Medicare HMOs: Potential Effects of a Limited Enrollment Period Policy
(Letter Report, 02/28/97, GAO/HEHS-97-50).

Pursuant to a congressional request, GAO reviewed how a limited
enrollment period would affect the Medicare program, private health
plans, beneficiaries, and employers who provide Medicare supplemental
benefits to retirees, focusing on: (1) the growth of Medicare's managed
care program; (2) employers' attempts to administer their respective
benefits seasons; (3) taxpayer savings measured against beneficiary
protections; and (4) the resources needed by the Health Care Financing
Administration (HCFA), which runs Medicare's day-to-day operations.

GAO noted that: (1) changing Medicare's current policy that allows
beneficiaries to switch among health maintenance organizations (HMO) or
between an HMO and fee for service monthly would have far-reaching
consequences for the Medicare program, beneficiaries, HMOs, employers,
and HCFA; (2) the specific effects would depend on the limits placed on
switching plans; (3) any change that restricts beneficiary opportunities
to enroll or disenroll would likely slow the growth of Medicare managed
care; (4) a limited enrollment period for Medicare could have two
principal advantages: (a) to improve the quality and distribution of
managed care information to beneficiaries: a focused enrollment period
would create a natural opportunity for HCFA to provide objective,
comparative information about health plans; and (b) to make impractical
the current practice of in-home sales of HMOs, a source of marketing
abuses, which are difficult for HCFA to deter; (5) a limited enrollment
period could also have several of the following disadvantages, the
combined effect of which could slow Medicare managed care enrollment
growth: (a) lessen the effectiveness of marketing Medicare HMOs: HMOs
would likely focus more of their marketing dollars on mass media
campaigns concentrated around Medicare's enrollment season, but
beneficiaries unfamiliar with managed care might not receive enough
specifics through mass marketing to appreciate any advantages offered by
an HMO over traditional fee-for-service Medicare; (b) lessen the
attractiveness of HMOs to beneficiaries: the only choice available to
dissatisfied HMO enrollees might be to change to fee for service and pay
either Medicare's deductibles and coinsurance or, if available, premiums
for a supplemental Medigap policy; and (c) pose considerable
administrative obstacles for employers: accommodating Medicare's
schedule could be so administratively difficult that some employers
might simply stop offering a managed care option to their retirees; (6)
limiting beneficiaries' option to change to fee-for-service Medicare
except during the officially appointed open season could also produce
the following mixed effects: (a) Medicare might achieve modest savings
on money now spent on services for HMO members who change to fee for
service; and (b) beneficiaries would lose an important consumer
protection and might be less willing to enroll in managed care; (7) ult*

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-97-50
     TITLE:  Medicare HMOs: Potential Effects of a Limited Enrollment 
             Period Policy
      DATE:  02/28/97
   SUBJECT:  Health care programs
             Health insurance cost control
             Managed health care
             Health maintenance organizations
             Employee benefit plans
             Marketing
             Employee medical benefits
             Consumer protection
IDENTIFIER:  Medicare Program
             California
             Florida
             New York
             Medigap
             Medicare Risk Contract Program
             Federal Employees Health Benefits Program
             HCFA Medicare Competitive Pricing Demonstration Project
             Miami (FL)
             Los Angeles (CA)
             CalPERS
             
******************************************************************
** 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, Committee on Finance, U.S.  Senate

February 1997

MEDICARE HMOS - POTENTIAL EFFECTS
OF A LIMITED ENROLLMENT PERIOD
POLICY

GAO/HEHS-97-50

Limited Enrollment Period for Medicare

(101392)


Abbreviations
=============================================================== ABBREV

  AAPCC - adjusted average per capita cost
  BBA - Balanced Budget Act of 1995
  CalPERS - California Public Employees' Retirement System
  FEHBP - Federal Employees' Health Benefits Program
  HCFA - Health Care Financing Administration
  HHS - Department of Health and Human Services
  HMO - health maintenance organization
  NCS - National Computer System
  OPM - Office of Personnel Management

Letter
=============================================================== LETTER


B-275832

February 28, 1997

The Honorable William V.  Roth, Jr.
Chairman, Committee on Finance
United States Senate

Dear Mr.  Chairman: 

Medicare, unlike most employer-sponsored health insurance coverage,
allows beneficiaries the flexibility to change managed care plans or
switch to fee-for-service arrangements monthly.  This flexibility can
cause problems for the Medicare program.  For example, under this
policy beneficiaries may decide to use managed care or other private
plans while in relatively good health but may disenroll to fee for
service when their health care needs increase.  The result can be a
disproportionate number of less healthy beneficiaries in fee for
service, excess payments to health maintenance organizations (HMO),
and unnecessary Medicare spending. 

Recently, the Congress has considered making Medicare's policies more
consistent with those of other large health care purchasing
organizations by establishing a limited time each year during which
Medicare beneficiaries could enroll in a particular plan and by
restricting disenrollment outside that period.  To help the Congress
in its consideration of the effects of such a policy change, you
asked us to assess how a limited enrollment period would affect the
Medicare program, private health plans, beneficiaries, and employers
who provide Medicare supplemental benefits to retirees.  To do this,
we examined the potential effects of policy changes on (1) the growth
of Medicare's managed care program, (2) employers' attempts to
administer their respective benefits seasons, (3) taxpayer savings
measured against beneficiary protections, and (4) the resources
needed by the federal agency that runs Medicare's day-to-day
operations.  Because a specific annual enrollment period could be
established without also limiting beneficiaries' opportunities to
change to fee for service, we have discussed the effects of the two
policy changes (limiting the enrollment period and limiting
disenrollment opportunities) separately. 

We assumed that a new Medicare enrollment policy might be similar,
but not necessarily identical, to the provisions contained in the
proposed Balanced Budget Act of 1995 (BBA), H.R.  2491, as discussed
in the act's accompanying conference report.\1 Therefore, we
developed and analyzed a limited enrollment period policy modeled on
the BBA.  Although the BBA would have provided for an expanded range
of health care delivery and insurance options, we focused our
attention on risk HMOs because they currently serve most
beneficiaries not in Medicare fee for service.\2 To assess the likely
effects of such a policy, we interviewed representatives of 10
Medicare risk HMOs, the Health Care Financing Administration (HCFA),
employers who offer managed care options to retirees, and Medicare
beneficiary advocacy organizations.  We surveyed HMOs with Medicare
risk contracts regarding their employer group business.  To develop
information on the potential financial effect on Medicare of limiting
disenrollment, we analyzed HMO disenrollment data and fee-for-service
claims in California.  For more detailed information on our
methodology, see appendix III.  We did our work from March through
November 1996 in accordance with generally accepted government
auditing standards. 


--------------------
\1 House Conference Report 104-350, pp.  1093-1102.  The act was
vetoed by the President; thus, these provisions did not become law. 

\2 The scope of our study reflects the fee-for-service alternatives
currently available to most beneficiaries, but many of the issues
discussed in our report pertain to analysis of other types of
insurance and delivery options under consideration. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

Changing Medicare's current policy that allows beneficiaries to
switch among HMOs or between an HMO and fee for service monthly would
have far-reaching consequences for the Medicare program,
beneficiaries, HMOs, employers, and HCFA.  The specific effects would
depend on the limits placed on switching plans.  Medicare could, for
example, emulate private insurance and establish a limited enrollment
period--that is, a set time each year when beneficiaries would choose
their health plan (a specific HMO or fee for service) for the coming
year--but not restrict opportunities for beneficiaries to change to
fee for service during the year.  Alternatively, Medicare could
combine a limited enrollment period with restrictions on changing to
fee for service.  Although both alternatives have advantages, any
change that restricts beneficiary opportunities to enroll or
disenroll would likely slow the growth of Medicare managed care. 

A limited enrollment period for Medicare could have two principal
advantages. 

  -- To improve the quality and distribution of managed care
     information to beneficiaries:  A focused enrollment period would
     create a natural opportunity for HCFA to provide objective,
     comparative information about health plans--information that
     beneficiaries now lack. 

  -- To make impractical the current practice of in-home sales of
     HMOs, a source of marketing abuses:  Although HCFA is alert to
     instances of HMO sales personnel misrepresenting HMO benefits
     and obtaining enrollment signatures under false pretenses, these
     abuses are difficult for HCFA to deter. 

A limited enrollment period could also have several of the following
disadvantages, the combined effect of which could slow Medicare
managed care enrollment growth. 

  -- Lessen the effectiveness of marketing of Medicare HMOs:  HMOs
     would likely focus more of their marketing dollars on mass media
     campaigns concentrated around Medicare's enrollment season, but
     beneficiaries unfamiliar with managed care might not receive
     enough specifics through mass marketing to appreciate any
     advantages offered by an HMO over traditional fee-for-service
     Medicare.  The Florida and New York Medicaid programs saw their
     managed care enrollment decline significantly after banning
     direct marketing by HMOs.  Third-party contractors, if given
     exclusive responsibility for informing and enrolling
     beneficiaries under a limited enrollment period policy, might
     not be effective substitutes for health plans' sales agents. 

  -- Lessen the attractiveness of HMOs to beneficiaries:  Under a
     policy precluding the beneficiary's option to switch plans
     during a 12-month period, the only choice available to
     dissatisfied HMO enrollees might be to change to fee for service
     and pay either Medicare's deductibles and coinsurance or, if
     available, premiums for a supplemental Medigap policy.  The new
     policy could also make Medicare HMOs impractical for
     beneficiaries who live in more than one part of the country
     during the year. 

  -- Pose considerable administrative obstacles for employers: 
     Regardless of the particular time of year selected for Medicare
     enrollment activities, some employers currently offering their
     retirees a managed care health insurance option could find that
     their health benefits seasons did not coincide with Medicare's. 
     Accommodating Medicare's schedule could be so administratively
     difficult that some employers might simply stop offering a
     managed care option to their retirees. 

Limiting beneficiaries' option to change to fee-for-service Medicare
except during the officially appointed open season could also produce
the following mixed effects: 

  -- Medicare might achieve modest savings on money now spent on
     services for HMO members who change to fee for service. 
     Medicare HMO members who disenroll and change to fee for service
     tend to use more services and more costly procedures than the
     average beneficiary under fee for service.  Consequently,
     Medicare spends more money to serve an HMO member who changes to
     fee for service than it would have paid to the HMO to care for
     that beneficiary.  For example, we found that, for beneficiaries
     who switched from managed care to fee for service in California
     during 1994, Medicare paid almost $30 million more than it would
     have paid had these beneficiaries not been permitted to switch
     to fee for service mid-year.  However, these savings may appear
     modest when measured against total 1994 California Medicare HMO
     outlays of $4.2 billion. 

  -- Beneficiaries would lose an important consumer protection and
     might be less willing to enroll in managed care.  HMO members
     who are dissatisfied with their HMO may now change plans or
     switch to fee for service at the end of each month.  HMO
     representatives, HCFA officials, and beneficiary advocates
     believe that eliminating this option would deter some
     beneficiaries from joining a managed care plan.  HCFA and HMO
     officials predict that, because dissatisfied HMO members could
     not disenroll until the next open season, the number of managed
     care complaints, grievances, and appeals would rise
     dramatically. 

Ultimately, changing Medicare's HMO enrollment and disenrollment
policies could have unintended effects.  Although Medicare might
achieve modest savings, these savings could be offset if policy
changes also led to slowing or reducing the enrollment of Medicare
beneficiaries in HMOs. 


   BACKGROUND
------------------------------------------------------------ Letter :2

Consistent with the national trend toward managed care, the number of
Medicare beneficiaries enrolled in HMOs has grown significantly--from
about 1 million in 1987 to about 4 million in 1996.  This growth
represents an increase from about 3 percent of all Medicare
beneficiaries to about 10 percent.  About 90 percent of Medicare
beneficiaries enrolled in managed care are in risk-contract HMOs.\3
The largest growth in Medicare managed care enrollment has occurred
in the risk program.  (See fig.  1.)

   Figure 1:  Medicare Risk HMO
   Enrollment, 1987-96

   (See figure in printed
   edition.)

Note:  Enrollment is as of December of each year except for 1996, for
which November data are used. 

Source:  Monthly reports, Office of Managed Care, HCFA. 

The number of HMOs in the risk program fluctuated somewhat in the
program's first 5 years, but since 1992 the number of risk HMOs has
grown steadily.  (See fig.  2.) As of November 1996, HCFA had entered
into 238 risk contracts.  Most beneficiaries have at least one risk
HMO available in their area, and, in some markets, beneficiaries can
choose from as many as 14 different HMOs. 

   Figure 2:  Number of Risk HMO
   Contracts, 1987-96

   (See figure in printed
   edition.)

Note:  Enrollment is as of December of each year except for 1996, for
which November data are used. 

Source:  Medicare Managed Care Program Update, HCFA, 1995. 

Risk HMOs are required to offer at least one 30-day enrollment period
each year, but, in practice, most accept enrollment continuously. 
Although HCFA provides beneficiaries some general information about
HMOs when beneficiaries first become eligible for Medicare, they
typically learn about their options from the HMOs.  Unlike leading
private and public health care purchasing organizations, Medicare
does not provide its beneficiaries with comparative information about
available HMOs.  HMOs provide beneficiaries with enrollment forms,
collect the forms, and notify HCFA of enrollments.  Beneficiaries may
disenroll from a plan as often as once each month. 

As discussed in the BBA conference report, the BBA included
provisions that would have amended Medicare's enrollment policy in
the following ways: 

  -- Each October, Medicare would have an annual, coordinated
     election period, or "open season," during which beneficiaries
     could change their Medicare election.  Elections of coverage
     would become effective the following January 1.  However, newly
     eligible Medicare beneficiaries could elect coverage and have
     their choice become effective when they first became eligible
     for benefits. 

  -- The Secretary of the Department of Health and Human Services
     (HHS) would conduct a nationally coordinated educational and
     publicity campaign during October.  At least 15 days before the
     election period, the Secretary would mail all Medicare
     beneficiaries and prospective beneficiaries general election
     information and information comparing benefits, premiums, and
     measures of quality at available health plans. 

  -- Disenrollment could occur only within 90 days of the time
     elected coverage began.  Beneficiaries who disenrolled could
     elect a different HMO for the remainder of the year.  This
     disenrollment option would only apply the first time a
     beneficiary enrolled in a particular managed care plan and would
     not apply more than twice for any beneficiary in a calendar
     year.  Exceptions would include disenrollment for beneficiaries
     who moved out of a service area. 


--------------------
\3 HCFA, which administers the Medicare program, pays risk HMOs a per
capita premium to provide a full Medicare package of benefits,
regardless of the amount the HMO spends for each member's health
care.  Except for emergency and out-of-area urgent care, members must
receive all their medical care through the HMO's network of
providers.  The remaining 10 percent of beneficiaries enrolled in
managed care are in plans that Medicare reimburses on a cost basis
(cost HMOs) or in plans that only cover Medicare part B services and
may have restrictive enrollment policies (health care prepayment
plans). 


   A LIMITED ENROLLMENT PERIOD
   POLICY COULD SLOW MANAGED CARE
   GROWTH DESPITE BETTER CONSUMER
   INFORMATION
------------------------------------------------------------ Letter :3

Establishing a limited enrollment period could slow managed care
growth for two reasons.  First, marketing practices possible under a
limited enrollment policy might be less effective in attracting
beneficiaries to managed care.  These changes could have a positive
by-product, however, as the incidence of in-home sales and associated
abusive sales practices would likely diminish.  Second, restrictions
on health plan switches outside the established enrollment
period--even if no restrictions existed on changing to traditional
fee-for-service Medicare--could deter some beneficiaries from
enrolling in HMOs.  In particular, a limited enrollment period policy
would have three main disadvantages for beneficiaries:  (1)
dissatisfied beneficiaries and those encountering problems gaining
access to desired treatments could be exposed to higher health care
expenses, (2) beneficiaries who spend part of each year in a
different location ("snowbirds") could find they had no choice other
than fee for service, and (3) all beneficiaries enrolled in HMOs
could face delays in obtaining physician appointments at the start of
each benefit year because of a large volume of new beneficiaries
seeking services. 


      HMO MARKETING ABUSES
      ASSOCIATED WITH IN-HOME
      SALES WOULD PROBABLY BE
      REDUCED
---------------------------------------------------------- Letter :3.1

One-on-one sales presentations, often conducted in the privacy of
beneficiaries' homes, leave beneficiaries vulnerable to abusive sales
tactics and serious marketing problems.  Reported abuses include HMO
representatives' lying to prospective enrollees about the benefits of
HMO enrollment, pressuring beneficiaries to join HMOs, enrolling
beneficiaries who could not make informed enrollment decisions, and
obtaining enrollment signatures under false pretenses.  Although HCFA
cannot determine the frequency of these problems, agency officials
are concerned about the potential for in-home sales marketing abuses. 

According to our HMO survey results, about half of the beneficiaries
who enrolled in a Medicare HMO as individuals (not as members of an
employer group) in 1995 participated in a one-on-one sales
presentation.  However, the likelihood of a beneficiary's
participating in a one-on-one sales presentation varied greatly by
HMO.\4

A limited enrollment period lasting just 1 or 2 months each year
could make it impractical for HMOs to conduct as many in-home sales
presentations.  Each one-on-one meeting can last from 1/2 hour to 2
hours and is conducted by an HMO sales agent who sells only to
Medicare beneficiaries.  HMO representatives told us that sales
agents who sell Medicare plans sell them exclusively.  Agents are
trained not only in the details of their HMO's offering, but also in
traditional Medicare and the rules governing Medicare managed care. 
Some HMO representatives implied that maintaining a large, dedicated
Medicare sales force year-round would be impractical if most sales
would take place during a 1- or 2-month limited enrollment period.\5
Furthermore, HMO representatives said it would be unrealistic to
expect non-Medicare agents to be able to sell Medicare products. 
Because beneficiaries are particularly susceptible to abusive sales
practices in their homes, reducing or eliminating in-home sales
presentations would better protect beneficiaries from the possibility
of sales abuses.  This protection, however, would be a by-product of
the enrollment policy change and could be achieved by more direct
methods. 


--------------------
\4 For example, about one-fourth of the HMOs responding to our survey
reported that 90 percent or more of the beneficiaries who joined
their plan had participated in a one-on-one sales presentation.  In
contrast, one-fifth of HMOs reported that less than 10 percent of
their enrollees had participated in such presentations. 

\5 Some sales would occur outside the limited enrollment period. 
Depending upon how such a policy would be implemented, such sales
would probably include sales to newly entitled Medicare beneficiaries
and beneficiaries who had moved outside of their previous HMO's
service area. 


      COMPARATIVE INFORMATION
      DISTRIBUTED DURING A LIMITED
      ENROLLMENT PERIOD WOULD AID
      BENEFICIARY DECISION-MAKING
---------------------------------------------------------- Letter :3.2

Under the BBA, before the start of a limited enrollment period, the
Secretary of HHS would have been responsible for producing and
distributing (1) a list of plans available in a given area and (2)
comparative information about those plans, including benefits,
premiums, and measures of quality.  The Secretary would also have
been responsible for maintaining a toll-free number that
beneficiaries could call to receive specific information. 

Beneficiaries' ability to make informed health care choices would be
enhanced by the availability of objective, comparative information
and access to a hot line.  We recently reported that beneficiaries
who wish to compare plans face difficult, if not daunting, steps.\6
First, they must call a toll-free telephone number to obtain a list
of HMOs available in their area.  Next, they must contact those HMOs
and request marketing brochures.  Finally, they must compare plans'
benefit packages and cost information described in the brochures. 
The last step can be difficult because HMOs are not required to use
standard formats or terminology in describing their products. 

A limited enrollment period would facilitate an annual HMO marketing
campaign and create a natural opportunity for HCFA to distribute
comparative plan information to beneficiaries.  Some experts believe
that HMOs' concentrated advertising during the open season would help
inform beneficiaries of alternative Medicare options.  Another
potential advantage is that any comparative information produced by
HCFA would be up to date at the time most beneficiaries were making
health care choices.\7


--------------------
\6 Medicare:  HCFA Should Release Data to Aid Consumers, Prompt
Better HMO Performance (GAO/HEHS-97-23, Oct.  22, 1996). 

\7 HCFA cites the changing health care marketplace as one reason the
agency has no plans to distribute printed comparison charts directly
to beneficiaries.  However, HCFA is planning to make some basic HMO
comparative information available on the Internet.  The agency will
periodically update the information. 


      MASS MARKETING CAMPAIGN
      MIGHT NOT ADEQUATELY INFORM
      BENEFICIARIES UNFAMILIAR
      WITH MANAGED CARE
---------------------------------------------------------- Letter :3.3

HMO representatives told us that if Medicare established a limited
enrollment period, plans would turn to a marketing approach more
conducive to a limited enrollment time frame.  HMOs would focus more
of their marketing dollars on mass media campaigns--including print,
radio, and television advertising--concentrated around Medicare's
enrollment season. 

Some experts believe that a concentrated mass marketing campaign
could increase beneficiary awareness of Medicare options, including
managed care.  These experts suggest that the Medicare advertising
blitz could be similar to the advertising campaigns that occur in the
Washington, D.C., area during the Federal Employees' Health Benefits
Program (FEHBP) open season each fall.  Whether Medicare HMOs'
advertising campaigns would be as intense as FEHBP plans' is
uncertain.  FEHBP subscribers represent about 9 percent of the
Washington, D.C., metropolitan area's total population.\8 Nationwide,
Medicare beneficiaries represent about 14 percent of the total
population.  However, only about 1 in 10 Medicare beneficiaries
currently enrolls in managed care.  If advertising intensity is
driven by the proportion of potential customers, the intensity of a
campaign for Medicare beneficiaries would depend upon whether HMOs
believe the potential market is all Medicare beneficiaries or only 1
in 10. 

Representatives of HMOs, however, believe that an advertising
campaign without the benefit of one-on-one sales would be less
effective at convincing Medicare beneficiaries to try managed care. 
Representatives of most HMOs we contacted stated that limiting
Medicare's enrollment period would slow the growth of managed care
because plans would not (1) have time to educate beneficiaries about
Medicare's managed care option and (2) be able to hire enough trained
sales staff on a seasonal basis to answer beneficiary questions
during the limited enrollment period.  Although abuses have been
reported in conjunction with one-on-one sales, HMOs believe this
sales approach is both necessary and effective, in part because many
beneficiaries have had no experience with managed care. 

The effectiveness of an FEHBP-like mass marketing campaign for
Medicare may depend on whether HCFA develops ancillary mechanisms to
inform beneficiaries.  Participants in FEHBP do not rely exclusively
on mass marketing to obtain information.  All active and retired
FEHBP enrollees are given comparative information on available plans
and can obtain detailed, plan-specific information brochures that
follow a standard format.  Active federal workers can also discuss
their health care options with colleagues or their agency's benefits
administrator.  Furthermore, most workers can easily attend health
fairs sponsored by their agency, where health plan representatives
distribute literature and answer questions.  The 20 percent of FEHBP
members who are retired also have some advantages over individuals in
Medicare.  As former federal workers, FEHBP participants are familiar
with the program's enrollment and disenrollment rules.  In addition,
federal retirees receive guidance from the National Association of
Retired Federal Employees.  This organization, with over 1,700
chapters nationwide, works closely with FEHBP in answering questions
and resolving problems.  Finally, some members of the Congress
sponsor annual FEHBP health fairs attended by retirees. 


--------------------
\8 This percentage does not include dependents of active workers or
retirees. 


      THIRD-PARTY ENROLLMENT
      BROKERS ARE OBJECTIVE BUT
      MAY BE LESS EFFECTIVE
      PROMOTING HMOS
---------------------------------------------------------- Letter :3.4

Requiring third-party contractors, or brokers, to conduct all
enrollment activities would better protect beneficiaries from abusive
sales practices, minimize the opportunity for HMOs to favorably
select only the healthiest beneficiaries, and provide beneficiaries a
convenient source of objective information.  Beneficiaries might
welcome such a change in enrollment practices partly because they
would have the convenience of "one-stop shopping" and also appreciate
a source of objective, comparative information.  A recent focus group
conducted for HCFA found that most beneficiaries did not view
insurance plan representatives as trustworthy sources of impartial
information.  Nonetheless, HMO representatives maintain that personal
contact with an HMO sales agent can be reassuring to beneficiaries
and that industry sales abuses are few. 

HCFA plans to test the effect of third-party enrollment contractors
in a future Medicare demonstration project.  Scheduled to begin
sometime in 1997,\9 this project will use a third-party contractor to
conduct marketing, education, counseling, and enrollment activities. 
HCFA's design--as of August 1996--will permit HMOs to provide
information to beneficiaries directly and even help beneficiaries
fill out enrollment forms.  The third-party contractor will provide
comparative information about the plans, counsel beneficiaries who
want to consult with a neutral party, and perform all enrollment
transactions.  The potential effect of this approach on enrollment is
not clear, and the demonstration's effects may not be fully evaluated
for years. 

If the Medicare program relies solely on enrollment brokers and
prohibits HMOs from marketing to individual beneficiaries, however,
growth of Medicare managed care might slow.  HMO representatives with
whom we discussed this issue were concerned that brokers would be
less knowledgeable about the advantages of specific plans and thus
not as effective as sales agents in selling managed care to Medicare
beneficiaries.  Recent experience in the Medicaid program suggests
that prohibiting direct marketing by HMOs could slow enrollment
growth.\10 Because of abuses, Florida and New York prohibited HMOs
from marketing to beneficiaries directly.  Both states experienced
significant declines in Medicaid HMO enrollment.  Florida reported
that, in a recent 3-month period since banning direct marketing,
enrollment levels fell by an average of 10,000 enrollees per month. 
New York temporarily suspended its ban on direct marketing to help
increase HMO enrollment but implemented other steps to prevent HMO
marketing abuses.  In fact, in many Medicaid programs in which
beneficiary participation in managed care is voluntary, states rely
on HMOs to inform beneficiaries about managed care and encourage them
to enroll. 


--------------------
\9 HCFA's Medicare Competitive Pricing Demonstration Project is
designed to test both the feasibility and effects of (1) setting
Medicare payments for HMOs using competitive market forces, (2)
providing beneficiaries with comparative information, and (3) using
third-party contractors to enroll beneficiaries in HMOs.  The
demonstration also includes plans for a coordinated enrollment
period.  Currently, the demonstration is behind schedule, and HCFA
has encountered difficulties selecting a site. 

\10 Medicaid:  States' Efforts to Educate and Enroll Beneficiaries in
Managed Care (GAO/HEHS-96-184, Sept.  17, 1996). 


      POTENTIAL COSTS IF
      DISSATISFIED COULD DISSUADE
      SOME BENEFICIARIES FROM
      ENROLLING IN MEDICARE HMOS
---------------------------------------------------------- Letter :3.5

Although a limited enrollment period could add some consumer
protections for beneficiaries, it could expose dissatisfied
beneficiaries to additional out-of-pocket costs.  Under the limited
enrollment period policy discussed here, beneficiaries dissatisfied
with their HMOs would have three choices:  (1) remain in the HMO, (2)
switch to traditional Medicare fee for service and pay the deductible
and coinsurance for submitted claims, or (3) switch to traditional
Medicare fee for service and purchase a Medigap policy if one was
available to them.  Beneficiaries dissatisfied with access to desired
treatments could remain in their HMO and purchase those services
privately.  However, going outside the HMO for treatment or changing
to fee for service would cost most beneficiaries more money than they
would have spent had they been able to enroll in another HMO.\11

Changing to traditional fee for service could be an expensive option
for many dissatisfied Medicare HMO members.  HMOs are cheaper than
fee for service for many Medicare beneficiaries because 65 percent of
HMOs do not charge a monthly premium (so-called "zero premium
HMOs").\12 In addition, HMOs frequently offer benefits, such as
outpatient prescription drugs, that are not provided by traditional
Medicare.  Beneficiaries in HMOs are responsible for copayments for
certain services but often fewer services than in a fee-for- service
arrangement.  Beneficiaries in fee for service who need services
covered under Medicare part B must fulfill a deductible and pay a
portion of additional expenses.\13

Dissatisfied HMO members who change to fee for service may want to
purchase supplemental health insurance, known as Medigap, to help
cover out-of-pocket costs.\14 However, Medigap policies can cost over
$1,000 per year--more than most beneficiaries would pay to an HMO. 
Furthermore, beneficiaries have no guarantee that a Medigap policy
will be available upon disenrolling from an HMO.  During the 6 months
after a person turns age 65 and enrolls in Medicare part B, federal
law guarantees beneficiaries the opportunity to purchase a Medigap
policy.  After that, Medigap insurers are permitted to refuse to sell
policies because of an applicant's health history or status.  We
recently reported that, although some insurers do exercise their
option to refuse coverage, all beneficiaries currently have at least
one Medigap policy available to them after the 6-month guarantee
period, regardless of their health history or status.\15
Nevertheless, no federal requirement exists to ensure that
beneficiaries will always have such alternatives. 


--------------------
\11 Although switching HMOs during a 12-month period is not typical
beneficiary behavior, neither is it uncommon.  Of the 161,792
beneficiaries who enrolled in HMOs at the start of 1995, 9,727
switched to another HMO or left and returned to their original HMO in
less than 1 year--a choice that would not be permitted under the
limited enrollment scenario discussed here. 

\12 Monthly premiums for the remaining HMOs are often lower than
premiums for Medigap policies.  As of September 1996, less than 6
percent of risk HMOs had monthly premiums that exceeded $60.  In
addition to any premium charged by the HMO, beneficiaries must
continue to pay their Medicare part B premium ($42.50 per month in
1996). 

\13 Beneficiaries' payments under Medicare part A (hospital
insurance) vary depending on factors such as their length of hospital
stay or whether they receive care in a skilled nursing facility.  For
Medicare part B (medical insurance), beneficiaries must pay a $100
annual deductible, after which they are responsible for 20 percent of
the Medicare-approved amount for most services.  For outpatient
hospital services, beneficiaries are responsible for 20 percent of
the charges, regardless of the Medicare-
approved amount.  Beneficiaries are responsible for 50 percent of the
Medicare-approved amount of outpatient mental health services. 

\14 The 10 standard Medigap policies cover Medicare coinsurance. 
Some policies also cover Medicare deductibles and benefits not
covered by Medicare such as prescription drugs. 

\15 Medigap Insurance:  Alternatives for Medicare Beneficiaries to
Avoid Medical Underwriting (GAO/HEHS-96-180, Sept.  10, 1996). 


      FEE FOR SERVICE COULD BE
      ONLY OPTION FOR MEDICARE
      "SNOWBIRDS"
---------------------------------------------------------- Letter :3.6

Beneficiaries who temporarily relocate for the winter, commonly known
as "snowbirds," might find joining a Medicare HMO impractical and
would probably choose the fee-for-service option instead.  HMOs are
required to provide emergency, but not routine, care to members
outside the HMO service area.  Furthermore, HMOs are required to
disenroll any member who leaves his or her HMO's service area for
more than 90 days.  Currently, snowbirds can disenroll from an HMO
and switch to fee for service or another HMO each time they relocate. 
If a limited enrollment period policy prohibited such plan switching,
snowbirds would be left with only one realistic option--enrolling in
Medicare's fee-for-service program.  Although data are not available
on the number of Medicare snowbirds, their existence is widely
recognized. 

HMOs might respond to a limited enrollment period policy by offering
flexible service arrangements not commonly available today, such as
reciprocal agreements and point-of-service options, partly to attract
snowbirds.  Reciprocal agreements among health plans--which permit
HMO members traveling outside their plan service area to receive
routine care and nonemergency services from another HMO--would make
temporary relocations less problematic for beneficiaries who wished
to enroll in managed care.  Several HMOs now offer reciprocity but
only within their own companies or affiliates.  For example, a member
of the Kaiser Foundation Health Plan in Los Angeles may receive
services from Kaiser HMOs in other parts of the country.  A
representative of the American Association of Retired Persons said
her organization is interested in encouraging the development of
reciprocal agreements among plans, although no such agreements
currently exist.  Similarly, if many HMOs offer the point-of-service
option--a hybrid of HMOs and fee-for-service plans--a Medicare policy
limiting plan switching would be less of a deterrent to snowbirds who
wished to enroll in HMOs.\16


--------------------
\16 With a point-of-service plan, beneficiaries have the option of
receiving services within their HMO's provider network or, for an
additional cost, receiving some services from nonparticipating
providers. 


      HMO ENROLLEES COULD FACE
      DELAYS OBTAINING PHYSICIAN
      APPOINTMENTS AT START OF
      HEALTH BENEFITS YEAR
---------------------------------------------------------- Letter :3.7

Most of the HMOs we contacted believe that a limited enrollment
period would cause beneficiaries to face delays in receiving health
care services at the beginning of each health benefit year.  HMO
representatives said a heavy demand for services would be caused by
new Medicare members' "trying out" their new physicians soon after
enrolling.  One HMO told us that a large percentage of that HMO's new
members see their primary care physician within 60 days of enrolling
to receive health care or renew a prescription.  In fact, some plans
strongly suggest that new members undergo initial health assessments
within 30 days of joining.  Although demand for provider services
also increases after the start of a commercial contract, the effect
of an influx of new Medicare members is greater because Medicare
beneficiaries tend to use physician services more frequently than
younger HMO members. 

Beneficiaries who would likely face delays in scheduling physician
office visits might be those who join HMOs that employ providers
directly ("staff model" HMOs) or have exclusive contracts with
providers ("captive group model" HMOs) or those who join HMOs with
relatively small provider networks.  Beneficiaries who join HMOs with
exclusive provider arrangements will, by definition, change providers
when changing plans.  New members in HMOs with small provider
networks are more likely to need to select a new provider than
beneficiaries joining plans with large networks.  However, for some
beneficiaries, joining an HMO or switching among plans will not
require switching physicians and an introductory visit because
physicians often contract with multiple HMOs. 

Obtaining appointments at the start of each health benefit year might
be difficult for beneficiaries in some HMOs because a limited
enrollment period policy would probably result in dramatic,
once-a-year membership spikes.  From December 1994 to December 1995,
24 plans enrolled more than 10,000 new members, including 1 that
enrolled close to 55,000 members.  (Table 1 shows the distribution of
new members among plans.) These membership increases, however, were
absorbed by the plans over 12 months, not during a single month, as
might occur under a limited enrollment period policy. 



                                Table 1
                
                  Medicare Risk HMO Membership Growth,
                     December 1994 to December 1995

                                                                Number
                                                                    of
HMO membership growth                                             HMOs
--------------------------------------------------------------  ------
10,000 new members or more                                          24
5,000-9,999 new members                                             22
1,000-4,999 new members                                             54
Less than 1,000 new members                                         22
Membership decline                                                  14
----------------------------------------------------------------------
Note:  Based on 136 risk HMOs that had members in 1994 and 1995. 

Source:HCFA Medicare Market Penetration Report File. 

The annual enrollment change resulting from a limited enrollment
period could be difficult for HMOs to predict accurately; any
unanticipated HMO enrollment growth could contribute to provider
access problems.  Representatives of one large HMO described what
happened when they grossly underestimated the response to their
Medicare product in a new market area.  Although the plan had
contracted with a large number of physicians, it underestimated the
need for primary care physicians and certain specialists.  Demands on
plan physicians' time and the level of beneficiary complaints were so
high that some physicians quit.  The plan contracted with new
physicians (a process that took about 6 months) and cut back its
marketing efforts to hold down additional enrollment, but 1-1/2 years
passed before the plan's provider network could comfortably meet
members' demand for services. 

A January start date for the Medicare benefits year, as specified in
the BBA, could cause longer delays in receiving health services than
if another time of year was selected.  January is already a
particularly busy month for providers because so many members of
employer-based health plans begin their benefits years on January 1. 
Furthermore, according to HMO representatives, demand for physician
office visits is already high in January because of winter
respiratory illnesses.  However, choosing a month other than January
could increase the number of employers that are inconvenienced, as
discussed in the next section. 


   A LIMITED ENROLLMENT PERIOD
   POLICY COULD DISCOURAGE SOME
   EMPLOYERS FROM OFFERING MANAGED
   CARE TO RETIREES
------------------------------------------------------------ Letter :4

Limiting Medicare's enrollment period would create varying degrees of
administrative problems for employers and could, as a result,
discourage some employers from offering managed care to their
retirees.  Our survey results indicated that in January 1996 about 21
percent of all beneficiaries in Medicare risk HMOs enrolled through
employer groups.  Moreover, between January 1995 and January 1996,
the number of Medicare beneficiaries in HMOs sponsored by employer
groups grew by 17.5 percent.  The number of Medicare beneficiaries
individually enrolled in HMOs grew even more--by 36.2 percent.  (See
fig.  3.)

   Figure 3:  Number of Medicare
   Beneficiaries Enrolled in Risk
   HMOs Responding to Our Survey,
   by Type of Enrollment

   (See figure in printed
   edition.)

Source:  GAO survey. 


      START DATES OF MANY
      EMPLOYERS' HEALTH BENEFITS
      YEARS WOULD NOT COINCIDE
      WITH MEDICARE'S DATES
---------------------------------------------------------- Letter :4.1

Almost all employer groups offering coverage through the risk HMOs we
surveyed limit the period during which members can enroll, but not
all these groups choose the same times of year to enroll members and
to begin benefits.  Under a limited enrollment policy, unless
exempted from complying with Medicare's specific enrollment period
and effective date, some proportion of employers would need to shift
their health benefits calendar.  The BBA proposed an October
enrollment period with Medicare beneficiaries' choices effective
January 1.  This timing would have coincided with the dates used by
62 percent of the employers offering managed care to retirees in
1995.  (See fig.  4.)

   Figure 4:  Distribution of
   Employee Groups' Enrollment
   Dates and Membership in These
   Groups

   (See figure in printed
   edition.)

Source:  GAO survey. 

If legislation mandates a specific health benefits open season for
all Medicare beneficiaries, it is unlikely that employers with
different benefit seasons would all respond in the same manner. 
Rather, these employers could take one of three courses:  (1) shift
all employees' and retirees' benefits seasons and run a single season
that would coincide with Medicare's season, (2) shift seasons for
Medicare retirees only and run one season for retirees and another
for active employees, or (3) choose not to offer the Medicare risk
program to retirees. 


      JANUARY START DATE WOULD
      INCONVENIENCE SOME EMPLOYERS
---------------------------------------------------------- Letter :4.2

Some employers could face problems shifting their benefits season to
coincide with Medicare's.  Employers and benefit consulting firms we
contacted discussed two major reasons why nearly 4 of 10 employers
have their group coverage begin in a month other than January. 
First, employers often select a benefits year that coincides with the
start of their fiscal year, which may not be January.  Second,
employers with seasonal businesses often choose slow business months
to conduct an enrollment process.  For example, representatives of a
major benefits consulting firm and several national retailers told us
that because the winter holiday season is the busiest and most
demanding time of year for retailers, these employers try to avoid
other activity at that time.  One of the health benefits consultants
we contacted said that his firm had tried unsuccessfully to get some
of its clients to begin their coverage in a month other than January
to ease the firm's administrative burden. 


      RUNNING SEPARATE BENEFITS
      SEASON FOR RETIREES MAY
      PRESENT PROBLEMS FOR SOME
      EMPLOYERS
---------------------------------------------------------- Letter :4.3

To comply with a mandated health benefits season for Medicare, some
employers might choose to run two seasons--one for retirees and one
for active workers.  One business group told us that some employers
already run two separate seasons because retirees tend to take more
time and ask more questions of health benefits personnel than do
active employees.  However, executives of one national health
benefits consulting firm also said that running two separate seasons
costs employers more money than running a single season. 

Executives of one large national retailer anticipated that running
two health seasons would create serious administrative problems.  The
retailer would have to (1) untangle its contracts with HMOs so that
coverage for Medicare-eligible retirees could be separated from
coverage for active employees, retirees, and retirees' dependents
under age 65; (2) renegotiate contracts with plans; and (3) revise
internal policies and communications.  Executives said untangling
contracts could take 2 to 3 years to complete.  They further noted
that if they ran two seasons, members of the same family could find
themselves with different health benefit years.  Because of all these
problems, the executives said they probably would not offer Medicare
risk plans if they had to change benefit years.  They further
predicted that other employers whose benefits seasons would not
coincide with Medicare's would do the same. 


      EMPLOYERS WOULD NEED TIME TO
      TRANSITION TO NEW HEALTH
      BENEFITS CYCLE
---------------------------------------------------------- Letter :4.4

Employers who were willing to switch their health benefits season
would probably need 9 months to 1 year of planning time to make the
transition, according to representatives of employers and benefit
consulting firms.  For example, one retailer we contacted had been
operating a single season for employees and retirees with a benefits
year beginning at the start of its fiscal year on February 1.  This
company recently shifted the start of its benefits year for its
active employees because the February health benefits year required a
November or December enrollment period, which interfered with holiday
business.  The retailer started actively planning 1 year before the
change.  It encountered some administrative difficulties but found
that making the change was relatively inexpensive. 

The California Public Employees' Retirement System (CalPERS) also
recently shifted its health benefits season for both employees and
retirees.  Before this change, benefits became effective on August 1;
now benefits are effective January 1.\17 CalPERS changed its season
to coordinate with preferred provider organizations and other state
benefit programs that operate on a calendar year.  CalPERS found the
process of shifting its health benefits cycle manageable and not very
costly but did need about 15 months to prepare for the change. 


--------------------
\17 To accomplish this shift, CalPERS established an interim benefit
"year" 17 months long. 


   LIMITING DISENROLLMENT MIGHT
   SAVE MEDICARE MONEY BUT CAUSE
   PROBLEMS FOR BENEFICIARIES AND
   HMOS
------------------------------------------------------------ Letter :5

If a new enrollment policy also limited HMO members' opportunities to
disenroll and change to fee for service, the Medicare program might
save some money; however, the policy could also result in reduced
beneficiary protections, increased beneficiary dissatisfaction, and
slower HMO growth. 


      LIMITING DISENROLLMENT MIGHT
      GENERATE SOME MEDICARE
      SAVINGS
---------------------------------------------------------- Letter :5.1

Limiting opportunities for beneficiaries to disenroll from HMOs
mid-year might generate some cost savings for Medicare.  These
savings would occur because payments to HMOs are based on our
assumption that HMO enrollees' health and medical requirements are
the same as those of the average beneficiary in fee for service. 
However, beneficiaries who leave managed care plans and switch to a
fee-for-service arrangement are not average--they tend to use more
services and incur higher costs than the average fee-for-service
beneficiary. 

Nonetheless, our analysis indicates that Medicare's maximum potential
savings from limiting disenrollment might be small, relative to
overall program expenditures, because few managed care enrollees
change to fee for service.  To quantify potential savings, we studied
the behavior of all 738,000 California Medicare beneficiaries who
were enrolled in a risk HMO at the start of 1994.  Of the
beneficiaries who did not change residences, only 15,772 switched
from managed care to fee for service during 1994.\18 Medicare paid
fee-for-service claims for 11,382 of these beneficiaries, amounting
to almost $73 million.  If these beneficiaries had not been allowed
to disenroll from their plans, the Medicare program would have paid
$42 million in capitated payments to HMOs to cover these same
beneficiaries.  Thus, the potential savings of limiting disenrollment
would have been, at most, $31 million in California during
1994--compared with total Medicare risk HMO expenditures in
California of $4.2 billion. 



                                         Table 2
                         
                             Medicare's Potential Savings Had
                         Disenrollment Opportunities Been Limited
                          for Beneficiaries in California, 1994

                                                             Medicare
                                  Actual Medicare    expenditures had
                                expenditures, HMO       beneficiaries   Potential savings
                                      and fee for       stayed in HMO   from limiting HMO
                                      service (in         entire year       disenrollment
                                        millions)       (in millions)       (in millions)
-----------------------------  ------------------  ------------------  ------------------
15,772 beneficiaries in HMOs                $72.6                41.6                31.0
 in January 1994 who switched
 to fee for service for at
 least 1 month that year\a
11,684 beneficiaries in HMOs                 48.4                26.1                22.3
 in January 1994 who switched
 to fee for service April 1
 or later for at least 1
 month\a
-----------------------------------------------------------------------------------------
\a Excludes beneficiaries who changed county of residence. 

Source:  GAO analysis of Medicare Claims Database and HCFA Group
Health Plan Master File. 

Potential savings, as a percentage of payments to HMOs, may be
slightly higher in states other than California.  Beneficiaries in
California have many HMOs from which to choose and can readily join a
competing HMO if dissatisfied with their own.  In other states,
however, beneficiaries have fewer choices, and the rate of changing
to fee for service among dissatisfied beneficiaries may be higher
than in California.  Since limiting the opportunity to change to fee
for service during the year produces cost savings, the potential
savings, as a percentage of payments to HMOs, may be higher in states
with few HMOs.  However, national Medicare savings would still likely
be small because California represents about 44 percent of all
Medicare risk-contract HMO expenditures. 

The less restrictive the disenrollment policy is--in other words, the
more opportunities beneficiaries have to change to fee for
service--the smaller the potential savings.  For example, if
beneficiaries were permitted to disenroll and switch to fee for
service during the first 90 days of membership, Medicare would
realize some savings but fewer than under a more restrictive
disenrollment policy.  Our analysis of 1994 California data indicates
that Medicare would have saved, at most, $22 million if beneficiaries
had been permitted to disenroll to fee for service only within the
first 90 days. 

These estimated savings probably represent the upper limit of what
Medicare could have saved in California in 1994.  Our estimates
assume that beneficiary behavior and enrollment patterns would not
change as the result of a limited disenrollment policy.  However, as
the following section discusses, beneficiary behavior will likely be
affected by such a policy.  (See app.  III for further information
regarding our analyses of potential Medicare savings.)


--------------------
\18 We did not include in our analysis the 25,918 beneficiaries who
moved out of their county, or service area, during 1994.  These
beneficiaries could have left their HMOs' service areas and would
presumably have been able to change coverage mid-year, even under a
limited enrollment and disenrollment policy. 


      LIMITING DISENROLLMENT WOULD
      REDUCE BENEFICIARY
      PROTECTIONS AND COULD
      INCREASE BENEFICIARY
      DISSATISFACTION
---------------------------------------------------------- Letter :5.2

According to HCFA officials and beneficiary advocates, limiting
beneficiaries' ability to disenroll from plans would remove a
valuable beneficiary protection.  Medicare's current policy allows
any beneficiary who is dissatisfied to disenroll and join a new plan
or change to fee for service at the end of each month.  Changing the
disenrollment policy could also weaken plans' incentive to maintain
the quality of the services and care they provide.  Finally, without
the ability to disenroll, HCFA and HMOs believe that beneficiaries
are likely to file more grievances and appeals. 

Although most beneficiaries do not change plans frequently,\19 some
HMOs have high member disenrollment rates, which can signal member
dissatisfaction.  We recently reported that one HMO in Miami and one
in Los Angeles had 1995 disenrollment rates of 37 percent and 42
percent, respectively.\20 One Miami HMO with high disenrollment rates
had a 7-year history of Medicare deficiencies, including those
involving beneficiary appeal rights and quality assurance.  Thus,
although most members appear to be satisfied with their HMO, problems
do exist, and the freedom to disenroll provides a course of action
for dissatisfied plan members. 

Some beneficiary advocates believe that to ensure continuity of care,
beneficiaries should be able to disenroll from an HMO if their
physician leaves the plan.  In fact, this may be a common reason for
switching HMOs or changing to fee for service.  A 1992 study reported
that 26 percent of beneficiaries who disenrolled from an HMO cited
their doctor's leaving the HMO as a reason for disenrolling.\21 One
large HMO told us that after terminating a contract with one of its
physician groups, nearly all 1,668 members assigned to those
physicians disenrolled from the HMO.  Representatives of the plan
believe that these members followed their physicians to a competing
HMO that also contracted with the physician group.  The Physician
Payment Review Commission recently recommended that if a limited
disenrollment policy is established, beneficiaries have the right to
disenroll before year-end or to purchase services on a special
point-of-service basis for the rest of the year if a plan makes a
major change in its network of providers during the year.\22 However,
the Commission acknowledged that defining the precise circumstances
for permitting disenrollment could be difficult. 

HMO representatives believe that beneficiaries' current ability to
disenroll at the end of any month is good for competition and, thus,
good for consumers.  The need to retain members who can disenroll
motivates plans to maintain quality, work for member satisfaction,
and improve benefits continuously throughout the year.  For example,
officials at one large HMO told us that it increased benefits three
times in 1995 to remain competitive.  The HMO increased its
pharmaceutical benefit, reduced beneficiary copayments for office
visits, and improved its dental coverage. 

Representatives of several HMOs told us that an enrollment policy
that includes a 90-day disenrollment option would be better for
beneficiaries than no disenrollment option at all but that the
current practice of permitting monthly disenrollment is far better
for industry competition and for beneficiaries.  Many beneficiaries
who disenroll from their risk HMO do so within the first 90 days. 
For example, of about 326,000 beneficiaries who joined a risk HMO
during the first 3 months of 1995, 14.4 percent disenrolled within 1
year or less, but a disproportionate amount--
5.6 percent--disenrolled in less than 90 days. 

On the other hand, representatives of one HMO speculated that if
beneficiaries were permitted just 90 days to disenroll, short-term
disenrollment rates would soar.  Beneficiaries who are less than
completely satisfied with their HMO might quickly disenroll, rather
than give their plan a chance to address their complaints. 

HCFA officials predict that without the option of disenrolling,
dissatisfaction among HMO members would manifest itself in other
ways, such as an increase in grievances to HMOs and appeals to
HCFA--a prediction that was echoed throughout our visits to HMOs. 
This prediction is supported by data we obtained from one HMO.  In
1995, over 90 percent of this plan's Medicare group membership was
"locked into" the HMO for the year.  Because of conditions set by the
beneficiaries' former employers, these members could change plans
only during annual enrollment periods.\23 Group members filed
grievances at a rate 100 times greater than that of individual
members who could disenroll monthly.  Group members filed 60 times
more appeals than individual members.  HMO representatives speculated
that individual members who were dissatisfied simply disenrolled,
rather than file grievances or appeals. 


--------------------
\19 Of the beneficiaries who enrolled in a risk HMO in January 1995,
about 82 percent were still in their same plan in May 1996.  Of the
18 percent of beneficiaries no longer in their original plan, some
had moved out of their HMO's service area or had died. 

\20 GAO/HEHS-97-23, Oct.  22, 1996. 

\21 Frank W.  Porell and others, Factors Associated With
Disenrollment From Medicare HMOs:  Findings From a Survey of
Disenrollees (Boston:  Health Policy Research Consortium of Brandeis
University, 1992). 

\22 Physician Payment Review Commission, 1996 Annual Report to
Congress (Washington, D.C.:  Physician Payment Review Commission,
1996). 

\23 These beneficiaries could have disenrolled at any time and joined
another HMO as individuals but would have lost the additional health
coverage benefits offered by their previous employer. 


      LIMITED DISENROLLMENT COULD
      FURTHER SLOW MANAGED CARE
      GROWTH
---------------------------------------------------------- Letter :5.3

HCFA officials and nearly all the HMOs we contacted shared a strong
belief that limiting disenrollment opportunities would deter some
beneficiaries from joining managed care, although none of the
representatives could quantify the extent to which this would occur. 
Managed care is a relatively new concept to some Medicare
beneficiaries, and a 1-year lock-in requirement could discourage
beneficiaries from trying managed care.  Some beneficiaries might not
join HMOs because, even if dissatisfied with the care they received
or denied a procedure they believed was critical, they would have
little recourse available.  Medicare has an appeals process in place,
but, of course, beneficiaries have no guarantee that the appeal will
be resolved in their favor.  Some beneficiaries might not enroll in a
plan if they knew they would not be able to follow their physicians,
should the physicians leave the plan mid-year. 


   HCFA WOULD FACE PEAK LOAD
   PROBLEMS AND ADDITIONAL
   RESPONSIBILITIES
------------------------------------------------------------ Letter :6

Implementation of a limited enrollment period could strain HCFA's
resources by creating a peak load and by increasing HCFA's
responsibilities.  HCFA's enrollment and disenrollment activities
would be concentrated in a short period of time, rather than spread
out during the year.  Also, HCFA would need to provide beneficiaries
access to a consumer hot line and comparative plan information, both
of which would likely be required under a limited enrollment period
policy. 


      HCFA WOULD NEED TO ADAPT TO
      PEAK WORKLOAD
---------------------------------------------------------- Letter :6.1

HCFA could face problems in completing tasks such as processing
enrollments.  Currently, HCFA processes about 100,000 transactions a
month, or 1.2 million transactions a year, which include enrollments,
disenrollments, and status changes.  Plans electronically submit
these data, which are processed by computers at HCFA--generally
within a few days.  However, problems could arise, such as incomplete
data or discrepancies in data, which could require follow-up work by
HCFA.  Some HCFA officials told us that the agency could manage the
peak workload associated with a limited enrollment period.  However,
representatives of HMOs, other organizations, and even some HCFA
officials said the agency sometimes had difficulty managing its
current workload and meeting deadlines; they were skeptical of HCFA's
ability to handle a peak workload with current resources. 

CalPERS\24 and FEHBP\25 both operate a single annual enrollment
period and face a peak load each year.  CalPERS hires temporary
workers and allows the permanent staff to work overtime hours.  FEHBP
contracts with a private firm to handle enrollment changes for
federal retirees.  (Each federal agency handles enrollment changes
for its current employees.) HMOs that experience a peak load from
their commercial business often hire temporary workers or shift
employees from other departments within the HMO. 

HCFA might need to change other activities to accommodate the timing
of a limited enrollment period.  For example, every year HCFA
announces risk HMO capitation payment rates in September.\26 This
allows HMOs time to decide whether they will renew their contract and
to adjust premiums and benefits before the new contract cycle begins
in January.  Depending on the timing of the enrollment period, the
announcement of the payment rates might need to occur earlier in the
year so that HMOs could set premiums and benefits before Medicare's
open season.  Sufficient time would also be needed for HCFA to
produce and publish comparison charts as well as to review HMOs'
marketing materials.  (See fig.  5.)

   Figure 5:  Timing of HCFA and
   HMO Activities

   (See figure in printed
   edition.)

Source:  GAO analysis. 


--------------------
\24 CalPERS manages health benefits for about 1 million public
employees and retirees.  It operates one open season, during which
employees and retirees make about 120,000 enrollment changes
annually.  See app.  II for additional information on CalPERS. 

\25 FEHBP manages health benefits for about 9 million employees,
retirees, and dependents.  See app.  I for additional information on
FEHBP. 

\26 The capitation payment is based on the adjusted average per
capita cost (AAPCC), an actuarial projection of what Medicare
expenses will be for a given category of Medicare beneficiaries in
traditional fee-for-service Medicare.  The rates change each calendar
year. 


      HCFA WOULD HAVE ADDITIONAL
      RESPONSIBILITIES UNDER
      LIMITED ENROLLMENT PERIOD
      POLICY
---------------------------------------------------------- Letter :6.2

Under a limited enrollment period policy, HCFA would likely be
responsible for additional tasks.  Some tasks would be new for HCFA;
for example, the BBA envisioned that the agency would prepare and
distribute comparative information.  Other tasks would represent
expansions of HCFA's current role--for example, operating an
information hot line for beneficiaries and resolving an increased
volume of beneficiary complaints.  The amount and extent of these
tasks would, of course, depend on the specifics of the limited
enrollment period policy enacted. 

HCFA has efforts under way to produce comparative health plan
information but would need to take additional steps to distribute
that information to beneficiaries.  Two of HCFA's regional offices
have developed charts that compare local HMOs' premiums and benefits,
but these charts--
although available upon request--are not widely distributed.\27 The
agency is working to make some HMO comparative information available
on the Internet but has no plans to distribute printed information
directly to beneficiaries.  Currently, HCFA intends to leave
information distribution to beneficiary advocates and federally
supported insurance counselors.\28

Although HCFA has an information hot line for Medicare beneficiaries
with questions about Medicare, the system would likely be inadequate
to handle the volume of calls generated under a limited enrollment
period policy.\29

Representatives of HMOs, beneficiary advocacy groups, and benefit
consulting firms cautioned us that older people need time to
understand their options.  Older people also seek considerable
information before deciding to join an HMO.  Some large national
brokers operate hot lines for their client companies.  These hot
lines, staffed by trained counselors who are familiar with Medicare
and the company's specific plan, answer questions posed by the
company's retirees.  Officials told us that these hot lines need to
be able to handle a large volume of calls.  For example, the hot line
for one company (with 57,000 retirees) received about 1,000 calls a
day from the retirees during the 1995 enrollment season--even though
retirees not changing plans did not have to re-enroll.\30 Some
retirees called repeatedly with questions about each step of the
application and enrollment process. 

HCFA plans to test the distribution of special handbooks and detailed
comparison charts as part of its Medicare Competitive Pricing
Demonstration Project.\31 These documents would contain information
on managed care plans and fee for service with Medigap that would
help beneficiaries make enrollment choices.  HCFA also intends to
make a telephone counseling center and educational seminars available
to beneficiaries with questions.  However, the demonstration project
has already been postponed once.  According to HCFA officials, it is
now scheduled to begin during 1997. 

In addition to preparing comparative information and operating a hot
line, HCFA would need both guidelines and procedures under which it
would allow beneficiaries to change plans outside the open season. 
With a limited enrollment period, beneficiaries would be expected to
change plans only during the designated open season.  However, as in
other programs with limited enrollment periods, exceptions would
likely be allowed.\32 The BBA specified several conditions under
which beneficiaries could change plans outside the enrollment period. 
Some conditions--for example, a beneficiary's moving out of a plan's
service area--would be easy for HCFA to evaluate and determine
whether a plan switch would be allowed.  However, other conditions
specified in the BBA would require HCFA to investigate the specific
case before making a determination.  For example, the BBA would have
allowed beneficiaries to disenroll if they could demonstrate that the
health plan had materially misrepresented the plan's provisions in
its marketing. 


--------------------
\27 HCFA's regional offices in San Francisco and Philadelphia have
developed charts comparing the benefits and premiums of local HMOs. 
The charts are distributed to news organizations and insurance
counselors primarily.  (See GAO/HEHS-97-23, Oct.  22, 1996, for
additional information.)

\28 These counselors, many of whom are volunteers, are available
through the federally supported, but state-managed, Information,
Counseling, and Assistance program.  Counselors can provide
beneficiaries with general information about Medicare, Medicaid,
managed care plans, and various types of health insurance to
supplement Medicare. 

\29 Currently, the hot line receives about 50,000 calls a month from
beneficiaries with questions on various issues--including managed
care. 

\30 Officials explained that this type of enrollment system,
so-called "passive enrollment," tends to reduce the number of
beneficiary hot line calls. 

\31 See footnote 9 for information on the Competitive Pricing
Demonstration Project. 

\32 For example, the 1995 criteria FEHBP used to allow enrollment
changes outside open season included 27 events, such as changes in
marital or family status, changes in employment status, relocation to
another part of the country or to another country, the member
becoming eligible for Medicare, the member's or a family member's
losing Medicaid coverage, and changes in the member's health plan. 


   CONCLUSIONS
------------------------------------------------------------ Letter :7

Encouraging enrollment in a managed care plan can help the
government's efforts to reduce high service utilization in the
Medicare program without unduly diminishing beneficiary access to
services.  To the extent that enrollment and disenrollment policy
revisions force health plans to retain and serve Medicare's more
costly beneficiaries, the government can battle the effects of the
high utilization tendency inherent in unmanaged fee-for-service
reimbursement.  However, these same policy revisions could produce
disincentives and obstacles to greater managed care enrollment--for
beneficiaries, health plans, employers, and HCFA--thereby undermining
the government's very effort to lower utilization. 

In fact, an annual limited enrollment period, along with restricted
disenrollment options, could have little impact on overall Medicare
spending.  Although such a policy would reinforce the concept of
managed care and reduce the opportunities for less healthy HMO
enrollees to change to Medicare fee for service, our analysis
suggests that the savings might be relatively small.  For example, if
enrollment and disenrollment had been limited for California
beneficiaries in 1994, Medicare savings would have been--at most--$20
million to $30 million.  In contrast, Medicare spent $4.2 billion on
payments to California HMOs during that year. 

Moreover, an enrollment policy change would likely have several
unintended consequences, including the loss of important beneficiary
protections and complications for many employers who offer managed
care to their retirees.  The result could well be substantially
slower growth in Medicare managed care and increased beneficiary
dissatisfaction.  The magnitude of these impacts would depend,
however, on the details of the adopted policy, beneficiary and
employer reaction to those details, and the effects of any other
policy changes made at the same time. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

We provided copies of this report to officials of HCFA's Office of
Managed Care.  HCFA agreed that the monthly disenrollment option is
an important consumer protection.  Our report indicates that changing
Medicare's current policy of allowing beneficiaries to switch among
HMOs or between an HMO and fee for service could have far-reaching
consequences.  We reported that this view is shared by beneficiary
advocates and HMO officials, who also believe that eliminating this
option would deter some beneficiaries from joining a managed care
plan. 

HCFA also stated that any analysis of beneficiary choice issues
should examine Medigap policy.  Our report notes that under current
law, beneficiaries have no guarantee that a Medigap policy will
always be available to them when they disenroll from an HMO.  As a
result, they may be reluctant to join an HMO.  HCFA commented that it
supports changes to the Medigap statute so that beneficiaries
dissatisfied with their managed care plan would be able to return to
fee for service and to the Medigap policy of their choice.  In a 1996
report, we made a similar recommendation.\33 HCFA's comments appear
in appendix IV. 


--------------------
\33 Medigap Insurance:  Alternatives for Medicare Beneficiaries to
Avoid Medical Underwriting (GAO/HEHS-96-180, Sept.  10, 1996). 


---------------------------------------------------------- Letter :8.1

As arranged with your office, unless you announce its contents
earlier, we plan no further distribution of this report until 5 days
after the date of this letter.  At that time, we will send copies to
the Secretary of Health and Human Services.  We will make copies
available to others on request. 

Please contact me at (202) 512-7114 if you or your staff have any
questions.  Major contributors to this report are listed in appendix
V. 

Sincerely yours,

William J.  Scanlon
Director, Health Financing and
 Systems Issues


THE FEDERAL EMPLOYEES' HEALTH
BENEFITS PROGRAM
=========================================================== Appendix I


   OVERVIEW
--------------------------------------------------------- Appendix I:1

To help us analyze the impact of a limited enrollment period, with
limited disenrollment, we looked at the Federal Employees' Health
Benefits Program (FEHBP).  We selected FEHBP because it is a large
employer-
sponsored health insurance program that conducts an annual enrollment
period (called an "open season") and, like Medicare, offers members a
choice of health plans. 

FEHBP is the largest employer-sponsored health insurance program in
the world.  The Office of Personnel Management (OPM) administers the
program, which went into effect on July 1, 1960.  FEHBP currently
provides voluntary health insurance coverage for about 9 million
people, including 2.3 million active employees, 1.8 million retirees,
and 5 million dependents.  In fiscal year 1995, FEHBP spent about
$17.7 billion to cover its members.  FEHBP outperforms Medicare--and
probably private plans--in controlling health care costs. 

The federal government and FEHBP members share program costs.  The
government contribution is readjusted annually.  For 1997, the
federal government's maximum annual contribution is $1,630 for
individuals and $3,510 for families.\34 The beneficiary's
contribution for individual coverage ranges from about $400 to $2,000
or more, and family coverage ranges from $800 to almost $5,000. 

In 1994, 59 percent of FEHBP retirees aged 65 or older also enrolled
in Medicare, although enrollment is not mandatory.\35 When a retiree
enrolls in Medicare, FEHBP serves as a supplemental insurance policy. 
FEHBP plans must waive deductibles, copayments, and coinsurance for
services covered by both programs.  Retirees pay the same premiums as
current employees. 

FEHBP offers a selection of several types of health plans, including
many managed care plans.  As in Medicare, the number of plans offered
to members varies by location.  Most of the fee-for-service plans (12
of 15) offer a preferred provider organization option.  Under FEHBP,
individual health plans establish their own relationships with
providers, process individual claims, develop benefits, and devise
marketing strategies. 

Since 1980, the number of HMOs has increased significantly.  By 1995,
about 29 percent of FEHBP members were enrolled in HMOs.\36 Like the
Medicare population, however, a much lower proportion of older
retirees were enrolled in HMOs; by 1996, only about 12 percent of
FEHBP members aged 65 and older and 10 percent of Medicare
beneficiaries were enrolled in HMOs.  Studies have shown that FEHBP
members who choose an HMO are generally younger and healthier than
members who select fee-for-service plans. 


--------------------
\34 The federal government cannot pay more than 75 percent of a
plan's total premium. 

\35 A 1983 amendment to the FEHBP legislation extended Medicare
coverage to FEHBP members aged 65 or older.  Before 1983, such
members were prohibited from participating in the Medicare program. 

\36 Enrollment by Medicare beneficiaries in HMOs has increased from 1
million in 1987 to about 4 million in 1996. 


      ROLE OF OPM AND EMPLOYING
      AGENCIES
------------------------------------------------------- Appendix I:1.1

OPM administers FEHBP, although each federal agency collects
information and premiums from employees.  OPM also interprets the
health insurance laws, writes regulations, and resolves disputed
claims.  OPM approves qualified plans for participation in the
program and negotiates with plans nationwide to determine benefits
and premiums for the following year.  OPM also publishes enrollment
and health plan information, including charts that compare benefits
and premiums.  OPM requires that the same premium be offered to
employees and retirees, regardless of their age, gender, or health
status.  It also requires that national plans offer the same premium
nationwide.  Local plans may offer local rates. 


      ENROLLMENT
------------------------------------------------------- Appendix I:1.2

FEHBP holds one annual open season, during which employees and
retirees may voluntarily enroll in a plan, change plans or options
within a plan, or change from individual to family coverage.  Most
changes by retirees occur during the first 2 weeks of the enrollment
season.  In 1996, open season occurred between November 11 and
December 9; changes made during open season became effective on
January 5, 1997--the first day of the next insurance year.  Each
year, about 5 to 10 percent of beneficiaries change plans. 


      ENROLLMENT FOR RETIREES
------------------------------------------------------- Appendix I:1.3

Most federal employees remain members of FEHBP when they retire; they
are familiar with how the open season works and with how to obtain
their health plan information.  However, retirees who choose to
disenroll from the program cannot return unless they had joined a
Medicare risk HMO.  They are required to sign a form to show they
understand that they cannot subsequently rejoin. 

Retirees can receive health plan information from health fairs; from
FEHBP directly; and from the National Computer System (NCS), an Iowa
City, Iowa, contractor that conducts retiree enrollment activities. 
Each year, some members of the Congress sponsor local health fairs
for federal employees and retirees.  Most of the people who attend
such fairs are retirees, in part because current employees attend
employer-sponsored health fairs.  Retirees can also call FEHBP
directly to request information or to discuss their options.  The
Retirement Information Office receives about 6,000 calls a month,
with about 25 percent of the calls focusing on health plans.  During
open season, the Health Benefits Branch receives about 500 calls a
day requesting information.  NCS, however, does not deal directly
with retirees.  Occasionally, it receives calls from retirees but
refers them to OPM. 

For the past 10 years, OPM has had a contract with NCS to handle
printing, distribution, processing, and brochure requests.  OPM
sought a contractor because it wanted to use technology, such as
scanning and other automated equipment, that OPM did not have.  Also,
NCS can hire temporary workers during busy times of the year; OPM
does not have the staff to handle retiree enrollment.  OPM believes
that the third-party contract with NCS is more efficient and less
expensive than if OPM was to do the work in house. 

About June of each year, OPM designs a health benefits application
form and sends it and a computer tape of the retiree rolls to NCS. 
NCS waits until approximately the first week in September, when the
OPM Policy and Information Office produces the final list of plans
and premiums.  Then, NCS prints a final list of available plans.  In
addition, it prints the comparative information with a rate sheet and
envelopes with addresses.  At the end of October, NCS mails to
retirees an E-Z application form, an instruction form with the rates,
and a return envelope.  Retirees who want to change plans return
their forms to NCS, which enters the change on its computer and sends
the information to OPM weekly during open season.\37 OPM notifies
plans of any changes. 

When retirees receive the information from NCS, they can request an
enrollment change or request additional information on specific
plans.  Unless they request information from NCS, they will only
receive it from their current plans.  Those who do not return their
forms automatically remain in the plan to which they belonged the
previous year. 


--------------------
\37 Currently, OPM also allows retirees to make enrollment changes by
telephone. 


      ROLE OF HMOS
------------------------------------------------------- Appendix I:1.4

HMOs supply plan information to FEHBP, which distributes it to
retirees through NCS.  HMOs can also market to retirees through
advertisements in newspapers, radio, and on television.  However,
they generally do not contact retirees directly unless a retiree is
already a member of the HMO.  In contrast, Medicare risk HMOs are
responsible for marketing to prospective members; HCFA does little to
provide plan information directly to beneficiaries.  In addition to
doing the same kind of mass media advertising as FEHBP HMOs, Medicare
risk HMOs are permitted to conduct one-on-one and group meetings. 
Medicare HMOs rely heavily on these techniques to attract new
members. 


THE CALIFORNIA PUBLIC EMPLOYEES'
RETIREMENT SYSTEM
========================================================== Appendix II


   OVERVIEW
-------------------------------------------------------- Appendix II:1

To help us understand the impact of a limited enrollment period, we
examined the California Public Employees' Retirement System
(CalPERS).  As with FEHBP, CalPERS is a large organization that
conducts an open season each year and offers members a choice of
health plans.  For about 35 years, CalPERS has offered health
insurance to employees of public agencies.  In 1995, CalPERS had
about 1 million members and paid $1.5 billion in health care
premiums. 

The organization has two divisions.  The Health Plan Administration
Division negotiates contracts and rates with the HMOs.  The Health
Benefit Services Division handles enrollments or changes in plans and
conducts educational activities for members.\38 Each year, the Health
Benefit Services Division processes about 120,000 enrollment
documents. 

CalPERS offers members a choice of 22 plans.  During the open season,
plans must accept enrollees regardless of health status, age, or
previous medical condition.  CalPERS encourages its members to join
an HMO by allowing members to choose from among 16 HMOs, including 9
Medicare risk HMOs.  Currently, about 76 percent of CalPERS members
are enrolled in HMOs.  For people who are eligible for Medicare, the
advantage of enrolling in an HMO through CalPERS is that CalPERS will
reimburse them for the Medicare part B premium.  If retirees were not
enrolled in a CalPERS health plan at the time they retired, they are
not eligible to enroll during their retirement.  Also, CalPERS offers
HMO benefits, such as prescription drugs, that are better than the
benefits people could obtain individually. 

To make comparisons easier for members, CalPERS requires HMOs to
offer similar coverage.  In addition, plans cannot charge more than
the standard premium, which is the same for anyone enrolling in the
specific plan.  The amount an employer contributes to a premium
varies among the public agencies participating in CalPERS. 


--------------------
\38 In 1995, about 4 percent of CalPERS members changed plans. 


   ENROLLMENT
-------------------------------------------------------- Appendix II:2

CalPERS has one annual open season.  During 1996, the dates were
changed from an open season beginning May 1 with an effective date of
August 1 to an open season beginning September 1 with an effective
date of January 1, 1997.  CalPERS changed its season to coordinate
its deductibles with its preferred provider organizations and with
other state benefits such as the vision and dental care programs. 
The preferred provider organizations with which CalPERS contracts and
the other state programs operate on a calendar year.  CalPERS
officials told us that they found the process of shifting the health
benefits cycle manageable and not very costly but that the
organization needed about 15 months to prepare for the change. 

Retirees who want to change plans visit the CalPERS office in person
or submit a written request.  Medicare beneficiaries must notify
CalPERS in writing of a change in enrollment.  CalPERS instructs
Medicare beneficiaries to mail their enrollment information directly
to the HMO of their choice during open season.  The plan sends the
new enrollment information to HCFA. 

CalPERS officials characterized the peak load associated with open
season as a time when the staff members are "basically busier." To
handle the peak load, the organization hires temporary workers and
allows its permanent staff to work overtime hours.\39

Educating members is an important task for CalPERS, especially
educating older people who fear signing over their Medicare cards to
an HMO.  CalPERS sponsors retirement seminars for active employees
who are within 5 years of retirement.  It also offers 4-hour
individual sessions for people who will retire soon.  During the open
season, CalPERS provides generic educational information to its
members.  For example, CalPERS publishes a booklet annually that
describes the features of each plan.  It also publishes a companion
booklet that contains comparisons of the quality and performance of
plans.  In 1995, CalPERS sent the books directly to all members.  In
past years, CalPERS held quarterly informational seminars for
retirees; however, the seminars were discontinued because of poor
attendance. 

CalPERS mails an exit survey to members who leave a plan to determine
why they left.  Last year, it mailed 15,227 surveys to members with
basic coverage and 1,535 to members with supplemental and managed
care plans.  In 1995, CalPERS also sent members a survey that
measured member satisfaction.  This survey was sent to a random
sample of members of various plans.  Findings from the exit survey
allow CalPERS staff to evaluate the medical care and services the
members receive as well as discuss areas of dissatisfaction with HMO
representatives during contract negotiations.  CalPERS officials
believe that the two surveys provide members with a balanced
perspective of member experience with their health plan. 

CalPERS, like FEHBP, restricts HMOs' ability to market directly to
members, although general marketing takes place statewide.  Plans are
not allowed to use gifts as incentives and are prohibited from
directly soliciting people who are not members of their plan. 

CalPERS officials have no data on the number of members who travel
seasonally ("snowbirds").  However, they estimate that between 8 and
10 percent of their Medicare enrollees might be snowbirds.  To assist
such members in receiving health services, CalPERS has encouraged
HMOs to develop reciprocal agreements with other plans. 


--------------------
\39 Officials could not provide us with data on the amount of
overtime worked. 


METHODOLOGY
========================================================= Appendix III

We assumed that a new Medicare enrollment policy might be similar,
but not necessarily identical, to the provisions contained in the
conference report that accompanied the Balanced Budget Act of 1995
(BBA), H.R.  2491.\40 Therefore, we developed and analyzed a limited
enrollment period policy modeled on the BBA.  Although other
alternatives are available to Medicare beneficiaries, we focused our
attention on enrollment in risk HMOs because they currently serve
most beneficiaries not in Medicare fee for service.  The hypothetical
policy we used to guide our analysis had three basic characteristics: 

  -- One enrollment period and one date when benefits became
     effective would be specified.  However, beneficiaries could
     elect coverage when they first became eligible for Medicare
     benefits regardless of the time of year this occurred. 

  -- The Secretary of HHS would be responsible for producing and
     distributing comparative plan information to beneficiaries as
     well as making a hot line available to them. 

  -- Beneficiaries could disenroll from an HMO during the year, but
     they would automatically be enrolled in fee for service. 
     Beneficiaries could switch to another HMO during the year only
     under limited circumstances, including moving out of their HMO's
     service area. 

We also analyzed the effect of limiting beneficiaries' disenrollment
options under two alternative scenarios: 

  -- no disenrollment would be allowed, except under specified
     circumstances, such as moving out of the health plan's service
     area; and

  -- disenrollment would be allowed for any reason during the first
     90 days after coverage was effective, but no disenrollment would
     be allowed after 90 days except under specified circumstances. 

To gather information on the likely effects of a limited enrollment
period and limited disenrollment opportunities, we interviewed
representatives of 10 Medicare risk HMOs, the American Association of
Health Plans, HCFA, national benefits consulting firms, selected
large employers who offer managed care options to retirees, Medicare
beneficiary advocacy organizations, FEHBP, and CalPERS.  In addition,
we surveyed HMOs with Medicare risk contracts regarding their
employer group business. 

We analyzed HMO disenrollment data and fee-for-service claims in
California to estimate potential Medicare savings from limiting
disenrollment. 


--------------------
\40 House Conference Report 104-350, pp.  1093-1102.  Because the act
was vetoed by the President, these provisions did not become law. 


   ESTIMATE OF POTENTIAL SAVINGS
   FROM LIMITING DISENROLLMENT
------------------------------------------------------- Appendix III:1

To estimate the potential Medicare savings that a policy limiting
disenrollment opportunities might generate, we compared 1994 Medicare
expenditures for California beneficiaries who changed from an HMO to
fee for service with the expenditures that Medicare would have
incurred had these beneficiaries been required to remain in their HMO
throughout the year.  We limited our analysis to California
beneficiaries to reduce the computational burden.  Nonetheless,
because Medicare HMO enrollment is concentrated in a relatively small
number of states--including California--our analysis covers about 36
percent of all Medicare beneficiaries enrolled in a risk HMO in 1994. 

We selected our sample population using 1994 data from HCFA's
Enrollment Database.  We identified 738,000 Medicare beneficiaries
who met the following criteria:  in January 1994 they belonged to a
risk HMO, they were eligible for Medicare parts A and B, and they
reported living in the same county 1 year later (in January 1995).\41

We then identified a subset of 15,772 beneficiaries who changed to
fee for service for 1 or more months during 1994. 

We computed the amount that Medicare would have paid for each of the
15,772 beneficiaries if they had remained in their HMO for the entire
year.  This amount varies by beneficiaries' county of residence and
demographic and other factors.\42 We then calculated the amount
Medicare actually spent on these beneficiaries in 1994--that is, the
capitation payments for the period they were enrolled in an HMO plus
their claims payments\43 for the period they were in fee for service. 

Finally, we estimated potential savings by subtracting the amount
Medicare would have paid if the 15,772 beneficiaries had remained in
HMOs from the amount Medicare actually paid during the year.  To
estimate potential savings of a policy that would allow beneficiaries
to return to fee for service during the first 90 days, we followed
the same steps, but included only those 11,684 beneficiaries who
changed to fee for service on April 1, 1994, or later.  (These
estimates are reported in table 2.)

Our estimates are probably upper bounds on potential savings in
California.  If a limited disenrollment policy discouraged some
beneficiaries from initially enrolling in an HMO, potential savings
could be lower.  Whether potential national savings can be
extrapolated using our estimates for California depends on whether
beneficiaries switch to fee for service at the same rate in other
states as they do in California.  Nonetheless, the behavior of
Californians would heavily influence estimates of national savings
because that state accounted for 44 percent of all payments to
Medicare risk HMOs in 1994. 


--------------------
\41 Because the BBA would have allowed beneficiaries who moved out of
their HMOs' service areas to return to fee for service, we excluded
25,918 beneficiaries who reported living in a county in 1995 that was
different from the county they reported in January 1994. 

\42 The demographic characteristics that affected HMO capitation
payments in 1994 were age, sex, institutional status, and Medicaid
status.  Capitation rates also depend on whether a beneficiary is
disabled or has end-stage renal disease. 

\43 The claims payments covered inpatient services, outpatient
services, physician/supplier services, care in a skilled nursing
facility, care in a hospice, home health care, and durable medical
equipment. 


   SURVEY OF MEDICARE RISK HMOS
------------------------------------------------------- Appendix III:2

To collect information on contracts between Medicare HMOs and
employer groups, we mailed a survey to all 118 HMOs that had risk
contracts in effect on January 1, 1995.  Eighty-three percent of the
HMOs responded to our survey and provided us with summary data on
retiree group contracts, including whether the contracts had a
limited enrollment period and a lock-in requirement. 




(See figure in printed edition.)Appendix IV
COMMENTS FROM THE HEALTH CARE
FINANCING ADMINISTRATION
========================================================= Appendix III


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

James C.  Cosgrove, Assistant Director, (202) 512-7029
Maryanne Keenan, Senior Social Science Analyst
Andrea Rozner, Senior Evaluator
Lori Weiss, Senior Evaluator



RELATED GAO PRODUCTS
=========================================================== Appendix 0

Medicare:  HCFA Should Release Data to Aid Consumers, Prompt Better
HMO Performance (GAO/HEHS-97-23, Oct.  22, 1996). 

Medicaid:  States' Efforts to Educate and Enroll Beneficiaries in
Managed Care (GAO/HEHS-96-184, Sept.  17, 1996). 

Medigap Insurance:  Alternatives for Medicare Beneficiaries to Avoid
Medical Underwriting (GAO/HEHS-96-180, Sept.  10, 1996). 

Medicare HMOs:  Rapid Enrollment Growth Concentrated in Selected
States (GAO/HEHS-96-75, Jan.  18, 1996). 

Medicare:  Increased HMO Oversight Could Improve Quality and Access
to Care (GAO/HEHS-95-155, Aug.  3, 1995). 


*** End of document. ***