cycle. AHCCCS awarded a total of 42
contracts to 14 of the 21 bidders. The contract recipients included
both commercial and noncommercial health plans, showing widespread
interest in participating in the AHCCCS program.
Health plan officials cited reported profits as a key reason for the
increased interest in obtaining an AHCCCS contract. AHCCCS health
plans earned a combined $44 million in profits, representing 5.4
percent of revenue, in 1993. As table 1 shows, aggregate profits for
AHCCCS health plans have been positive since 1990 and have grown
steadily. Aggregate profits in 1994 reached $56 million,
representing 6.7 percent of plans' total revenue.
Table 1
AHCCCS Health Plans' Revenues and
Profits, 1986-94
Total Net income Net income
Year revenue (dollars) (percent)
---------------- ------------ ------------ ------------
1986 $208,811,761 ($3,062,316) (1.47)
1987 220,636,178 5,088,259 2.31
1988 298,086,180 (8,237,696) (2.76)
1989 339,063,375 (347,058) (0.10)
1990 445,346,952 1,299,891 0.29
1991 561,492,337 17,585,993 3.13
1992 713,781,503 34,167,143 4.79
1993 809,463,837 43,528,836 5.38
1994 843,354,415 56,141,598 6.66
----------------------------------------------------------
Source: AHCCCS Office of Managed Care.
Health plans compete to serve Medicaid beneficiaries in Arizona's
rural counties because plans can profit in these areas, as well as
urban areas. It is impossible to measure profits for all rural
counties because profits are reported for each health plan, not each
county, and some health plans serve both urban and rural counties.
However, in 1994, six health plans served only rural counties, and
five of these plans earned profits ranging from 2 to 3 percent of
gross revenues. The remaining plan had unusually low medical costs
and earned a profit of over 23 percent. This plan, however, did not
win a contract in 1995. The ability of Medicaid managed care to
thrive in rural areas stems from the way capitation rates are set.
AHCCCS' capitation rates are set by the market on the basis of health
plans' costs of providing services. Consequently, capitation rates
are only slightly lower in rural counties compared with rates in the
two urban counties.
In contrast, relatively few rural Medicare beneficiaries in Arizona
receive their health care from health maintenance organizations
(HMO). This likely results from the way Medicare capitation rates
are set. Unlike the competitive bidding process used by AHCCCS, HCFA
sets Medicare rates for each county using a formula based on Medicare
fee-for-service costs in that county. In earlier testimonies and
reports, we noted several flaws in the Medicare rate-setting
formula,\12 which can understate or overstate the actual cost of
providing health care in a managed care setting. For example, if
Medicare beneficiaries lack adequate transportation and thus
consequently receive less care, Medicare fee-for-service expenditures
will be artificially low and the formula will set capitation rates
too low for the entire county.\13 Low rates discourage health plans
from participating in Medicare managed care (risk contracts) and
beneficiaries from enrolling in the plans that do exist.\14
--------------------
\12 Medicare: Rapid Spending Growth Calls for More Prudent
Purchasing (GAO/T-HEHS-95-193, June 28, 1995); Medicare Managed Care:
Program Growth Highlights Need to Fix HMO Payment Problems
(GAO/T-HEHS-95-174, May 24, 1995); and Medicare: Changes to HMO Rate
Setting Method Are Needed to Reduce Program Costs (GAO/HEHS-94-119,
Sept. 2, 1994).
\13 Under AHCCCS, rural Medicaid capitation rates are approximately
90 percent of urban capitation rates. The divergence of Arizona
Medicare capitation rates is much larger: rural rates amount to only
about 50 percent of the rates paid in urban counties.
\14 Low capitation rates also discourage Medicare beneficiaries from
enrolling in the HMOs that do exist. In areas with low rates, HMOs
offer fewer inducements for beneficiaries to enroll, such as
prescription drug benefits, compared with HMOs in areas with high
capitation rates. Unlike Arizona's Medicaid beneficiaries, its
Medicare beneficiaries are not required to join HMOs.
FLEXIBILITY TO WAIVE FEDERAL
REQUIREMENTS HELPS ARIZONA
ENCOURAGE PLAN PARTICIPATION
---------------------------------------------------------- Letter :4.3
AHCCCS has used the flexibility granted by its section 1115 waiver to
address federal requirements that tend to limit health plan
participation in Medicaid managed care. For example, the waiver
allows AHCCCS to award contracts to health plans that serve only
Medicaid beneficiaries. Without a waiver, current Medicaid standards
require that more than 25 percent of a health plan's enrollees be
non-Medicaid, non-Medicare beneficiaries. All 14 of AHCCCS' current
health plans serve only Medicaid beneficiaries or other AHCCCS
populations, although some of the plans are subsidiaries of larger
organizations. The waiver also allows AHCCCS to mandate that
beneficiaries enroll in a managed care plan. AHCCCS guarantees an
initial enrollment of at least 5 months--even if a beneficiary should
lose Medicaid eligibility before that time. This guarantee reduces
enrollee turnover and increases the attractiveness of Medicaid
contracts to the health plans.
FOCUS ON COST DOES NOT APPEAR
TO HARM ACCESS TO APPROPRIATE
CARE
------------------------------------------------------------ Letter :5
Recent comprehensive measures are not available, but several
indicators suggest that beneficiary satisfaction is high. AHCCCS'
beneficiary satisfaction suggests that the program has established
effective controls to ensure beneficiary access to appropriate care.
The program, (1) in its competitive bidding process, places more
weight on access and quality factors than it does on capitation
rates; (2) requires networks to meet certain standards for primary
care coverage; and (3) routinely monitors health plans' financial and
operational performance. Finally, AHCCCS is about to implement a new
quality management system that will emphasize the use of
outcome-based clinical measures.
BENEFICIARIES SATISFIED WITH
AHCCCS
---------------------------------------------------------- Letter :5.1
On the basis of our discussions with beneficiary representatives and
HCFA officials, our reviews of patient satisfaction surveys and
quality studies, and the data on how often beneficiaries change
health plans, the program's emphasis on cost containment appears not
to have adversely affected the care provided to Arizona Medicaid
beneficiaries. According to officials of advocacy groups
representing Arizona's indigent population, Medicaid beneficiaries
receive the same level of medical care as others in the state. HCFA
officials said that they did not know of any concerns about
beneficiary dissatisfaction with the care provided under the program.
Further, they said that Arizona has a quality management program in
place that fulfills HCFA regulations and all conditions of the
waiver.
A Flinn Foundation beneficiary satisfaction survey conducted in
1989--the most recent comprehensive survey available--found that
nearly 90 percent of adult current and former beneficiaries were
satisfied with the care they received through AHCCCS.\15 A February
1995 survey also indicates beneficiary satisfaction with AHCCCS,
although this survey is limited in scope. The survey, conducted by
the Arizona State University Survey Research Laboratory, found that
over 85 percent of contacted beneficiaries were very or completely
satisfied with the prenatal and maternity care they received.
Relatively few beneficiaries voluntarily change health plans during
the annual open season, which, according to AHCCCS officials, is
another indicator of beneficiary satisfaction. In 1993, only 6
percent of beneficiaries changed health plans during the annual open
season. In 1994, mainly because of new health contracts, 16 percent
of the beneficiaries voluntarily changed plans, but in many cases
their providers remained the same. In 1995, only 4.4 percent of the
total acute care population changed health plans during open
enrollment.
--------------------
\15 Another Flinn Foundation survey scheduled for publication in 1995
will include a measure of AHCCCS beneficiary satisfaction.
IN AHCCCS' BID PROCESS,
ACCESS AND QUALITY FACTORS
OUTWEIGH COST
---------------------------------------------------------- Letter :5.2
Although the competitive bidding process is important in keeping
capitation rates low, two-thirds of the scoring comprises access and
quality items. For example, the process considers the extent of a
health plan's provider network, including the number, type, and
geographic location of its physicians. AHCCCS also reviews a health
plan's ability to meet the contract requirements in areas such as
member and provider services and quality management. AHCCCS also
considers a bidder's ability to perform the administrative tasks of
the contract and its financial ability to meet all the contract
terms.
By establishing a minimum allowable capitation rate, AHCCCS prohibits
unreasonably low bids that might force health plans to curtail
services and adversely affect the quality of care provided. This
precludes harmful price competition among health plans.
WELL-DEVELOPED NETWORKS
ENSURE ACCESS TO NEEDED
SERVICES
---------------------------------------------------------- Letter :5.3
Two factors are responsible for Arizona's Medicaid beneficiaries'
good access to medical care--a high level of provider participation
in AHCCCS and AHCCCS' minimum provider network requirements.
In Arizona, 81 percent of licensed and practicing physicians are
registered Medicaid physician providers. Similarly, 70 percent of
obstetricians and 71 percent of pediatricians participate in
Medicaid. This level of participation is generally found in all of
the state's 15 counties. Of Arizona's 104 hospitals, 96 participate
in AHCCCS. According to health providers and their representatives,
reasonable payment rates and the assurance of getting paid for the
services provided through a monthly prospective payment system
account for the high level of physician and hospital
participation.\16
AHCCCS' minimum network requirements ensure beneficiaries' access to
appropriate care, even in rural areas. Contracts stipulate the type
and location of providers in each county, with minimum
provider-enrollee ratios. Beneficiaries have access to all levels of
services, including obstetrics and other specialist services, through
their primary care provider. The state also requires that plans make
transportation available, when necessary, so that beneficiaries have
access to needed services.
--------------------
\16 The willingness of Arizona providers to participate in AHCCCS may
also be partly due to the market alternatives they face. As of 1993,
penetration of private managed care was high in the state, 28 percent
compared with a national rate of 19 percent. Thus, Arizona
physicians, facing the alternative of a large private managed care
market, may be more willing to participate in Medicaid managed care
than physicians in other states who derive a greater share of their
business from the fee-for-service sector.
AHCCCS OVERSEES HEALTH PLAN
PERFORMANCE
---------------------------------------------------------- Letter :5.4
As mandated by HCFA, AHCCCS monitors health plans' financial and
operational performance. Health plans are required to submit
periodic financial and encounter data to the state. Annual reviews
by AHCCCS ensure that plans are financially stable and help the state
assess the level of care that Medicaid beneficiaries are receiving.
Studies are also conducted to measure the appropriateness of care in
specific areas--such as immunizations and prenatal care.
AHCCCS continues to refine its data collection efforts and improve
its ability to monitor health plans. For example, AHCCCS discovered
that health plans reimbursed providers for each patient's prenatal
care and delivery in one lump sum, which discouraged providers from
recording each separate encounter with the patient. As of October
1995, AHCCCS requires providers to report each encounter separately
so that it can accurately monitor the frequency of patients' prenatal
visits.
AHCCCS is developing a new quality management system that will gather
standardized encounter data from all plans and increase the emphasis
on outcome-based clinical standards. This will further enhance its
capacity to ensure that plans provide appropriate care. AHCCCS
expects this system to be operational in October 1997. According to
HCFA officials, Arizona is a leader among all the states in
developing encounter-based data to support a new quality management
system. This system would produce such indicators as prenatal care
and birth weights.
SEVERAL CONDITIONS CONTRIBUTE
TO AHCCCS' SUCCESS
------------------------------------------------------------ Letter :6
Arizona's Medicaid program, operating under a waiver from certain
federal requirements, has succeeded in containing costs while
providing beneficiaries access to what state officials and health
providers describe as mainstream medical care. Several findings
supply evidence of lower costs, widespread provider interest in
serving Medicaid patients, and beneficiary satisfaction with the
program.
Arizona's AHCCCS program can serve as a model for other Medicaid
programs. Rapid escalation in Medicaid costs has prompted many
states to search for new ways to control spending, including moving
more beneficiaries into managed care delivery systems. No state,
however, is as advanced as Arizona in using market forces to control
cost growth. Although each state Medicaid program is unique, states
converting from a fee-for-service to a managed care program can learn
from Arizona's experience. Our study of Arizona's Medicaid program
suggests that the following conditions, regardless of the state, are
necessary for an effective program.
Flexibility to deviate from constraining regulations: The
flexibility granted under the section 1115 waiver allowed
Arizona to encourage the development of managed care in ways
that are not possible under current HCFA regulations. Because
of the waiver, Arizona can mandate that beneficiaries enroll in
managed care plans. Also, health plans can be created that
serve Medicaid patients exclusively. Without a section 1115
waiver, more than 25 percent of each health plan's enrollees
would have to be non-Medicaid patients. This "75-25" rule was
originally designed to help ensure that Medicaid patients
received an acceptable quality of care because health plans
would have to attract a minimum share of private patients.
Arizona has decided to allow plans to have 100 percent of
Medicaid enrollees and instead sets quality standards and
monitors plan performance directly.
Development and use of market forces: In contrast to
Medicare--which uses administratively set capitation rates to
pay HMOs--and other Medicaid programs, Arizona's Medicaid
program relies on marketplace competition to determine
capitation rates. The state gives plans an incentive to submit
the lowest reasonable bid on capitation rates, excluding plans
that propose rates below an invisible floor--to protect against
impractically low bids. This system enables Arizona to buy the
best priced plans. It also allows health plans to earn
potential profits, which encourages the interest of both
existing and new plans in serving Medicaid patients.
Controls to protect beneficiaries from inadequate care: Arizona
takes several steps to protect beneficiaries. First, all health
plans must meet several basic requirements, including minimum
numbers of participating providers to ensure beneficiaries'
access to medical care. Second, in awarding contracts, Arizona
assigns more points to factors that influence the quality of
care that a plan can deliver than it does to the capitation rate
bid. Third, the state continuously monitors health plans'
performance.
Investment in data collection and analysis capabilities: Arizona
collects cost, profitability, and patient encounter data from
each health plan. This information is then analyzed to help the
state meet its goals of controlling costs and providing
appropriate health care. On the basis of these data,
independent actuaries estimate the cost of serving Medicaid
patients in each county. This estimated cost is then used to
evaluate the bids received at the start of each contract cycle.
These data also enable the state to oversee the care provided to
beneficiaries and protect against underprovision of medical
services.
The final lesson of Arizona's program is that other states are likely
to encounter problems changing from a fee-for-service delivery
system--in which the Medicaid administrators primarily determine
beneficiary eligibility and act as third-party payers--to a managed
care system that requires administrators to develop market forces and
carefully monitor the care provided. As noted in our earlier
reports, Arizona faced several difficulties along the way. In
particular, attempts to contract out the program's administration and
data information system failed. Furthermore, other states'
competitive managed care programs will likely need to evolve, as did
AHCCCS, to maximize their effectiveness in their own economic
environment. One such change Arizona made was to limit the number of
contracts awarded in each county. One of our earlier reports
recommended this: because it increases plans' possibility of not
winning a contract, it strengthens their incentives to submit low
bids. Arizona has been expanding and refining its market-driven
approach to Medicaid since 1982; it is unlikely that other states
could adopt equally successful programs without substantial
preparation and development of the capabilities Arizona now has.
AGENCY COMMENTS
------------------------------------------------------------ Letter :7
We obtained comments from HCFA and AHCCCS officials on a draft of
this report, and they agreed with our findings. Technical comments
that HCFA and AHCCCS provided have been incorporated where
appropriate.
---------------------------------------------------------- Letter :7.1
We are sending copies of this report to the Secretary of Health and
Human Services, the Administrator of HCFA, AHCCCS officials, and
other interested parties. We will also make copies available to
others upon request.
Please call me on (202) 512-7123 or James Cosgrove, Assistant
Director, on (202) 512-7029 if you or your staff have any questions
about this report. Karyn Papineau, Tony Padilla, and Mary Needham
were major contributors to this report.
Sincerely yours,
William J. Scanlon
Associate Director,
Health Financing Issues
ARIZONA HEALTH CARE COST
CONTAINMENT SYSTEM MANAGED CARE
CONTRACTS AWARD PROCESS
=========================================================== Appendix I
The Arizona Health Care Cost Containment System's (AHCCCS) Office of
Managed Care is responsible for the procurement process used to
competitively select the health plans that will provide medical
services to Medicaid beneficiaries. The nearly year-long process
starts with a bidders' conference where potential bidders learn how
the process works and continues with bid submissions, evaluation of
bids, contract awards, and readiness reviews before contract
implementation.
AHCCCS completed its most recent procurement of managed care
contracts in 1994, covering October 1, 1994, to September 30, 1997.
Previously, contracts were awarded for a 2-year period. In the last
cycle, contracts were competitively awarded on a county-by-county
basis. The major steps in the 1994 contract process are described
below.
BIDDERS' CONFERENCE
--------------------------------------------------------- Appendix I:1
After advertising statewide that it would issue a request for
proposal (RFP) to health plans to provide medical services to
Medicaid beneficiaries, AHCCCS invited all potential bidders to a
bidders' conference in March 1994. AHCCCS officials briefed
attendees and answered questions on the Medicaid program, contract
requirements, the award process, and responding to the RFP.
To facilitate the preparation of capitation proposals, AHCCCS
provided each bidder with a Data Supplement Book. AHCCCS officials
said that this book contained historical medical service utilization
data. However, AHCCCS stated that each bidder would be solely
responsible for researching, preparing, and documenting its
capitation proposal and that this book should not be used as the sole
source of information in making decisions about the capitation
proposal.
ISSUANCE OF THE RFP
--------------------------------------------------------- Appendix I:2
On March 7, 1994, AHCCCS issued its RFP describing the requirements
that a health plan must meet to deliver health care services to
eligible recipients. The RFP instructed bidders in preparing a
response to the state. Bidders were required to submit a proposal
for each county in which they wished to compete for a contract. All
interested health plans were required to respond to the RFP by June
1, 1994. For the contract cycle beginning October 1, 1994, 21 health
plans submitted 95 bids for 42 Medicaid managed care contracts in 15
counties. The 95 bids were more than twice the number (44 bids)
received in the previous procurement cycle (1992).
EVALUATION OF PROPOSALS
--------------------------------------------------------- Appendix I:3
AHCCCS selected in-house staff with managed care expertise to serve
on teams to evaluate all bids submitted. AHCCCS evaluated the bids
submitted in four areas, which are assigned a percentage value of the
total score: (1) provider network, (2) capitation rates, (3)
program, and (4) organization. AHCCCS also contracted with private
consulting firms that independently verified the evaluation and
scoring done by AHCCCS' internal evaluation teams. The scores
received in each of the four areas were combined and weighted to get
a final score for each bidder, by county.
PROVIDER NETWORK SCORES
------------------------------------------------------- Appendix I:3.1
The provider network score included two categories--network
development and network management. Network development involves
developing contractual arrangements with a sufficient number of
providers capable of delivering high-quality covered contract
services to eligible recipients in a specified service area according
to AHCCCS standards. AHCCCS identified service area minimum network
standards describing location requirements by county. Network
management involves the health plan's process of communicating with
its network and monitoring and evaluating its providers.
CAPITATION RATE SCORES
------------------------------------------------------- Appendix I:3.2
Following HCFA requirements, AHCCCS contracted with an actuary to
analyze historical utilization data and certify the capitation rate
ranges that AHCCCS used to evaluate the capitation rates (prices)
submitted by each bidder. The actuaries developed ranges for each
eligibility group (for example, AFDC and SSI) in each of Arizona's 15
counties.
AHCCCS did not share with bidders the capitation rate ranges used to
evaluate the capitation rate bids. Competitors may have been asked
to submit two bids, an initial bid and a best and final bid. AHCCCS
scored both bids. The closer the bid was to the low end of the
range, the more points a bidder received. The higher the bid, the
fewer points the bidder received. When the bids are below the range,
but, in AHCCCS' view, not unreasonably low, the state considers the
bid at the low point on the range and assigns the corresponding
points. If a bid is excessively below the range, the state may
consider it nonresponsive and unacceptable. When an initial bid is
above the range (considered an unacceptable bid), a bidder may be
allowed to submit a second bid, the best and final offer. In
scoring, the first bid (the unacceptable bid) received the least
points. The best and final bid was scored on the basis of the bid's
ranking within the acceptable range.
Between submission of initial and final bids, AHCCCS officials
reviewed bidders' detailed cost and utilization data (for example,
estimates of hospital days, physician visits, administrative costs,
and profits) supporting the bidders' initial capitation rate bids.
Although AHCCCS pointed out to bidders specific items that required
adjustments on the basis of AHCCCS' knowledge of historical medical
service utilization and cost data, bidders are solely responsible for
deciding the capitation rate that they wish to bid.
AHCCCS evaluated the final bids the same way it scored the initial
bids. The initial and final bids received equal weight for
evaluation and scoring.
PROGRAM SCORES
------------------------------------------------------- Appendix I:3.3
For program issues, AHCCCS evaluated the following: (1) executive
management and staff, (2) medical director's role, (3) member
services, (4) provider services, (5) quality management, (6) maternal
child health/early periodic screening diagnosis and treatment/family
planning, (7) grievance and appeals, and (8) behavioral health.
AHCCCS also considered its past experience with continuing program
bidders in evaluating proposals in this area. It did this by
reviewing the results of its most recent operational and financial
reviews of these bidders. AHCCCS adds points to a score if a
bidder's past AHCCCS performance has significantly exceeded the
requirements of the RFP. AHCCCS subtracts points from a score if a
bidder's past performance has been poor. If AHCCCS does not have
past program experience with a bidder, then the program score is
based on the bidder's response to the RFP.
ORGANIZATION SCORES
------------------------------------------------------- Appendix I:3.4
For organization, reviewers examined the bidders' prospective ability
to perform the administrative tasks necessary to support the
contract's requirements. AHCCCS used financial planning and
financial viability criteria to evaluate and score this bid factor.
AHCCCS reviewed in detail the many management and administrative
systems, including a bidder's encounter data reporting system,
financial reporting system, contracting and subcontracting process,
grievance standards and process, and medical records.
CONTRACTS AWARDED
--------------------------------------------------------- Appendix I:4
On July 11, 1994, AHCCCS awarded 42 contracts to 14 health plans.
AHCCCS awarded 2 contracts in each of its 13 rural counties. In
Maricopa County, where Phoenix is located, 10 contracts were awarded.
In Pima County, where Tucson is located, six contracts were awarded.
The current contract period runs from October 1, 1994, to September
30, 1997. Seven bidders, including two health plans that previously
had managed care contracts, did not win any contracts.
READINESS REVIEW
--------------------------------------------------------- Appendix I:5
Between submission of contract award and the start of health care
delivery, (for example, July 11, 1994, to Oct. 1, 1994) AHCCCS
conducted operational and financial readiness reviews for new
successful bidders. The purpose of readiness reviews is to assess
new contractors' readiness and ability to provide contract services
to members at the start of the contract period.
Although AHCCCS gives new contractors until the end of the first
quarter of the contract year (for example, Oct. 1, 1994, to Dec.
31, 1994) to complete pending readiness items, a new contractor is
permitted to begin operations only if the readiness review concerns
have been addressed to AHCCCS' satisfaction. An example of a pending
item would be the unsuccessful testing of a health plan's data system
to collect encounter information.
AHCCCS' OVERSIGHT OF HEALTH PLAN
PERFORMANCE
========================================================== Appendix II
HEALTH PLAN REPORTING
REQUIREMENTS
-------------------------------------------------------- Appendix II:1
HCFA mandates that AHCCCS conduct annual reviews of its contracted
health plans to determine their financial and operational stability.
AHCCCS has established financial and operational standards for the
health plans and their providers to ensure that beneficiaries receive
appropriate care. The state conducts on-site financial and
operational reviews at least annually to ensure that health plans
comply with operational and financial standards. In addition, AHCCCS
monitors health plans' data (for example, financial and encounter
data) submitted through monthly, quarterly, and annual reports.
The on-site reviews identify potential deficiencies that affect the
delivery, quality, or integrity of services or care. Corrective
action plans are developed to resolve any identified deficiencies.
These plans must state and describe the corrective action, the agency
responsible for implementing the corrective action, and the suggested
date for implementing the corrective action. AHCCCS monitors a
contractor's progress in implementing these corrections and provides
technical assistance if necessary.
AHCCCS' financial review ensures the reliability of a plan's
accounting systems, claims processing and encounter reporting
systems, and recovery systems and the stability of its financial
position. It also reviews financial management operations, such as
reinsurance, and evaluates any proposed changes to the program such
as new subcontractors and new management companies. AHCCCS also
requires each health plan to submit an annual audited financial
statement.
The operational segment of AHCCCS' annual on-site review documents
how a health plan manages and monitors its delivery system and the
quality of care as measured by standards developed by AHCCCS' Office
of the Medical Director (OMD). According to a HCFA official, these
standards reflect proposed standards issued by HCFA in 1993 for
Medicaid managed care programs. The OMD policy requires health plans
to develop quality management plans that articulate requirements;
promote quality improvement; comply with federal, state, and AHCCCS
requirements; ensure the participation of a health plan's members and
providers; and ensure health plan leadership participation.
QUALITY AND UTILIZATION
MANAGEMENT REQUIREMENTS
-------------------------------------------------------- Appendix II:2
AHCCCS requires that health plans' quality management plans
incorporate quality indicators to monitor performance in areas such
as preventive health, women's health, birth outcomes, prenatal care,
and accessibility and availability of services. OMD has established
standards adopted from commonly accepted medical standards of care
and relies on performance goals established by the Department of
Health and Human Services. Health plans must also have systems to
detect underutilization as well as overutilization. Health plans are
responsible for providing feedback to health professionals and health
plan staff regarding performance.
Health plans are also required to complete an annual quality of care
study on a topic approved by AHCCCS. Topics have varied, including
subjects such as the review of pharmacy utilization by plan members,
specialty physician referrals documentation, and the appropriateness
of emergency room use. According to AHCCCS officials, the study
results are used to improve the quality of services provided.
AHCCCS oversees health plans' quality management by monitoring member
and provider grievances; monitoring networks; and reviewing various
monthly, quarterly, and annual reports on plans' adherence to
performance measures.
QUALITY STUDIES COMPLEMENT
QUALITY MANAGEMENT EFFORTS
------------------------------------------------------ Appendix II:2.1
AHCCCS has completed various studies on quality issues, some of which
have been completed to meet federal or state legislative
requirements. These studies compare the care provided to AHCCCS
members with national standards and point out the need to develop
reliable data and constantly improve the system of care to
beneficiaries.
In 1995 AHCCCS completed the second year of a state-required annual
review comparing the immunization levels for AHCCCS members under 6
years of age with national data and Centers for Disease Control and
Prevention standards. The review found that the percentage of
children receiving immunizations increased from 1993 to 1994,
although the rates need further improvement to meet HHS' 90-percent
immunization goal. In addition to reviewing immunization rates, the
second year's study also identified reasons for beneficiaries' not
receiving recommended immunizations. The researchers found that
providers missed opportunities to give immunizations. AHCCCS is
sharing this information with health plans for use in their provider
education efforts.
In another study, AHCCCS looked at the underutilization of dental
services. The study found that AHCCCS' requirement to have health
plans refer beneficiaries to dentists was deterring beneficiaries
from using available dental services. AHCCCS implemented a new
policy on October 1, 1994, to allow beneficiaries direct access to
dentists.
NEW QUALITY PROGRAM BEING
DEVELOPED
------------------------------------------------------ Appendix II:2.2
AHCCCS is developing a new quality management program that reflects
current national trends on quality management in Medicaid programs.
With its most recent waiver extensions, HCFA has encouraged Arizona
and other states to improve their use of encounter data in quality
management programs. According to HCFA officials, the recommendation
to change AHCCCS' program reflects increased interest in quality
standards for Medicaid programs at the national level rather than
problems with Arizona's current system.
AHCCCS is adopting new quality standards to reflect HCFA's national
initiative to develop standards for the Medicaid population. Changes
to the current standards for commercial managed care plans are needed
to reflect, for example, Medicaid's enrollment fluctuations. In
addition, AHCCCS plans to standardize the information received from
its health plans. This effort will ultimately be used to monitor and
compare participating health plans' performance.
This new standardizing of information has four components that
measure both outcomes and health plan operations. Clinical
indicators, financial indicators, member satisfaction, and provider
satisfaction will be used to evaluate health plans. The
standardizing of information from the health plans, the main
difference between current efforts and the new plan, will improve
monitoring efforts and reduce the current time-intensive reliance on
on-site record review.
As of July 1995, AHCCCS was completing the clinical indicators for
monitoring the health plans and establishing baseline data. It is
updating its computer systems to support the increased reliance on
encounter data. AHCCCS plans to implement the new standards for the
next health plan contract cycle beginning on October 1, 1997.
HISTORY OF AHCCCS
========================================================= Appendix III
Before 1982, Arizona was the only state in the nation that did not
participate in the federal Medicaid program. Health care for
low-income people was provided and funded by county governments, with
care provided by county hospitals and clinics or through contracted
providers. Each county set its own income and resource guidelines
with certain minimum standards set by the state. The range of
services also varied from county to county.
In 1981 Arizona legislators recognized that the counties could not
continue to pay the full cost for health care and began to explore
options that would relieve the counties and, for the first time,
bring federal Medicaid dollars to the state. In 1981, the state
government passed legislation to create AHCCCS as the first statewide
Medicaid managed care system. In 1982 Arizona sought approval from
HCFA to operate AHCCCS under a section 1115 demonstration waiver.
Named for section 1115 (a) of the Social Security Act, 1115 waivers
enable states to operate differently from conventional Medicaid in
certain ways, such as limiting beneficiaries' choice of providers.
On July 13, 1982, HCFA approved AHCCCS as a 3-year acute medical care
demonstration project, and the state implemented the program on
October 1, 1982. AHCCCS became the first statewide Medicaid managed
care system in the nation based on a prepaid, capitated financing
arrangement with private health plans. Arizona's intent was to
create a managed care delivery system that would deliver quality
services, control costs, discourage the use of emergency rooms for
primary care, and avoid the fraud and abuse reported in
fee-for-service programs.
Initially, AHCCCS was required to provide medical services only to
the federally mandated eligible group for which Arizona received
federal matching funds (recipients of AFDC and SSI), recipients
commonly referred to as "categorical."
Today, AHCCCS still serves these groups as well as women and children
covered under the 1987 Sixth Omnibus Reconciliation Act.
Medicaid-eligible American Indians can choose to receive services
from the Indian Health Service or, off reservation, from an AHCCCS
health plan. The Arizona Department of Economic Security determines
eligibility for AFDC. The Social Security Administration determines
eligibility for SSI. As of August 1, 1995, over 432,000
beneficiaries were enrolled in the acute care program.
Initially, AHCCCS was also required to provide all the federally
mandated Medicaid services except for skilled nursing facility care,
home health care, nurse midwife services, family planning services,
and nonacute mental health services. Today, AHCCCS provides these
services.
AHCCCS also provides services to groups of low-income people who do
not qualify for Medicaid and who are funded entirely by state and
county dollars. These groups are the medically needy/medically
indigent, eligible assistance children; eligible low-income children;
and undocumented individuals, who receive some emergency services.
These groups of recipients are commonly referred to as
"noncategorical." These state-funded groups generally receive the
same acute care services available to the federally funded
populations except for comprehensive behavioral health services and,
until recently, heart, liver, and bone marrow transplants.
Initially, AHCCCS' day-to-day operations were carried out by a
private contractor selected through a competitive procurement to act
as the AHCCCS administrator. The administrator's responsibilities
included procuring and monitoring providers, establishing and
monitoring medical quality assurance systems, enrolling
beneficiaries, maintaining provider relations, providing technical
assistance to health plans, and collecting and compiling reports
using claim and utilization data.
After a little more than a year, on March 15, 1984, Arizona was
forced to cancel the contract because of disputes. The AHCCCS
Division of Arizona's Department of Health Services took over as
administrator. After this, the state terminated two health plan
contracts due to plan insolvency. In addition, another plan was
reorganized with new management under the federal bankruptcy
statutes.
Subsequently, the AHCCCS Division became a separate agency reporting
directly to the governor. The AHCCCS structure was also changed so
that the Division could assume a stronger regulatory role. AHCCCS'
new challenges included financial and contractual compliance reviews
of health plans; quality control review of the county eligibility
systems; medical quality of care audits of the health plans; and
increased staffing for the audit, compliance, and utilization review
functions.
We first reported on AHCCCS\17 in June 1984. We testified on HCFA's
monitoring of certain aspects of the AHCCCS program before the
Subcommittee on Health and the Environment, Committee on Energy and
Commerce, U.S. House of Representatives. At that time, we reported
that AHCCCS had not generated the program information necessary to
render an opinion on the financial performance of health plans, the
quality of care provided, or the reasonableness of payments to
providers.
In response to GAO's recommendations, both HCFA and AHCCCS began the
steps necessary to obtain data essential to evaluating AHCCCS.
Although HCFA informed AHCCCS that it would not approve an extension
of its waiver unless the state produced complete and accurate
utilization data, HCFA told us that as long as AHCCCS made
significant progress in implementing reforms and improvements
necessary for producing the information, HCFA would not terminate the
program. In addition to providing HCFA with some utilization data
dating back to 1982, AHCCCS submitted a plan detailing the steps it
would take to meet HCFA's requirements.
In November 1985, we reported that many AHCCCS health plans had not
complied with federal requirements for disclosure of ownership
information. We reported that some AHCCCS plans either had not
disclosed direct or indirect ownership interests or had not disclosed
officers or directors. We recommended that the Secretary of HHS
direct the Administrator of HCFA to review AHCCCS plan contract
proposals and renewal submissions and determine the extent to which
federal financial participation should not be available for plans
that did not comply with disclosure laws for ownership and control
information or related-party transactions. Our other recommendations
were that AHCCCS and HCFA institute procedures to ensure that AHCCCS
plans comply with these disclosure requirements in the future.
In response to GAO's recommendations, HCFA and AHCCCS said that they
would act to ensure that future health plan contracts follow
disclosure requirements. AHCCCS currently has a reporting guide that
establishes the monthly, quarterly, and annual reporting requirements
for acute care contracting health plans. The guide requires reports
on owners, officers, directors, related-party transactions, and full
financial disclosure and also establishes financial penalties if the
health plan contractors fail to comply with these requirements.
In March 1987, we reported on AHCCCS' first 3 years of operation
(Oct. 1982 through Sept. 1985). We addressed Arizona's approach to
(1) competitive bidding for procuring health plan contracts, (2)
collection of utilization data from the prepaid plans on the health
care services provided, and (3) financial oversight of the prepaid
health plans. In our March 1987 report, we stated that Arizona had
experienced many start-up problems that prevented an assessment of
the effectiveness of its cost containment features.
We reported that states planning on using prepaid health programs
should, among other things, (1) develop adequate financial and
utilization reporting systems and program controls before
implementing the program, (2) establish penalties for noncompliance
with reporting requirements, (3) establish requirements to
demonstrate the financial viability of prepaid health plans and
devote adequate resources to monitoring health plans' performance,
and (4) design health plan procurements to promote competition. As
previously indicated, AHCCCS officials have acted to resolve these
problems. Today, for example, AHCCCS limits the number of contracts
it awards in each rural county to two--a recommendation we made to
promote competition and reduce costs.
In 1986 AHCCCS contracted with a private consulting firm to design
and develop a prepaid medical management information system (PMMIS).
Contract disputes resulted in the termination of this contract in
1990 as the system was set to enter the testing phase. In May 1990,
AHCCCS took over the testing, user training, conversion, and
implementation of PMMIS. AHCCCS officials told us that its
management information system has for several years enabled them to
collect and analyze adequate financial and utilization data to assess
plans' finances and set sound capitation rates. They added that such
data are beginning to be used to conduct outcome-based monitoring and
evaluation of the quality of care being provided to beneficiaries.
In 1981 Arizona passed authorizing legislation allowing small
employers with up to 25 employees to purchase health care from
AHCCCS. This program began in 1988, and in 1991 the employee limit
was raised from 25 to 40. AHCCCS currently has contracts with four
health plans to provide health coverage under its small employer
program called Healthcare Group of Arizona. Currently, about 5,800
employers are enrolled providing care for over 18,000 employees and
dependents. AHCCCS does not receive federal financial assistance for
this program.
In 1989 AHCCCS implemented the Arizona Long Term Care System (ALTCS),
which offers long-term care, acute care, and home and community-based
services to the elderly or physically and developmentally disabled.
As of August 1995, over 20,000 people were enrolled in ALTCS.
In March 1993, we reported on various states' managed care program
initiatives including Arizona's. We described (1) states' use of
managed care programs; (2) the difficulty states face in implementing
certain program components; (3) the effect of the managed care
approach on health care access, quality, and cost; and (4) the
presence of features that ensure the quality of health services and
providers' financial stability.
AHCCCS began its 14th year on October 1, 1995. It continues today
under an August 1994 waiver extension as a statewide Medicaid managed
care demonstration project. The most recent waiver extension
authorizes AHCCCS to operate until October 1, 1997.
In March 1995, AHCCCS submitted to HCFA an amendment to its waiver,
contingent on approval by the Arizona legislature, which would
streamline the eligibility determination process and offer health
care services to low-income and working poor individuals who have
income levels up to 100 percent of the federal poverty level. The
Arizona legislature did not approve the AHCCCS proposal.
--------------------
\17 The Health Care Financing Administration's Monitoring of the
Arizona Health Care Cost Containment System (GAO/HRD testimony, June
15, 1984; Arizona Medicaid: Nondisclosure of Ownership Information
by Health Plans (GAO/HRD-86-10, Nov. 22, 1985); Medicaid: Lessons
Learned From Arizona's Prepaid Program (GAO/HRD-87-14, Mar. 6,
1987); and Medicaid: States Turn to Managed Care to Improve Access
and Control Costs (GAO/HRD-93-46, Mar. 17, 1993).