Health Care Quality: Implications of Purchasers' Experiences for HCFA
(Chapter Report, 06/18/98, GAO/HEHS-98-69).

Pursuant to a congressional request, GAO determined how large purchasers
use quality-related data to seek or promote better quality of care and
lessons that can be learned from their experiences for the Health Care
Financing Administration (HCFA).

GAO noted that: (1) after collecting and making use of quality-related
data, the purchasers GAO studied reported that in addition to cost
savings, they saw improvements in access to care and health plan
services, as well as in employee satisfaction with health plan
performance; (2) they realized such improvement by identifying
opportunities to use quality-related data in selecting health plans,
monitoring health plan performance, developing quality improvement
initiatives with plans and taking other actions, and providing
information on health plans to their employees; (3) while HCFA is a
unique purchaser of managed care--by virtue of the size of the Medicare
program and the freedom of choice provided to beneficiaries--a number of
private purchasers' quality of care strategies could be relevant to
HCFA's administration of the Medicare program; and (4) major lessons
from large purchasers' experiences relate to the importance of: (a)
educating employees as to the meaning of quality-related measures when
providing comparative information on health plan quality; (b) using
collaborative- and compliance-oriented approaches to achieve
improvements in plan performance; and (c) continually looking for
additional opportunities to make use of quality-related data, such as
developing standards and benchmarks for plan performance.

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

 REPORTNUM:  HEHS-98-69
     TITLE:  Health Care Quality: Implications of Purchasers' 
             Experiences for HCFA
      DATE:  06/18/98
   SUBJECT:  Health maintenance organizations
             Quality assurance
             Health care programs
             Health care cost control
             Data collection
             Employee medical benefits
             Health insurance
IDENTIFIER:  Medicare Program
             NCQA Health Plan Employer Data and Information Set
             California Public Employees Retirement System
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Chairman and Ranking Minority, Special Committee on
Aging, U.S.  Senate

June 1998

HEALTH CARE QUALITY - IMPLICATIONS
OF PURCHASERS' EXPERIENCES FOR
HCFA

GAO/HEHS-98-69

Purchasers and Quality-Related Data

(108291)


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

  CBO - Congressional Budget Office
  FACCT - Foundation for Accountability
  HCFA - Health Care Financing Administration
  HEDIS - Health Plan Employer Data and Information Set
  HMO - health maintenance organization
  JCAHO - Joint Commission on Accreditation of Healthcare
     Organizations
  NCQA - National Committee for Quality Assurance
  POS - point-of-service
  PPO - preferred provider organization
  WBGH - Washington Business Group on Health
  HIV/AIDS - human immunodeficiency virus acquired immunodeficiency
     disease syndrome

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


B-279631

June 18, 1998

The Honorable Charles Grassley
Chairman
The Honorable John Breaux
Ranking Minority Member
Special Committee on Aging
United States Senate

As you requested, we have determined how large purchasers use
quality-related data to seek or promote better quality of care and
lessons that can be learned from their experiences for the Health
Care Financing Administration. 

As arranged with your office, unless you publicly announce the
contents of this report earlier, we plan no further distribution
until 30 days after its issue date.  At that time, we will send
copies of this report to the Administrator of the Health Care
Financing Administration, appropriate congressional committees, and
other interested parties.  If you have any questions about this
report, please call me at (202) 512-6543.  Other major contributors
are listed in appendix II. 

Bernice Steinhardt
Director, Health Services and
 Quality Issues


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

To help control costs associated with health care, public and private
sectors have moved toward managed care.  Medicare, the nation's
largest purchaser of health care, contracts with nearly 400 managed
care plans that serve over 5.5 million elderly beneficiaries--about
14 percent of the total Medicare population.  Over the past several
years, the number of Medicare beneficiaries enrolled in managed care
plans and the number of plans contracting with the Medicare program
have increased dramatically.  With more and more beneficiaries
enrolling in managed care, Medicare faces many of the same concerns
that other purchasers face--particularly those that relate to quality
of care.  Some large corporate purchasers have begun to examine
"value-based purchasing"--how best to achieve value as a balance
between cost and quality.  As they purchase care, some employers ask
for quality-related data from health plans and find opportunities to
take action on the basis of such data as they monitor plan
performance. 

To better understand how quality of care might be ensured for
Medicare beneficiaries, the Chairman and the Ranking Minority Member
of the Senate Special Committee on Aging asked GAO to determine (1)
how large purchasers use quality-related data to seek or promote
better quality of care and (2) lessons that can be learned from their
experiences for the Health Care Financing Administration (HCFA) as it
administers the nation's Medicare program.  In conducting this work,
GAO performed case studies of two national purchasers--one private
and one public--and two regional purchasers.  (See app.  I for a
brief description of each purchaser.)


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

Title IV of the Balanced Budget Act of 1997 (P.L.  105-33) encourages
increased enrollment of Medicare beneficiaries in managed care.  HCFA
will also be required to provide quality-related and other
comparative information to beneficiaries to help them make informed
health plan choices.  In addition, the act requires plans to take
action to improve the quality of care received by Medicare
beneficiaries and to assess the effects of such actions. 

Value-based purchasing, a concept being examined by large corporate
purchasers to improve quality of care, is based on the idea that
inexpensive health care has little value if employees get sick more
often, stay sick longer, or suffer more disabilities due to poor
quality care.  To evaluate the quality of care received from health
plans, doctors, and other professionals, the following must be
determined:  (1) the appropriateness of the care provided, (2) the
technical excellence of the providers' knowledge and their delivery
of care, (3) patient accessibility to care, and (4) patient
satisfaction with the care received.  These purchasers are requiring
plans to provide quality-related data and are taking action on the
basis of such data. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

After collecting and making use of quality-related data, the
purchasers GAO studied reported that in addition to cost savings,
they saw improvements in access to care and health plan services, as
well as in employee satisfaction with health plan performance.  They
realized such improvements by identifying opportunities to use
quality-related data in selecting health plans, monitoring health
plan performance, developing quality improvement initiatives with
plans and taking other actions, and providing information on health
plans to their employees. 

While HCFA is a unique purchaser of managed care--by virtue of the
size of the Medicare program and the freedom of choice provided to
beneficiaries--a number of purchasers' quality of care strategies
could be relevant to HCFA's administration of the Medicare program. 
Major lessons from large purchasers' experiences relate to the
importance of (1) educating employees as to the meaning of
quality-related measures when providing comparative information on
health plan quality; (2) using collaborative- and compliance-oriented
approaches to achieve improvements in plan performance; and (3)
continually looking for additional opportunities to make use of
quality-related data, such as developing standards and benchmarks for
plan performance. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      PURCHASERS USE
      QUALITY-RELATED DATA TO
      ACHIEVE EARLY RESULTS
-------------------------------------------------------- Chapter 0:4.1

Purchasers can use quality-related data to inform their assessment of
whether or not to contract with health plans; purchasers can also use
such data to monitor plan performance and provide information to
their employees.  The purchasers GAO studied used a variety of
quality-related data from different sources, including (1) evidence
of accreditation from the National Committee on Quality Assurance and
the Joint Commission on Accreditation of Healthcare Organizations;
(2) reports on how well plans address selected measures from the
Health Plan Employer Data and Information Set (HEDIS), a standardized
set of performance measures; and (3) surveys of employee satisfaction
with the quality of care received from health plans.  Purchasers have
drawn on the resources of groups of employers banding together, known
as business coalitions, and participated in a cooperative HEDIS
reporting initiative to collect, analyze, and report audited data on
HEDIS measures. 

By reviewing data from multiple sources--especially those data
related to employee satisfaction--purchasers have been able to
identify problems with and improve health plan performance as well as
realize cost savings.  For example, one purchaser worked closely with
plans to achieve improvements in customer service and referrals to
specialists.  The purchaser identified problems in these areas using
a satisfaction survey administered to its employees, complaints data,
and feedback from a committee established to improve communications
between the purchaser and its employees.  The same purchaser used
HEDIS data from diabetes-related measures to encourage a plan to
develop an educational project targeted to the purchaser's employees
with diabetes. 

In addition to collaborating with plans, purchasers have used
financial incentives to reward or penalize plans for their
performance.  In one case, a purchaser noted that financial penalties
motivated a plan to institute changes that improved employee
satisfaction with plan providers.  Another purchaser cited cost
reductions as a result of using HEDIS and other data in rate
negotiations, with a 4-percent decrease in premiums achieved at 21
health plans targeted as having below-average performance on selected
HEDIS measures. 

Two of the purchasers that we reviewed also used quality-related data
in distributing health plan "report cards" on the characteristics and
performance of plans, which employees can use to compare plans and
make informed choices when selecting a health plan.  In addition to
providing the report cards, the two purchasers said they were careful
to explain the meaning of the quality-related measures.  After
distributing this information, one purchaser saw a relatively modest
shift by employees into a plan with a higher quality ranking. 
However, when the purchaser subsequently froze enrollment in plans
with lower quality rankings, there was a more significant shift into
the plan with the higher quality ranking.  The other purchaser found
that 66 percent of those responding to an employee survey viewed the
purchaser's report card as very or somewhat important in assisting
them in selecting a plan.  This same purchaser used feedback from
employees to make revisions to a subsequent report card. 

Each of the four purchasers plans to increase its use of
quality-related data in a variety of ways--some of which are already
in use by other purchasers.  Such use includes identifying poorly
performing plans on the basis of quality-related data and declining
to contract with them and linking plan performance to financial
incentives for employees to encourage them to choose plans with
higher quality rankings.  In addition, purchasers may use
quality-related data to contract with--and thereby monitor--fewer
plans.  Some of the purchasers have implemented standards and
benchmarks for health plan performance--which, in some cases, are
based on multiple types of quality-related data, such as HEDIS and
satisfaction data--as part of an effort to provide comprehensive
feedback to health plans on how their performance compares with
earlier years and how they compare against other plans.  Other
purchasers plan to develop such standards and benchmarks. 


      THE BALANCED BUDGET ACT
      MOVES HCFA IN THE DIRECTION
      OF OTHER PURCHASERS
-------------------------------------------------------- Chapter 0:4.2

The Medicare program stands apart from other purchasers by virtue of
its size and the freedom of choice provided to beneficiaries.  Where
the largest private purchaser that we visited serves 1 million
people, HCFA serves over 5.5 million beneficiaries through Medicare
health maintenance organizations (HMO).  In making their health care
decisions, Medicare beneficiaries have enjoyed more choice than much
of the employed population under 65.  They can choose fee-for-service
or managed care, select any of the Medicare-approved HMOs in their
area, and switch plans monthly. 

Legislation has shaped and continues to shape how quality of care is
monitored under the Medicare program through monitoring programs and
reviews by peer review organizations--also known as quality
improvement organizations.  For example, any eligible organization
that agrees to meet minimum standards may participate in the Medicare
program.  In contrast, employers can decide not to contract with
plans.  They have greater ability to command the attention of health
plans because of their ability to exclude them from contracts and
because employers can share quality-related data with employees that
could spur plans to improve. 

While the Medicare program will retain several of its distinguishing
characteristics--most notably, its status as the nation's largest
purchaser of health care--title IV of the Balanced Budget Act of 1997
begins to move HCFA in the direction already taken by the four
purchasers GAO reviewed.  While HCFA had authority and planned to
provide comparative information about Medicare plans to beneficiaries
prior to the passage of the act, it now has specific time frames for
doing so within the context of an annual open enrollment season.  In
addition, the act requires that Medicare-contracted health plans take
actions to improve quality as part of their internal quality
assurance programs.  To address this requirement, HCFA is considering
using standardized measures to determine whether plans are achieving
results from their quality assurance programs. 

With these changes, the experiences of the purchasers GAO reviewed
become more relevant to HCFA.  For example, the two purchasers that
disseminated quality-related data in report cards also explained and
interpreted the data to their employees, and they refined the
information that they provide employees on health plans.  All four
have used quality-related data to provide feedback on performance to
the plans as well as reevaluate how they provide plans with such
feedback.  In addition, the purchasers have used quality-related data
to improve performance through collaborative- and compliance-oriented
approaches.  In order to implement certain purchaser practices in
using quality-related data, such as the use of quality-related data
to select and negotiate rates with health plans, HCFA would need new
legislative authority. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

GAO is making no recommendations in this report. 


   COMMENTS FROM HCFA AND
   PURCHASERS
---------------------------------------------------------- Chapter 0:6

HCFA and the four purchasers visited by GAO commented on a draft of
this report.  They agreed with our presentation of the information
and our observations.  They also provided technical suggestions,
which we incorporated where appropriate. 


INTRODUCTION
============================================================ Chapter 1

In the public and private sectors alike, concerns about quality of
health care are intensifying as purchasers of health insurance shift
from traditional indemnity plans to managed care.  With plans'
increased focus on controlling the skyrocketing costs of health care
benefits, there are concerns about the value of the health benefits
purchased.  As a result, several large private purchasers have begun
to examine "value-based purchasing." Key to value-based purchasing is
the measurement of health plan quality using different types of
quality-related data to hold plans accountable and encourage
improvements. 

Lessons learned from the experience of large purchasers may be
applicable to the Health Care Financing Administration (HCFA), the
nation's single largest payer for health care.  HCFA administers the
Medicare program, which provides care for about 38 million
beneficiaries, over 5.5 million of whom are currently in health
maintenance organizations (HMO).  It purchases health care coverage
for almost all of the nation's elderly population and more than 4
million disabled beneficiaries.  Like purchasers in the private
sector, the federal government has looked to managed care as a way to
help contain costs associated with providing health care to Medicare
beneficiaries.  At the same time, the agency wants to ensure that the
beneficiaries currently enrolled in health plans and those who enroll
in the future are receiving high-quality care.  With the passage of
the Balanced Budget Act of 1997--a major piece of legislation
affecting the Medicare program--HCFA will have more plans and more
types of plans to monitor for the quality of care provided to
beneficiaries. 


   PURCHASERS SHIFT TO MANAGED
   CARE
---------------------------------------------------------- Chapter 1:1

In an effort to curb the double-digit inflation in health care costs
of the 1980s, large purchasers increasingly turned to managed care. 
The rise in managed care enrollment has been swift.  From 1987 to
1996, enrollment in managed care provided through private employers
nearly tripled.  According to a 1997 survey of health benefits
offered by firms with 200 or more workers,\1 only 19 percent of
employees are still enrolled in indemnity programs, which allow a
free choice of providers and reimburse physicians and hospitals with
limited or no review of the appropriateness of services rendered.  In
addition, traditional indemnity coverage uses a fee-for-service
payment mechanism to reimburse providers.  The remainder of employees
with health insurance receive care through a variety of health plans. 
These can include (1) HMOs, (2) preferred provider organizations
(PPO), and (3) point-of-service (POS) plans.\2

HCFA has also seen a rapid increase in managed care enrollment in
Medicare.  However, unlike the private sector, the vast majority of
Medicare beneficiaries still receive care through fee-for-service
arrangements.  In the early 1970s, the Congress encouraged commercial
and Medicare use of HMOs by authorizing federal standards and
oversight to ensure reasonable care and service.  Between 1994 and
1997, enrollment in Medicare HMOs increased by 75 percent.  There has
also been a dramatic increase in the number of plans Medicare
contracts with.  Currently, HCFA contracts with close to 400 health
plans to provide health care to over 5.5 million beneficiaries, about
14 percent of the total Medicare population. 

With the passage of the Balanced Budget Act, even greater growth in
Medicare beneficiary enrollment in managed care can be expected.  The
act permits contracts between HCFA and a variety of different managed
care entities, including PPO and POS plans, which are similar to HMOs
but are directly controlled by groups of providers.  The
Congressional Budget Office (CBO) projects that as a result of the
passage of the act, all types of managed care organizations will
account for 25 percent of Medicare enrollees in 2002, 38 percent in
2008, and about 50 percent by 2030. 


--------------------
\1 KPMG Peat Marwick LLP, Health Benefits in 1997 (June 1997).  KPMG
conducts annual surveys of employer-sponsored health benefits in
firms with 200 or more workers. 

\2 HMOs require patients to use a limited number of affiliated
physicians who may be salaried, paid on a per-capita (capitated)
basis, or reimbursed for each service.  Typically, a patient's care,
especially referrals to specialists and hospitalization, is
coordinated by a primary care physician--often called a "gatekeeper."
PPOs provide care to enrollees through a network of providers that
are normally reimbursed at a discounted rate, generally with higher
out-of-pocket costs to enrollees who choose to go to providers
outside the network.  Finally, POS plans generally resemble HMOs but,
like PPOs, allow enrollees to see nonaffiliated physicians if they
are willing to incur higher out-of-pocket costs.  Fee-for-service
payment is also used in PPOs and to some extent in POS and HMO plans. 


   NEED TO MONITOR QUALITY OF CARE
---------------------------------------------------------- Chapter 1:2

With increased use of managed care, public and private purchasers
must consider strategies to monitor plans and ensure the quality of
the care they provide.  The Institute of Medicine has formally
defined quality of care as "the degree to which health services for
individuals and populations increase the likelihood of desired health
outcomes and are consistent with current professional knowledge."\3
In evaluating plans, one or more of the following dimensions of
quality can be measured: 

  -- Appropriateness:  Are providers giving patients the care they
     need? 

  -- Technical excellence:  How well are providers using medical
     science and knowledge to deliver care to patients? 

  -- Accessibility:  Are patients able to obtain care when needed and
     within reasonable proximity to where they live or work? 

  -- Acceptability:  Are patients satisfied with the care they
     receive? 

Since the concept of quality is multidimensional, experts describe
the importance of using different types of measures to evaluate care. 
For example, the Foundation for Accountability (FACCT)--a forum for
consumers and purchasers, including HCFA--argues for the importance
of balancing the use of quality measures to reflect (1) the results
of care, (2) whether patients are satisfied with the care received,
and (3) whether the appropriate processes have been followed. 

Performance indicators are used to measure the various attributes of
quality.  For example, for clinical attributes, they can measure
appropriateness and technical excellence--that is, providers' actions
and the outcomes of those actions.  Process-related indicators refer
to clinical interventions, such as the diagnostic tests performed by
a physician when examining a patient.  In contrast, outcome
indicators measure the results of providers' activities, such as
mortality and morbidity.  Outcome measures are critical to evaluating
the quality of care, but experts recognize that these measures are
not fully developed. 

A number of questions have been raised about the reliability and
validity of certain measures and the data sources for performance
indicators.  For example, data from computerized administrative
databases maintained by managed care plans and from individual
patient medical records kept in providers' offices may be inaccurate,
incomplete, or misleading.  This is because most administrative
databases were designed for financial--not clinical--purposes.  In
addition, providers may enter incorrect information in medical
records or not document certain interventions.  In an earlier report,
we expressed concerns about the reliability of satisfaction data,
since most people lack the knowledge needed to adequately evaluate
the appropriateness of the care that they receive or do not
receive.\4 We also noted in the report that plan-reported data on
access-related measures, such as what constitutes a sufficient
provider network, do not necessarily ensure that access to care is
received.  Such data must be checked by independent and systematic
monitoring efforts that go beyond plan-reported, paper-based
indications of compliance. 


--------------------
\3 Institute of Medicine, Medicare:  A Strategy for Quality
Assurance, Kathleen Lohr, ed.  (Washington, D.C.:  National Academy
Press, 1990). 

\4 Medicaid Managed Care:  Challenge of Holding Plans Accountable
Requires Greater State Effort (GAO/HEHS-97-96, May 16, 1997). 


   SOME PRIVATE CORPORATIONS MOVE
   TOWARD VALUE-BASED PURCHASING
---------------------------------------------------------- Chapter 1:3

Despite problems in measurement, some large companies--concerned with
absenteeism and reduced productivity from illness--have begun to
apply value-based purchasing concepts when purchasing health plan
services.  For example, these companies have considered information
about quality to assess, rank, and select health plans and to monitor
ongoing plan performance against standards and negotiate rates based
on these standards.  In addition, these companies are providing
information on plan performance to employees to help inform their
selection of health plans.  Large purchasers have spearheaded several
initiatives as they search for credible tools to help them identify
and demonstrate to others the "value" resulting from premiums paid to
managed care plans.  For purchasers, standardized measures can help
them to set desirable goals or "benchmarks" for health plans in
different areas of interest or concern to the purchaser, provide
feedback to plans on the results of such performance, and monitor the
progress of plans against these goals. 

In the early 1990s, a committee of health plan representatives and
corporate purchasers began to work on a set of standardized
performance measures, which were later revised by the National
Committee for Quality Assurance (NCQA)--a nonprofit institution that
reviews and accredits health plans.\5 The result of these efforts,
the Health Plan Employer Data and Information Set (HEDIS), is now in
its third generation and currently covers the following categories: 
effectiveness of care, access and availability of care, satisfaction
with the experience of care, informed health care choices,
descriptive information on health plans, the cost of care, health
plan stability, and the use of services.\6

Another major effort by purchasers, with participation by HCFA and
other government agencies, was the creation of FACCT to develop
standardized outcome measures.  In 1996 and 1997, FACCT endorsed
comprehensive measurement sets for asthma, diabetes, breast cancer,
major depression, as well as other areas; some of these indicators
focus on outcomes.  Now FACCT is coordinating efforts with NCQA and
others to create comprehensive measures for children's health,
HIV/AIDS, end-of-life care, coronary artery disease, and alcohol
misuse.  FACCT has also developed a "consumer information framework"
for purchasers, which emphasizes the importance of a consistent and
understandable framework for presenting quality-related information
to consumers.  One example of this information is the ability of
health care organizations to maximize functioning and quality of life
when a consumer faces chronic, incurable illnesses, such as diabetes
and asthma. 

Despite the involvement of some major purchasers in the development
of quality-related measures, surveys conducted by the Watson Wyatt
consulting firm with the Washington Business Group on Health (WBGH)
in 1996 and 1997 concluded that cost still prevails as the principal
concern when most employers evaluate a managed care plan.\7 The
surveyed employers noted, however, that they are beginning to look
more closely at issues such as plan coverage and access in judging
health plan value.  And a significant number of employers are
requiring plans to report HEDIS data, with some making it a
prerequisite for health plans that wish to contract with them.  They
also view accreditation as providing assurance that a health plan is
attempting to manage the quality of care.\8

While employers are beginning to make increased use of
quality-related data in screening plans with which to contract, they
may not necessarily be using it throughout the purchasing and
monitoring process to the extent desired by proponents of value-based
purchasing.  A recent mapping of activities by individual employers
and business coalitions concluded that only a limited number are
actually implementing the principles of value-based purchasing.\9


--------------------
\5 NCQA was founded in 1979 by two trade associations that represent
the managed care industry.  It became independent in 1990 and now
represents the interests of purchasers and consumers as well as
health care organizations. 

\6 As a measurement set, HEDIS has evolved over time.  Early HEDIS
indicators addressed quality, access and patient satisfaction,
membership and utilization, finance, and HMO management.  The
indicators addressing quality issues generally focused on providers'
actions rather than the outcomes of those actions.  For example, the
indicators measured the rate at which women received mammograms but
not the 5-year survival rate of women diagnosed with breast cancer. 
NCQA has made subsequent revisions to HEDIS.  The latest version
includes a standardized patient satisfaction survey and more
indicators bearing on high-prevalence diseases.  Quality measures are
still predominantly process-oriented. 

\7 In 1996, Watson Wyatt and the Washington Business Group on Health
conducted the first employer survey of 384 U.S.  employers on value
purchasing of health benefits.  A follow-up survey of 325 U.S. 
employers was conducted in spring 1997.  See Watson Wyatt, Reality
Check:  Is Cost Everything?  (Bethesda, Md.:  1996) and Getting What
You Pay For:  Purchasing Value in Health Care (Bethesda, Md.:  1997). 

\8 Accreditation is a formal designation granted by a third party. 
NCQA and the Joint Commission on Accreditation of Healthcare
Organizations (JCAHO) both accredit managed care plans. 
Accreditation has traditionally involved evaluating the extent to
which health plans meet standards that specify the resources and
organizational arrangements necessary to attain quality.  Both NCQA
and JCAHO now look at a health plan's efforts to continuously improve
the quality of care and service it delivers. 

\9 Jack Meyer and others, Theory and Reality of Value-Based
Purchasing:  Lessons From the Pioneers (Rockville, Md.:  Agency for
Health Care Policy and Research, Nov.  1997). 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:4

The Chairman and the Ranking Minority Member of the Senate Special
Committee on Aging asked us to study how large corporate purchasers
use quality-related information collected from health plans and the
applicability of purchasers' experiences to HCFA.  Specifically, we
agreed to describe (1) how large purchasers use quality-related data
to seek or promote better quality of care and (2) lessons that can be
learned from their experiences for HCFA in administering the Medicare
program. 

In conducting our review, we analyzed and synthesized relevant
literature about managed care and discussed value-based purchasing
and quality measurement with employers and with HCFA officials.  We
then conducted detailed case studies with four large purchasers of
managed care for employees.  During site visits with these
purchasers, we discussed how they incorporated quality-related data
into their purchasing and monitoring decisions and the results they
believe are attributable to their efforts.  We also reviewed
available data on results achieved through these efforts. 

For the purposes of the case study analysis, we defined "results" in
terms of improved health plan performance on dimensions measured;
increased health plan accountability to the purchaser or enrollee;
and actions taken by purchasers, health plans, providers, or
consumers in response to quality-related data.  As such, we defined
results not in terms of outcomes in the sense of clinical quality but
rather those that indicated improvement in the performance of health
plans in the dimensions measured by the purchaser. 

We selected purchasers for case studies that met the following
criteria:  the purchaser had (1) received performance measurement
information from managed care plans at least twice, (2) documentation
of specific examples of data uses and results, and (3) experience
with managed care markets in several regions of the country or was
able to exercise major leverage as a purchaser in at least one
market.  Also, we sought large purchasers that were willing to allow
us access to their information and to spend time responding to our
questions.  Given these criteria, we selected four purchasers that
represented a range of characteristics and experience with managed
care:  the California Public Employees' Retirement System (CalPERS),
Federal Express, Johnson & Johnson, and Southern California Edison. 

Of the four purchasers we studied, Federal Express and Johnson &
Johnson can be characterized as national, as they purchase care for
large concentrations of employees in multiple markets.  Southern
California Edison and CalPERS can be characterized as regional, as
the vast majority of the employees for whom they purchase care are
located in a single state or market.  Two of the purchasers began
offering managed care to their employees before 1994, and two began
offering managed care since 1994.  (See app.  I for additional
details on each purchaser.)

We performed our work for this study between August 1996 and May 1998
in accordance with generally accepted government auditing standards. 
We also provided a draft of the report to HCFA and the four
purchasers we visited for review and comment.  They provided
technical suggestions, which we have incorporated where appropriate. 


QUALITY DATA SERVE EMPLOYERS,
INFLUENCE PLAN BEHAVIOR, AND
INFORM EMPLOYEES
============================================================ Chapter 2

The four purchasers we studied achieved results--in health plan
access, service by health plans to employees of the purchaser,
satisfaction, and cost savings--by making use of multiple types of
quality-related data, primarily those relating to satisfaction with
care.  They used these data to negotiate increased services from
health plans, improve health plan performance, and inform employees
about their health care choices.  This chapter examines more closely
those uses that have achieved demonstrable results.  To date,
purchaser assessments of health plan quality have largely focused on
issues of accessibility and acceptability and whether health plans
effectively administer their daily operations.  As the four
purchasers evaluate the benefits derived from their and others' use
of quality measures, they anticipate making even greater use of
quality-related data. 


   EARLY RESULTS FROM REQUIRING
   HEALTH PLAN DATA
---------------------------------------------------------- Chapter 2:1

Purchasers can require quality-related data from health plans as a
contracting requirement in order to focus the plans' attention on
purchaser priorities and set the stage for subsequent quality
improvement and accountability activities.  To collect and analyze
quality-related data, purchasers use different types of information
from a variety of sources.  Improvements in access to services and in
health plan capacity to report on HEDIS measures are some of the
results from these activities, according to the purchasers we
visited. 


      PURCHASERS USE DIFFERENT
      TYPES OF DATA FROM DIFFERENT
      SOURCES
-------------------------------------------------------- Chapter 2:1.1

Purchasers use a variety of data sources to assess whether or not to
contract with health plans, monitor their ongoing performance, and
develop quality-related information to provide employees.  Data
sources range from formal data on whether health plans have met
accreditation standards set by entities such as NCQA and JCAHO, how
health plans perform on certain HEDIS measures, and surveys of
employees satisfaction to more qualitative data gathered through the
judgments made by health benefits staff in the process of assessing
health plans during the selection process.  According to the 1997
Watson Wyatt/WBGH survey, the use of health care data is a
resource-intensive activity; therefore, most purchasers who do so are
large companies.  As of 1997, 62 percent of large employers said they
use HEDIS data in making purchasing decisions.  In contrast, only 7
percent of small employers (those with fewer than 1,000 employees)
use HEDIS data. 

Two of the purchasers we visited augment quality-related measures
with site visits when selecting a health plan.\10 To screen and
conduct initial rankings of plans, these two purchasers requested
evidence of NCQA and JCAHO accreditation, various HEDIS measures, and
patient satisfaction surveys.  They also used benefits consulting
firms to assist them in selecting quality-related measures and
analyzing health plan performance against targets, using HEDIS and
other data.  Once plans were screened and ranked, benefits staff
conducted site visits.  For example, one purchaser that we visited
used these visits to observe plan operations, touring plan facilities
including the customer service and claims processing centers and
receiving an overview of the plan's internal quality assurance
processes. 

Site visits can weight heavily when final decisions on health plan
selection are made.  For example, one purchaser ultimately selected a
plan that had not received the highest quality rankings based on the
analysis of quality-related data.  According to the purchaser's
staff, observations during site visits changed the ranking of the
plans.  For example, during site visits at one plan that had received
a high ranking, the purchaser's staff found that medical directors at
some locations in the state did not always know what medical
directors at other locations in the state were doing.  At a site
visit at another plan, the purchaser's staff began to question the
plan's commitment to customer service, given the plan's reaction to
the purchaser's concerns about the process for employee selection of
a primary care physician.  As a result of these site visits, the
purchaser did not select either of these plans. 

Purchasers also acquire data from other sources, such as regional
business coalitions.  One purchaser we visited participated in a
business coalition to augment quality efforts in areas with small
populations of employees.  Two other purchasers we visited said they
benefited from a regional reporting initiative to collect, analyze,
and report audited HEDIS data.  One of these purchasers stressed a
philosophy of building on information that is already publicly
available rather than imposing another reporting requirement on
health plans. 


--------------------
\10 The other two purchasers did not select their health plans using
quality-related criteria, since this selection occurred before their
formulation of a strategy for ensuring quality in managed care. 


      DATA ARE USEFUL FOR
      IMPROVEMENTS IN ACCESS AND
      HEALTH PLAN REPORTING
-------------------------------------------------------- Chapter 2:1.2

As purchasers move into managed care, their first step often is to
ensure access to care.  Purchasers consider data on access as well as
customer service to be particularly important--both to their
employees and as indicators of quality.  The two purchasers that used
quality-related data to select health plans said they had required
health plans to submit data on access-related measures.  One
purchaser, for example, required plans to report on the percentage of
employees who would have access to at least two primary care
physicians within 8 miles of their residence, the average time to
obtain appointments, the percent of primary care providers who were
not accepting new patients, and the timeliness of response to
telephone and member inquiries.  In this case, the purchaser required
a commitment from plans to undertake actions to fill gaps in provider
networks. 

Several purchasers we visited required plans to continue to submit
data on HEDIS measures to ensure the plans gathered and maintained
data on quality.  One purchaser found that the initial HEDIS data
received from plans during the plan selection process may have lacked
validity and reliability.  After requiring HEDIS data for 3 years
from plans and contracting with a consultant to perform a data
quality assessment, the purchaser described significant improvement
in the plans' ability to report and in the reliability of the data
reported.  For example, in 1993, only 50 percent of the managed care
plans under contract could submit the HEDIS data requested, and
purchaser officials described these data as only poor to fair in
quality.  In 1994, over 90 percent of the plans could provide HEDIS
data of "fair quality." By 1995, 100 percent of the plans under
contract reported HEDIS data, and the data submitted by all but three
plans were judged to be of acceptable quality.  The purchaser now
plans to make more use of these improved data during performance
monitoring. 


   DIFFERENT APPROACHES ELICIT
   RESULTS IN HEALTH PLAN
   PERFORMANCE
---------------------------------------------------------- Chapter 2:2

The four purchasers we visited suggested that their philosophies
about their relationship with health plans helped shape the
approaches they use to hold plans accountable for providing quality
health care and bring about improvements in plan performance.  The
four purchasers generally used a combination of collaborative- and
compliance-oriented approaches.  The collaborative approach, based on
a "quality partnering" philosophy, is characterized by a close and
informal relationship between purchaser and plan staff, frequent
discussions about progress made against performance goals and
benchmarks, and jointly developed plans for performance improvement. 
The compliance approach is characterized by techniques such as the
establishment of specific and quantifiable performance standards,
periodic assessment of plan performance against the standards, and
financial penalties for failure to meet the standards. 

Each of the four purchasers were able to identify results achieved
from both approaches, including projects to streamline member access
to specialty care and improvements in employee satisfaction and cost
savings.  While each purchaser tended to use a blend of both
approaches--working collaboratively with plans to improve performance
while holding the same plans accountable against contractual
standards and penalizing them if they do not meet these
standards--all four cited the importance of close interaction with
plans to influence changes in behavior and said that close and
continuous interaction is easier when dealing with a small number of
plans. 


      COLLABORATIVE APPROACH
      TRIGGERS CHANGES AT HEALTH
      PLANS
-------------------------------------------------------- Chapter 2:2.1

In employing a collaborative approach, several purchasers we visited
used quality-related data to highlight problems for discussion with
health plans.  These discussions then triggered actions for
improvement at an individual plan or resulted in the dissemination of
best practices at various plans.  Results achieved through this
approach included the creation of a provider directory to assist
employees in accessing care, the development of joint projects
between purchasers and health plans to enhance ease of referrals to
specialists and to educate employees with diabetes, and streamlining
of procedures for complaints and grievances. 

One purchaser, for example, has been working closely with plans to
improve in areas related to customer service and referrals to
specialists.  The purchaser identified problems in these areas using
an employee satisfaction survey, employee complaints, and feedback
from employee committees established to improve communications
between the purchaser and employees.  For example, approximately 20
percent of employees surveyed were very dissatisfied with the
procedures for changing primary care physicians.  The purchaser
discussed these problems with the health plan during a site visit. 
One month later, the plan distributed listings of primary care
physicians and specialists, including their hospital affiliation. 
The plan also committed to meet weekly with the purchaser to continue
discussing the purchaser's concerns. 

Another purchaser began a joint activity with a health plan after
analyzing data from the purchaser's open enrollment survey and a
member satisfaction survey.  Results of the survey revealed, among
other items, that only 55 percent of employees were satisfied with
ease of referral to a specialist.  In response to these concerns and
the plan's own satisfaction data, the purchaser and one of its health
plans designed a specialist referral project to streamline member
access to specialty care.  Telephone surveys and focus groups were
conducted with four provider groups and members receiving services
from those groups to evaluate the impact of this project.  All
parties--providers, the purchaser, the plan, and member
representatives--are currently meeting with provider groups to design
solutions to member and physician concerns. 

The first purchaser also addressed the issue of specialty referrals
on the basis of data from a satisfaction survey.  These data
indicated that employees perceived specialty referrals as being too
slow and too hard to get.  In some cases, members had to wait for a
review committee at the health plan to approve a referral to a
specialist.  The purchaser's analysis of satisfaction survey data,
coupled with a health plan's own analysis, prompted the appointment
of a task force to develop a referral system.  The system developed
by the health plan gives primary care physicians the authority to
approve referrals on the spot. 

This purchaser also collaborated with a health plan in developing a
diabetes management program, designed to improve patient quality of
life and to reduce emergency room visits.  This program was developed
in response to the prevalence of diabetes among employees and the
purchaser's examination of quality-related data from HEDIS measures. 
After the purchaser initiated discussions with the health plan as
part of its collaborative approach, the plan used its pharmacy
database to identify diabetic employees of the purchaser.  Employees
recruited to participate in the program received educational
materials on diabetes as well as the opportunity to participate in
classes at various work sites.  The plan subsequently surveyed
participants to obtain information on their evaluation of materials
provided and classes attended as well as outcome measures, such as
perception of health status and diabetes-related quality-of-life
measures. 

The purchasers also used quality-related data to identify and
disseminate best practices after holding discussions with a health
plan.  One of the four purchasers, for example, conducts annual
visits at the various sites operated by the plan that serves most of
the purchaser's employees nationwide.  During these visits, the
purchaser and plan managers evaluate plan policies and procedures,
review HEDIS data, conduct clinical audits, and analyze satisfaction
survey data.  At one site, the purchaser's staff identified what they
viewed as an exceptional process for handling appeals.  After they
suggested that this site share its process with other sites managed
by the health plan, the process was implemented in other locations. 


      COMPLIANCE-ORIENTED APPROACH
      HOLDS HEALTH PLANS
      ACCOUNTABLE
-------------------------------------------------------- Chapter 2:2.2

Purchasers have also achieved results by using quality-related data
to assess plan compliance with established contractual standards and
to discipline or reward plan performance.  After applying financial
penalties, one purchaser said it achieved improvements in employee
satisfaction.  This purchaser also documented that it used HEDIS data
as part of the rate negotiation process.  Through this process, the
purchaser communicated its unwillingness to accept higher rate
increases from plans that had not performed as well as others. 

Purchasers often held health plans accountable against contractually
specified standards to meet the purchasers' goals.  For example, one
purchaser developed standards to meet its goal of enhancing the value
of health care services delivered to its members by the year 2000. 
Purchaser standards measuring performance included timeliness of
identification card issuance, evidence of coverage booklet
distribution, speed of written responses, and average time for the
telephone to be answered by a person and telephone abandonment rates. 
In the case of one plan, its performance deteriorated over 2 quarters
on two specific standards:  having a plan representative answer the
telephone within 35 seconds after a caller opted to speak with the
representative and a telephone abandonment rate of less than 5
percent.\11 As a result, the purchaser sent a letter to the plan
requesting that it explain its poor performance and outline its
corrective action.  The plan was also asked to send continuing
commentary on performance in these areas when submitting its
quarterly results on required performance measures.  The plan
responded by consolidating the management of its member services and
by improving the capability of its database and has since improved
its performance.  Other standards developed by purchasers to address
areas of particular concern included identification card accuracy,
appeal and grievance turnaround times, timeliness of data
submissions, and physician turnover. 

Two purchasers imposed financial penalties when specific contractual
standards described in the contract were not met.  The standards
selected described, among other items, specific purchaser
expectations related to the plans' ability to maintain or improve
access and employee satisfaction.  Purchasers also used the rate
negotiation process to reward or penalize plans for their
performance. 

Since its 1994 move into managed care, one purchaser has required the
five health plans covering a majority of its enrolled population to
meet standards in the areas of appeals and grievances; customer
service, including member satisfaction, call abandonment rate, and
telephone response rate; and data reporting, including accuracy and
timeliness.  These standards are specified in a partnership
agreement.  The success of individual health plans at meeting these
standards is subsequently captured in a purchaser scorecard on
individual plan performance.  A distinctive feature of this scorecard
is a subjective, collective assessment by health benefits staff of
how well plans respond to purchaser demands.  If the standards are
not met, this purchaser assesses a financial penalty equal to a
designated small percentage of total revenues under the contract. 
According to the purchaser and health plans, this minor penalty has
helped effect changes in the behavior of health plans, since they
generally wish to avoid the embarrassment of a penalty. 

This purchaser annually evaluates health plan performance with regard
to how well the purchaser's staff thinks the plan responds to these
and other concerns.  The purchaser's staff base their ratings on
their interaction with plan staff during weekly meetings.  For
example, for 1995, one plan was penalized about $9,000 because the
purchaser was dissatisfied with, among other issues, its
responsiveness to purchaser concerns.  For 1996, the purchaser found
the plan to be more responsive and no penalties were levied for the
plan's failing to meet this performance standard.  However, during
the same 1995 to 1996 time period, the purchaser staff continued to
be dissatisfied with the plan's commitment to customer service.  For
1995, it was penalized approximately $6,000, and for 1996, it was
penalized about $7,000. 

Another purchaser also attributed improvements in quality to the use
of financial penalties.  As an example, this purchaser established a
contract standard requiring plans to maintain an 85-percent
satisfaction rate among its employees.  Data submitted for the plan's
midyear review showed that its rate fell from 91 to 84 percent.  When
the plan investigated, it found that the purchaser's employees felt
plan providers lacked empathy.  The plan instituted training for the
providers.  Six months later, employee satisfaction had risen to 93
percent. 

Another tool used by purchasers to evaluate health plans at the end
of a period is the annual contract renewal and rate negotiation
process.  Purchasers can use quality-related data to reward or
penalize plans as part of this process and, as a result, believe that
they are improving the value of their health care purchasing
decisions.  Officials at one purchaser said they were able to improve
the quality of health care while holding the line on costs because
the purchaser's rates are based in part on rewarding health plans for
high performance. 

Beginning in 1996, another purchaser began to target plans that
proposed rate increases but had low overall HEDIS scores for further
conversation.\12 According to the consultant hired by the purchaser,
the first year this strategy was used, four targeted managed care
plans had proposed a 2-percent increase in premiums.  After rate
negotiations, the premiums decreased by 7 percent.  For the next
year, 21 targeted plans had proposed a 6-percent increase.  A
4-percent decrease was achieved through rate negotiations.  The
purchaser attributes the premium decreases to the use of HEDIS data
in negotiations in addition to the analysis of administrative fees,
average charge per member per month, and a comparison with similar
plans in the same geographic area and with their regional claims
experience.  The purchaser is currently studying the relationship
between the cost savings achieved through rate negotiations and
quality of care. 


--------------------
\11 A telephone call is considered abandoned when a caller hangs up
after requesting to speak with a customer representative but before
the representative answers. 

\12 This purchaser contracts with over 40 managed care plans. 


   PURCHASERS SHARE QUALITY DATA
   WITH EMPLOYEES
---------------------------------------------------------- Chapter 2:3

In addition to taking actions to elicit changes at health plans,
purchasers can also use data about quality to help employees make
informed choices in selecting plans.  Report cards provide the
results of cost and quality indicators, as well as other descriptive
information, comparing the performance of competing health plans. 
Some believe that as consumers become better informed and decide not
to select health plans of lesser quality, such plans may be motivated
to initiate improvements in the quality of care they provide. 
Research on report cards indicates that these formats are continuing
to evolve as a way of presenting quality-related data.  We found that
of the two purchasers using report cards, one purchaser surveyed
employees and concluded that employees found the information useful. 
The other saw only a modest increase in employee selection of the
plan with the highest quality ranking in the report card.\13

One purchaser that disseminated information to employees collaborated
with the magazine Health Pages to report information about the
quality of health plans the purchaser offered to its employees. 
Information in this magazine included general descriptions of the
plans; characteristics of the plans; physician and hospital networks;
information about preventive care, such as the rates at which plans
administer childhood immunizations or perform cholesterol screenings;
and satisfaction ratings.  The other purchaser that disseminated
information to its employees produced and distributed its own report
cards comparing offered health plans during the open enrollment
period.  For example, one report card gave prospective enrollees
comparative information on HEDIS measures in three areas:  preventive
health services (childhood immunizations and cholesterol screening),
women's preventive health (prenatal care and pap smear and
mammographies), and care for chronic illness (diabetic eye
examinations).  The report cards used by each purchaser also
contained narrative material explaining the importance of such
measures. 

The purchasers using report cards to educate employees as part of the
enrollment process saw some initial results from their decision to
disseminate comparative information.  For example, from an employee
survey intended to assess the effect of its first report card on
enrollment behavior, one purchaser found that 66 percent of those
responding viewed the purchaser's report card as very or somewhat
important in assisting members in selecting their plan.\14 The
purchaser did not use a survey to assess the effect of its second
report card; however, they did examine several hundred write-in
responses returned on an enclosed tear-out sheet.  The most frequent
employee recommendation for future report cards was to include more
data about the quality of each plan.  Members also recommended
providing (1) easy-to-read comparisons, such as those found in
Consumer Reports; (2) feedback from existing or previous plan
members; and (3) information on complaints filed against physicians
or hospitals.  A subsequent report card reflected the first two
recommendations.  This report card also contained information based
on most frequently asked questions in such areas as administrative
policies, prescription drugs, disenrollment statistics, type of
physician specialties offered in the plan, and NCQA accreditation
status.  The purchaser has not evaluated if employees moved into
health plans on the basis of report card information. 

The other purchaser assessed the effect of providing employees with
comparative information by examining the extent to which enrollees
actually shifted into the plan with the highest quality ranking.  The
purchaser concluded that a modest shift had occurred.  The purchaser
subsequently froze enrollment in plans with continuing quality
problems and saw a more significant shift as a result. 


--------------------
\13 For the purchasers that did not disseminate information, one
concluded that it had exercised sufficient care in health plan
selection and did not need to disseminate quality-related information
to its employees.  The other purchaser decided not to disseminate
HEDIS information because of concerns about the reliability of
self-reported data by health plans. 

\14 However, an almost equally high percentage of respondents (63.9
percent) also cited recommendations of friends, coworkers, and family
members as important sources of information in their decisionmaking
process.  The response rate to this survey was less than 50 percent
(7,990 of 16,762 surveyed). 


   WHERE PURCHASERS ARE HEADED
   WITH QUALITY DATA
---------------------------------------------------------- Chapter 2:4

To achieve greater results from the use of quality-related data in
the future, the purchasers we visited see future opportunities to
rely on such data for selecting and monitoring the performance of
health plans, rewarding or penalizing plans through rate
negotiations, as well as informing and educating their employees. 
They have already begun or are planning to use quality-related data
to (1) discriminate among and contract with fewer plans to make
quality oversight and monitoring efforts more effective; (2) decide
whether to renew contracts with plans; (3) translate performance
goals into contractual standards; (4) present multiple types of data
to health plans through combined formats, known as scorecards; and
(5) negotiate rates with and provide financial incentives to
employees to choose plans with higher quality rankings. 

The purchasers we visited are also beginning to use quality-related
data to more closely focus their efforts on issues of particular
concern to them, such as provider and health plan relationships. 
Despite concerns over existing measures, several purchasers plan to
make greater use of HEDIS measures.  Another approach to care has
been taken by a national purchasing coalition, which conducts
in-depth reviews of costly and seriously ill cases for their
purchaser members as part of ensuring health plan quality. 


      HEALTH PLAN SELECTION
-------------------------------------------------------- Chapter 2:4.1

Purchasers intend to make various changes in how they select health
plans in the future.  For example, one purchaser first focused use of
quality-related data to select plans in areas where its employees are
geographically concentrated.  This purchaser now plans to begin using
such data in selecting plans in areas with fewer employees.  Other
purchasers would like to use quality-related data to contract with
fewer plans to make quality oversight and monitoring efforts more
effective and cost efficient and to eliminate poorly performing
health plans that are unable to demonstrate improvement.  For some
purchasers, quality-related data have not been sufficiently reliable
and valid for decisionmaking.  However, once these concerns are
resolved, several purchasers may move to use quality-related data as
a basis for not renewing contracts with poorly performing plans. 
Contracting with fewer plans may mean that the purchaser does not
need to expend as many resources for monitoring. 


      PERFORMANCE MONITORING
-------------------------------------------------------- Chapter 2:4.2

Purchasers see numerous ways to increase the use of quality-related
data when monitoring health plan performance.  Several purchasers we
visited plan to develop new contractual standards to more effectively
hold health plans accountable.  For example, one purchaser plans to
translate its existing performance goals into contractual standards. 
Originally, it had issued these goals with the expectation that
health plans would continuously strive to address areas of importance
to the purchaser regardless of whether the goals appeared in the
contract.  By translating performance goals into contractual
standards, this purchaser hopes health plan accountability will
improve. 

For another purchaser, if a plan's general satisfaction performance
falls below a certain level, the health plan becomes a candidate for
quality improvement dialogues and may be selected for more in-depth
surveys or reviews of employee satisfaction.  This purchaser, which
has had multiple measurement initiatives, also plans to analyze
satisfaction, HEDIS, and other measures that need to be consolidated
to create an overall scorecard--an approach already taken by another
purchaser we visited.  By assigning weights to various indicators of
performance--including financial, clinical, and customer
service--purchasers can give health plans an overall quality index
score and present the results in a quality assessment instrument, or
scorecard.  The advantage of this approach is that multiple sources
of information can be presented in a comprehensive format, which can
be used by purchasers to discuss health plan performance.  Intended
to reflect an employer's specific health care benefits strategy, in
some cases, these scorecards can be associated with rewards for good
performance and incentives for improving poor performance.  A
purchaser other than the four we visited, for example, will use its
scorecard to reward those plans that have performed well with
incentive payments and give plans with low scores the opportunity to
improve over a reasonable time frame.  However, if such plans do not
improve their performance, they could risk losing this company's
business. 


      RATE NEGOTIATIONS
-------------------------------------------------------- Chapter 2:4.3

The four purchasers we visited all recognize the need to incorporate
additional performance and quality measures into the annual contract
renewal and rate negotiation process.  As explained by two
purchasers, future negotiations of rates with health plans must
achieve a balance between cost and quality.  For these purchasers, a
focus on costs to the exclusion of quality will result in a decline
in the overall value of care.  One purchaser plans to incorporate
results of its health plan performance scorecard into rate
negotiations. 


      REPORT CARDS
-------------------------------------------------------- Chapter 2:4.4

As data improve, purchasers plan to improve their report cards that
compare plan quality.  Both our study and the Watson Wyatt/WBGH
survey have found increasing use of these reports by large
purchasers.  We have reported that many purchasers are moving toward
greater use of report cards and that others plan to do so in the near
future.\15 According to the Watson Wyatt/WBGH survey, 33 percent of
large purchasers give their employees information about accreditation
status and 26 percent give their employees HEDIS information.  While
many purchasers are moving toward using report cards, there are
concerns about performance reports, such as the reliability and
validity of data, the need for more readily available and
standardized information, and a greater emphasis on outcome
measures.\16


--------------------
\15 Health Insurance:  Management Strategies Used by Large Employers
to Control Costs (GAO/HEHS-97-71, May 6, 1997).  Our review of large
purchasers found that about half of the purchasers in our sample
currently provide employees a report card on the HMOs that they
offer, and others were planning to do so.  These report cards focused
on the results of employee satisfaction surveys and, to a lesser
extent, health plan performance in delivering HEDIS preventive
services, such as immunizations or cancer screenings. 

\16 Health Care:  Employers and Individual Consumers Want Additional
Information on Quality (GAO/HEHS-95-201, Sept.  29, 1995). 


      PREFERRED PRICING
-------------------------------------------------------- Chapter 2:4.5

One purchaser that published a report on plans for employees has not
seen desired movement to the most highly ranked plan.  Therefore, it
intends to implement preferred pricing to encourage employees to move
to more highly ranked plans by setting lower employee premiums for
these plans.  In the meantime, this purchaser froze enrollment in one
plan, which had continuing quality problems.  According to the Watson
Wyatt/WBGH survey, 32 percent of the large purchasers who responded
to the survey offered some type of financial incentive to employees
to choose plans deemed to be of exceptional quality by the purchaser. 
This technique also rewards plans designated by a purchaser as being
of high quality because it encourages enrollment in these plans.  One
purchaser that we did not visit attributes desirable results to
preferred pricing.  The purchaser ranked performance in eight
selected quality categories for managed care plans and disseminated
this information as part of a medical plan guide during the annual
health care and benefits enrollment process.  The purchaser claims
significant enrollment increases in top-rated plans and decreases for
below-average plans.  According to a purchaser official, its efforts
to reward workers for selecting good plans led to an almost
13-percent increase in enrollment for these plans. 


      PROVIDER ISSUES
-------------------------------------------------------- Chapter 2:4.6

In considering the next step in the use of quality-related data, some
purchasers plan to move from a focus on health plan quality to
exploring the use of data related to provider quality.  Two
purchasers plan initiatives based on the use of satisfaction data to
identify problems with specialist physician referrals.  One
purchaser, for example, will launch an initiative to collect
quality-related data on providers and to later issue report cards on
provider performance.  In addition, this purchaser has begun
conversations with providers to gain a better understanding of how
health plans and providers could relate more effectively.  The
purchaser hopes to develop an approach that will financially reward
health plans for prompting desired changes in provider behavior. 


      HEDIS MEASURES
-------------------------------------------------------- Chapter 2:4.7

The purchasers that we visited told us that they plan to make more
use of data from HEDIS and other measures as they become available. 
One purchaser noted that NCQA's database of managed health care
information, Quality Compass, will be helpful in producing
user-friendly reports for employers.  Quality Compass contains
performance, accreditation, and patient satisfaction information from
more than 300 managed health care plans throughout the United States. 

In general, purchasers appear to have mixed views on the use of HEDIS
measures.  We have found that some purchasers are reluctant to
disseminate information on HEDIS measures to their employees.\17
Purchasers have expressed concerns over self-reported data that are
not independently audited, and a recent study notes that many health
plans are struggling to provide data on all of the measures and some
fail to produce any data.\18 However, NCQA recently announced that it
will certify organizations to perform audits of HEDIS data.  This may
further improve the quality of data that purchasers receive from
health plans. 

Purchasers other than those we visited appear to have made much more
extensive use of HEDIS measures.  They use the data to select plans,
monitor changes in performance over time, and establish benchmarks
and minimum standards.  According to one study, many work with their
health plans to identify best practices and develop strategies for
quality improvement.  Also, some companies have incorporated
performance on HEDIS measures as part of their pricing strategies.\19

The purchasers we visited plan to continue making use of HEDIS and
other quality-related measures as they are refined and new ones
become available.  In contrast, a national purchasing coalition does
not rely exclusively on existing quality-related measures but rather
uses medical audits to determine whether managed care plans have the
systems in place to respond to and appropriately manage patients with
potentially serious and costly episodes of illness. 


--------------------
\17 GAO/HEHS-97-71, May 6, 1997. 

\18 Jack Meyer and others, Theory and Reality of Value-Based
Purchasing. 

\19 Lise Rybowski, Putting HEDIS to Work:  Employer Initiatives to
Promote Quality in Health Care (Washington, D.C.:  Employers' Managed
Health Care Association, 1996). 


OPPORTUNITIES FOR HCFA TO BENEFIT
FROM PURCHASERS' EXPERIENCES
============================================================ Chapter 3

Although the characteristics of the Medicare program have
distinguished it from other purchasers and shaped HCFA's major
strategies for ensuring quality care for beneficiaries enrolled in
HMOs, the passage of the Balanced Budget Act of 1997 makes the
experience of other purchasers more relevant to HCFA.  This
legislation gives the Medicare program authority to contract with new
types of managed care plans and calls for the program to provide
quality-related and other comparative data to beneficiaries to
promote a more informed selection of health plans.  It is expected to
result in more plans contracting with the program and more
beneficiaries enrolling in plans.  The legislation also requires
managed care plans to take action to improve quality of care.  As a
result, HCFA will now begin to look more like purchasers that enroll
most of their employees in managed care and those that provide
comparative information on health plans to their employees as well as
use quality-related data to prompt health plans to improve their
performance. 

The experiences of purchasers we visited have implications for HCFA
in three ways:  (1) educating beneficiaries as to the meaning of
quality-related measures when providing comparative information on
health plan quality; (2) interacting with health plans to take
action, either through a collaborative- or a compliance-oriented
approach, when problems with health plan performance are surfaced by
quality-related data; and (3) continually looking for additional
opportunities to make use of quality-related data. 


   HCFA IS UNIQUE AS A PURCHASER
---------------------------------------------------------- Chapter 3:1

Perhaps the most striking difference between HCFA and other
purchasers has to do with the enormity of HCFA's presence in the
marketplace.  Although HCFA is the nation's largest purchaser of
health care, only a small percentage of Medicare beneficiaries have
decided to enroll in HMOs, although this number has been rising
sharply in recent years.  Nevertheless, the sheer number of Medicare
beneficiaries in managed care far exceeds the number of employees who
would be enrolled in managed care by a private company.  The
purchasers we visited now enroll most of their employees in managed
care health plans.  The largest purchaser we visited serves 1 million
people, while HCFA in its Medicare managed care program currently
serves over 5.5 million beneficiaries, with potentially many more
expected according to CBO estimates. 

HCFA also differs from other purchasers in the freedom of choice
enjoyed by Medicare beneficiaries who have had far more latitude in
selecting options for health care than others.  Much of the privately
insured population under 65 only has access to those health plans
selected by their employer, and in many cases, the employer just
chooses one plan.  They also only have the option to enroll or
disenroll during a specified "open season." In contrast, Medicare
beneficiaries have been able to select any of the Medicare-approved
HMOs in their area and may switch plans monthly or choose the
fee-for-service program.  HMOs have been able to market their plans
to Medicare beneficiaries throughout the year, not only during the
required 30-day open enrollment period. 

The structure of the Medicare program, unlike private sector care, is
determined by law and regulation.  Any eligible health plan that
agrees to meet minimum standards may participate in the Medicare
program.  In contrast, private sector purchasers can engage in
"selective contracting" to select plans with lower costs and to use
quality-related and other data in their selection decisions.  As a
result, they can exclude plans as part of the selection process.  In
contrast, HCFA does not have the flexibility of refusing to contract
with plans that meet its minimum standards.  In markets such as Los
Angeles, HCFA contracts with 14 health plans; other large purchasers
in that area contract with a smaller number of plans and claim that
contracting with fewer plans enhances a purchaser's ability to more
effectively oversee the quality of health plan performance.  While
other purchasers have more flexibility than HCFA in selecting plans,
Medicare HMO beneficiaries in certain parts of the country have the
ability to choose among more managed care plans than may be available
to employees. 

HCFA also differs from other purchasers in how HMO prices are set. 
Other purchasers can negotiate rates with health plans on the basis
of performance measured against preestablished standards.  In
contrast, Medicare HMO rates are determined by statutory formula,
which does not allow the flexibility of negotiation.\20

Like other purchasers, HCFA monitors HMO performance but does so
according to law.  Its two principal strategies are its HMO
monitoring program and review by peer review organizations.  The
monitoring program implements requirements ranging from financial
solvency requirements to grievance procedures.  After a Medicare
contract is awarded, HCFA regional staff have the responsibility of
monitoring HMOs against federal statutory and regulatory requirements
as part of on-site biennial reviews.  HCFA is also required to
contract with peer review organizations, also known as quality
improvement organizations, which are physician organizations in each
state that review HMO quality of care.  In the past, these
organizations attempted to determine instances of poor care through
medical record reviews.  In recent years, quality improvement
organizations and plans have begun to conduct quality improvement
projects in different clinical areas.  For example, in one project,
quality-related measures have been used to collect information,
provide feedback to plans on their performance, and design
interventions to improve the quality of care in outpatient diabetes
management. 


--------------------
\20 The Balanced Budget Act of 1997 establishes a competitive pricing
demonstration to begin in 1999.  However, this project is temporary
and limited to a specified number of geographic areas. 


   PURCHASER EXPERIENCES WILL HAVE
   INCREASED RELEVANCE FOR HCFA
---------------------------------------------------------- Chapter 3:2

A goal of title IV of the Balanced Budget Act is to encourage
Medicare beneficiaries to enroll in managed care health plans.  CBO
estimates suggest that HCFA's presence as a purchaser of managed care
will become even more pronounced than at present and that HCFA will
need to become more active in its oversight functions.  In this
regard, the information the purchasers provide employees and
purchasers' monitoring experiences are especially relevant. 

The Balanced Budget Act establishes specific time frames for HCFA to
meet in providing beneficiaries comparative information on covered
benefits, premiums, and quality and performance of managed care plans
to guide their enrollment decisions.  Although HCFA had the authority
to provide beneficiaries with such information prior to the act, past
work by GAO found that HCFA was not doing so and recommended that
HCFA help elderly consumers choose among competing Medicare HMOs by
distributing comparative information on HMOs.\21

According to HCFA, the agency had already begun to move in this
direction prior to the passage of the Balanced Budget Act.  The new
legislation, however, not only establishes specific time frames for
HCFA to meet but also couples the provision of comparative
information with an annual open enrollment season.  By the year 2002,
with limited exceptions, Medicare beneficiaries who enroll in a
health plan will only be able to enroll in another plan during
periodic coordinated open enrollment seasons, whereas at present they
can switch at any time.\22 The reduced ability of Medicare HMO
enrollees to freely change plans places an additional responsibility
on HCFA for ensuring the quality of care that HMO enrollees receive. 

Purchasers have had a variety of experiences in distributing
comparative information on health plans to their employees.  One
purchaser we visited surveyed employees on its report card and
concluded that employees found the information to be useful.  The
purchaser later used feedback from employees to modify the report
card and enhance its usefulness.  The effect of the report cards,
however, is not yet clear.  For example, another purchaser found only
a modest shift of employees into the plan with the highest quality
ranking and decided to encourage changes in employee behavior by
freezing enrollment in plans with continuing quality problems.  In
designing these report cards, both purchasers provided explanatory
material so that employees would be able to better understand the
meaning of the measures employed. 

These experiences by purchasers are relevant to HCFA.  Not only do
they provide comparative information to their employees, but in using
feedback from employee surveys and by assessing the impact of such
information on employee behavior, purchasers demonstrate that they
continually review the value and utility of the information they
present to their employees.  For HCFA, this implies continual
monitoring of how consumer information is used.  Other lessons for
HCFA relate to the need for purchasers to educate employees on the
meaning of the measures contained in the report card and how to
interpret these measures when deciding between health plans. 

In addition to providing quality-related data to employees,
purchasers also provided this information to plans and expected the
plans to take action on the basis of this information.  The Balanced
Budget Act provides a more explicit listing of required elements for
plan quality assurance programs than had been required before.  These
elements include requirements for plans to take action to improve
quality and assess the effectiveness of such action through
systematic follow-up.  In relation to this requirement, HCFA is
considering how to use standardized measures to prompt quality
improvement activities.  Elements of this approach have already been
present in collaborative projects between quality improvement
organizations and health plans. 

Again, the experience of the four purchasers we visited can inform
how HCFA addresses the quality assurance provisions in the Balanced
Budget Act.  The four we visited did not simply provide health plans
with data from HEDIS measures, satisfaction surveys, and other
sources of information.  They also met and took follow-up steps to
ensure the plans were taking action to improve health plan
performance and achieve what they described to us as promising
results.  In addition, some purchasers used the information to
penalize plans that were not meeting their standards.  For example,
one purchaser alerted employees to problems in health plan
performance by freezing enrollments in a plan that had performance
problems. 

Purchasers emphasized the importance of interaction with plans and of
blending techniques from two purchasing philosophies--one oriented
toward quality partnering and the other toward ensuring compliance
with standards set by the purchaser.  In the same way that purchasers
refine the information provided to employees, they continue to
reevaluate the ways in which they provide information to health plans
on their performance. 

While HCFA can examine how other purchasers use quality-related data
in some areas, it would need new legislative authority to implement
other purchaser practices in using quality-related data.\23 These
include the use of quality-related data to selectively contract and
negotiate rates with health plans. 


--------------------
\21 Medicare:  HCFA Should Release Data to Aid Consumers, Prompt
Better HMO Performance (GAO/HEHS-97-23, Oct.  22, 1996). 

\22 From 1998 through 2001, Medicare beneficiaries can continue to
enroll, if the plan is open to new enrollees, or disenroll on a
monthly basis. 

\23 "Health Care Purchasing Strategies," Internal HCFA Report (Dec. 
1996). 


DESCRIPTIONS OF THE PURCHASERS GAO
VISITED
=========================================================== Appendix I

Tables I.1 through I.4 provide brief descriptions--including number
covered, enrollee locations, purchaser goals, and managed care
experience--of the large corporate purchasers we visited as well as
the purchasers' quality strategies. 



                               Table I.1
                
                 Brief Description of Johnson & Johnson
                                 (J&J)

----------------------  ----------------------------------------------
Number of covered       72,000
lives

Primary location of     California, Florida, New Jersey, and Texas
enrollees

Health care purchasing  --Reduce the rate of increase in health care
goals                   costs.
                        --Offer employees a choice of quality medical
                        plan options.
                        --Ensure provider choice and access.
                        --Ensure plan quality and employee
                        satisfaction.

Experience with         J&J first offered a managed care point-of-
managed care            service option to its employees in 1995; it
                        had previously offered a traditional indemnity
                        option along with HMOs. By 1997, nearly 80
                        percent of the company's enrollees were
                        covered by self-funded managed care health
                        plans, including point-of-service and HMO
                        options. J&J also contracts with fully insured
                        HMOs in areas of the country with fewer
                        employees and maintains its traditional
                        indemnity plan.

History of quality      J&J initially focused on selecting and
strategy                monitoring plans that enrolled a majority of
                        employees. J&J also required health plan
                        account representatives to attend a training
                        and orientation program. J&J assesses
                        financial penalties when expectations are not
                        met according to a performance scorecard. J&J
                        also analyzes individual complaints to
                        determine whether they are symptoms of an
                        underlying, systemwide problem and demands
                        documentation from health plans on how
                        complaints are resolved.

Future plans            J&J plans to extend monitoring efforts by
                        collecting more quality-related data on fully
                        insured HMOs, use multiple sources of data in
                        a balanced scorecard format, drop health plans
                        for poor performance based on data, gauge
                        enrollee satisfaction by reviewing enrollment
                        trends, and develop a methodology for
                        independent verification of quality-related
                        and other performance data.
----------------------------------------------------------------------


                               Table I.2
                
                  Brief Description of Federal Express
                                (FedEx)

----------------------  ----------------------------------------------
Number of covered       120,000
lives

Primary location of     California, Florida, Illinois, Indiana, New
enrollees               Jersey, New York, Tennessee, and Texas

Health care purchasing  Improve value and provide employees with a
goals                   choice of plans and providers.

Experience with         In 1982, FedEx offered the first "local HMO"
managed care            (an HMO serving a limited geographic area) in
                        the New York area. Between 1984 and 1991, the
                        company launched 46 local HMOs. Additional
                        options were rolled out by market; California-
                        based employees offered a self-funded POS/HMO
                        option in 1993 with the same option extended
                        in 1994 to employees in other locations. By
                        1997, about 78 percent of employees were
                        enrolled in one of three managed care options:
                        a self-funded POS/HMO, a basic preferred
                        provider organization, or a fully insured
                        local HMO; the remaining 20 percent were still
                        enrolled in the basic indemnity plan.

History of quality      FedEx established preferred pricing for its
strategy                national plan and has worked extensively with
                        a consultant to collect and analyze first
                        utilization data. More recently, FedEx has
                        used HEDIS data and provided feedback in
                        "dialogues" if plans scored poorly either due
                        to low HEDIS scores or incomplete data
                        submissions. FedEx was also an early
                        participant in the areawide business coalition
                        that acheived improvements in quality at area
                        hospitals.

Future plans            After achieving success in cost control
                        through managed care, FedEx has reorganized
                        its benefits function in a step to improve
                        measurement and improvement of quality at
                        health plans. FedEx plans to communicate more
                        quality-related data to employees (with an
                        emphasis on satisfaction data) and, through
                        scorecards combining HEDIS and other types of
                        quality-related data, provide consolidated
                        feedback to plans.
----------------------------------------------------------------------


                               Table I.3
                
                Brief Description of Southern California
                              Edison (SCE)

----------------------  ----------------------------------------------
Number of covered       54,000
lives

Primary location of     Arizona, California, and Nevada
enrollees

Health care purchasing  --Manage costs.
goals                   --Improve health plan quality and service.
                        --Promote consumer education.

Experience with         SCE introduced several HMOs in the mid-1970s
managed care            and from 1989 to 1995 administered a self-
                        insured PPO and acted as both purchaser and
                        provider, with on-site doctors and clinics. In
                        1995, the PPO was replaced with seven plans
                        offering standardized benefits. About 94
                        percent of the company's enrolled population
                        is now covered by four health plans, three of
                        which offer both a POS and HMO option.

History of quality      SCE emphasizes a quality partnering approach,
strategy                which it describes as relationship-driven,
                        with continued refinement of its quality
                        strategy over a 3-year period. Health plan
                        site visits to discuss HEDIS, satisfaction
                        survey results, complaints, report cards, and
                        measurement of plan performance against the
                        company's performance goals are a central tool
                        of this strategy. SCE participates with a
                        business coalition to administer a
                        comprehensive member satisfaction survey and
                        collaborates with other purchasers, health
                        plans, and medical groups to obtain audited
                        HEDIS data. SCE also uses report cards to
                        present quality-related data to employees and
                        holds meetings with consumer committees
                        (representing employees, retirees, and union
                        representatives) to discuss issues in health
                        plan performance and emerging trends in the
                        management of health care delivery.

Future plans            SCE plans to continue efforts to improve
                        performance and accountability at the health
                        plan, medical group, and provider levels and
                        to communicate health plan and medical group
                        quality indicators to plan participants. SCE
                        may implement incentives to encourage
                        enrollment in plans that are the highest
                        performers.
----------------------------------------------------------------------


                               Table I.4
                
                 Brief Description of California Public
                 Employees' Retirement System (CalPERS)

----------------------  ----------------------------------------------
Number of covered       1,000,000
lives

Primary location of     California
enrollees

Health care purchasing  --Ensure the availability of affordable,
goals                   quality health care for all participants.
                        --Provide leadership in health care purchasing
                        and quality.

Experience with         CalPERS has offered managed care since 1962.
managed care            In 1989, its fee-for-service plans were
                        consolidated into one PPO; in 1993, a second
                        PPO product was introduced. Currently, 19
                        percent of covered lives are enrolled in PPOs
                        and 81 percent in fully insured HMOs. In 1992,
                        a standard benefit design was implemented to
                        allow the purchaser and enrollees to make more
                        meaningful comparisons of HMOs. The number of
                        HMOs contracted has dropped from 23 to 10 due
                        to changes in the health care industry, such
                        as mergers; new plans will be added only if
                        they cover previously unserved areas.

History of quality      Successful cost containment efforts raised
strategy                concerns over impact on quality care. CalPERS,
                        however, found early efforts to measure
                        quality were inhibited by lack of reliable,
                        comparable data and characterizes its approach
                        as "conservative and incremental." CalPERS
                        distributed health plan report cards for the
                        1995, 1996, and 1997 benefit years, with the
                        first presenting comparative HEDIS and
                        satisfaction data; the second was expanded to
                        provide survey results on why members changed
                        health plans; and the third added answers to
                        frequently asked questions by enrollees,
                        including NCQA accreditation and other plan
                        information. CalPERS participates with a
                        business coalition in a comprehensive member
                        satisfaction survey and collaborates with
                        other purchasers, health plans, and medical
                        groups to obtain audited HEDIS data.

Future plans            CalPERS plans to expand its use of report
                        cards to include comparative disease
                        management outcomes and complaint monitoring
                        results. As data improve, CalPERS plans to
                        increase its use of contractual quality
                        standards and consider financial incentives
                        for plans to improve.
----------------------------------------------------------------------

MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II

Nancy Donovan, Evaluator-in-Charge, (202) 512-7136
Dawn Shorey, Senior Evaluator


*** End of document. ***