University of Southern California Dissertations and Theses

An analysis of health risk selection and quality of care under Medicare fee -for -service and Medicare managed care health care systems

(USC Thesis Other)

An analysis of health risk selection and quality of care under Medicare fee -for -service and Medicare managed care health care systems

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content AN ANALYSIS OF HEALTH RISK SELECTION AND QUALITY
OF CARE UNDER MEDICARE FEE-FOR-SERVICE AND
MEDICARE MANAGED CARE HEALTH
CARE SYSTEMS
by
John A. Salisbury
A Dissertation Presented to the
FACULTY OF THE SCHOOL OF POLICY, PLANNING
AND DEVELOPMENT
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PUBLIC ADMINISTRATION
May 2003
Copyright 2003 John A. Salisbury
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3103964
Copyright 2003 by
Salisbury, John Arnold
All rights reserved.
®
UMI
UMI Microform 3103964
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
SCHOOL OF POLICY, PLANNING, AND DEVELOPMENT
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90089
This dissertation, written by
JOHN ARNOLD SA L IS B U R Y
under the direction o f hi.s Dissertation
Committee, and approved by all its
members, has been presented to and
accepted by the Faculty of the School of
Policy, Planning, and Development, in
partial fulfillment o f requirements for the
degree o f
DOCTOR OF PUBLIC ADMINISTRATION
D ean
Date....
DJSSkK’fA JlSfCO M M ITTEE
Chairperson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
I would like to thank Michael Cousineau, Ph.D. for introducing me to and
affording me access to the 1997 Los Angles County Health Survey. I would also like
to thank Robert Stallings, Ph.D. for his patient guidance into and through the
statistics and data analysis.
I owe a significant debt of gratitude to my Dissertation Committee, Dowell
Myers, Ph.D., Kate Wilber, Ph.D. and Committee Chair Robert Myrtle, DPA. for
their forbearance as well as insight and direction in the creation and, finally, the
completion this study.
I would like to express particular appreciation to Dr. Myrtle who spent many
long and late hours guiding me thorough the academic as well as dissertation
processes.
Gratitude to Marion Thielmann, whose faith, love and backing ultimately
made this project viable.
Special appreciation and gratitude goes to Howard and Nedra Salisbury, my
parents, whose love of learning lit the way along this long, long journey.
Finally, love and appreciation go to Jean E. T. Salisbury, Ph.D., who blazed
the trail and set the standard. Her unwavering confidence and support, not to
mention innumerable hours of proofreading and editing, made this document a
comprehensible reality.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS................................................................................... ii
LIST OF TABLES............................................................................... vi
LIST OF FIGURES.............................................................................. viii
ABSTRACT............................................................................................................ ix
Chapter Page
1. INTRODUCTIONS......................................................................... 1
Medicare................................................................................................ 4
Managed Care...................................................................................... 6
Purpose of the Study............................................................................ 9
Research Question A .................................................................. 10
Research Question B .................................................................. 13
2. LITERATURE REVIEW..................................................................... 18
Early Literature.................................................................................... 18
Commercial HMOs.............................................................................. 22
Hospital Utilization.................................................................... 23
Physician Office Visits............................................................... 24
Prevention and Health Promotion............................................. 24
Quality of Care..................................................................................... 25
Enrollee Satisfaction.................................................................. 27
Costs............................................................................................. 28
Medicare HMOs................................................................................... 28
Growth of Risk Contracts.................................................................... 29
The Behavioral Model of Health Services Utilization...................... 34
System......................................................................................... 37
Population................................................................................... 37
Utilization.................................................................................... 38
Satisfaction.................................................................................. 39
Access and Managed Care.................................................................. 39
Satisfaction and Managed Care........................................................... 40
Outcomes and Managed Care............................................................. 41
Costs and Managed Care..................................................................... 44
Medicare’s Capitation Rate................................................................. 46
Risk Selection...................................................................................... 47
Importance of Risk Selection.............................................................. 49
Risk Selection and Nonelderly............................................................ 50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iv
Chapter Page
Risk Selection and Medicare............................................................... 52
Regression Toward the Mean.............................................................. 55
Summary of Literature......................................................................... 60
Research Questions.............................................................................. 62
Quality of Care, Service Access, and Patient Satisfaction................ 65
Geographically Explicit Sampling...................................................... 65
Summary............................................................................................... 67
3. METHODOLOGY..................................................................................... 68
Data Source and Method..................................................................... 68
The Survey Questionnaire................................................................... 68
Survey Data.......................................................................................... 71
Analysis of Data................................................................................... 72
Healthcare Variables............................................................................ 73
Specificity of Locale............................................................................ 79
The Telephone Survey......................................................................... 79
Statistical Software............................................................................... 82
4. FINDINGS.................................................................................................. 83
Definition of Variables......................................................................... 83
Demographics/Descriptive Statistics................................................. 84
Tests of the Hypotheses...................................................................... 89
Research Question A ..................................................................... 89
Health Risk Variables.......................................................................... 91
Hypothesis 1................................................................................ 91
Quality of Care Variables.................................................................... 97
Hypothesis II............................................................................... 98
Satisfaction.................................................................................. 102
Afford Medication Treatment................................................... 104
Summary............................................................................................... 110
Post Hoc Multivariate Analysis.......................................................... 110
Summary............................................................................................... 118
5. CONCLUSIONS, IMPLICATIONS AND RECOMMENDATIONS ... 119
Health Risk Variables.......................................................................... 120
Self-reported Health Status........................................................ 120
Chronic Conditions.................................................................... 120
Hospital Admissions.................................................................. 121
Regular Source of Care.............................................................. 122
Quality of Care..................................................................................... 122
Doctor Visit................................................................................. 123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
Chapter Page
Satisfaction.................................................................................. 124
Afford Medication or Treatment............................................... 124
Preventive Procedures................................................................ 126
Summary of Findings................................................................. 126
Policy Implications and Conclusions................................................. 127
Study Limitations and Future Research.............................................. 132
Limitations.................................................................................. 132
Future Research........................................................................... 133
BIBILIOGRAPHY.................................................................................................. 135
APPENDIX............................................................................................................. 148
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vi
Chapter Page
Post Hoc Multivariate Analysis........................................................... 110
Summary............................................................................................... 118
5. CONCLUSIONS, IMPLICATIONS AND RECOMMENDATIONS.... 119
Health Risk Variables.......................................................................... 120
Self-reported Health Status........................................................ 120
Chronic Conditions..................................................................... 120
Hospital Admissions................................................................... 121
Regular Source of Care............................................................... 122
Quality of Care...................................................................................... 122
Doctor Visit.................................................................................. 123
Satisfaction................................................................................... 124
Afford Medication or Treatment................................................ 124
Preventive Procedures................................................................. 126
Summary of Findings.................................................................. 126
Policy Implications and Conclusions.................................................. 127
Study Limitations and Future Research.............................................. 132
Limitations................................................................................... 132
Future Research........................................................................... 133
BIBILIOGRAPHY................................................................................................... 135
APPENDIX.............................................................................................................. 148
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vii
LIST OF TABLES
Table Page
1. Process and Outcome Variables for the Study of Health Care
Systems................................................................................................. 38
2. Number of Interviews Completed by Language........................................ 70
3. Disposition of Telephone Contact Attempts............................................. 72
4. Health Risk and Quality of Care Variables Used in Analysis of Fee-
for-Service and Managed Care Organizations................................... 75
5. Descriptive Statistics 1997 LA County Health Survey: Comparison of
Los Angeles County Population Aged 65 Years and Over and the
Los Angels County Health Survey Medicare Population Aged 65
Years and Over...................................................................................... 85
6. Self-reported Health Status of Los Angeles County Seniors by Health
Insurance Type...................................................................................... 92
7. Prevalence of Chronic Health Conditions: Los Angeles County
Seniors Reporting Having Chronic Health Conditions by
Insurance Type...................................................................................... 93
8. Hospital Admissions for Los Angeles County Seniors by Health
Insurance Type...................................................................................... 95
9. Regular Source of Care: Whether or not Los Angeles County Seniors
Have a Regular Source of Health Care by Insurance Type............... 96
10. Regular Source of Care For Los Angeles County Seniors by Health
Insurance Type...................................................................................... 96
11. Patient Ease of Access to Their Healthcare Provider For Los Angeles
County Seniors by Health Insurance Type......................................... 99
12. Waiting Time To Get An Appointment For Los Angeles County
Seniors by Health Insurance Type....................................................... 100
13. Waiting Time At Appointments For Los Angeles County Seniors by
Health Insurance Type.......................................................................... 101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
viii
Table Page
14. Doctor Visit in the Past Year: Los Angeles County Seniors Who
Reported Seeing a Healthcare Provider in the Past Year.................. 102
15. Patient Satisfaction with Their Healthcare Provider For Los Angeles
County Seniors by Health Insurance Type......................................... 103
16. Financial Difficulty Accessing Health Services: Los Angeles County
Seniors Who Reported Needing Healthcare Service But Could Not
Afford It................................................................................................. 106
17. Preventative Test or Examinations Administered in the Past Two
Years: Los Angeles County Seniors Who Reported Receiving
Particular Examinations....................................................................... 107
18. Results of Hypothesis Test of No Significant Difference Between
FFS andMCO Medicare Beneficiaries in Los Angles County I l l
19. Odds Ratios for Health Risk Variables Controlling for Health Plan
and Insurance Coverage....................................................................... 113
20. Comparison of Total and Average Number of Chronic Conditions for
FFS vs. M CO ........................................................................................ 117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ix
LIST OF FIGURES
Figure Page
1. Distribution of Risk HMO Enrollment by State, 1997 ............................ 32
2. Availability of Medicare Risk Plans by Percent of Population with
Plans Available, 1995-1997................................................................ 33
3. Medicare and Commercial HMO Penetration in Selected States,
1996....................................................................................................... 35
4. Medicare Risk Plan Enrollment as Percentage of Beneficiaries,
Years 1987-2007.................................................................................. 36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X
ABSTRACT
Managed health care is increasingly important to provision of health care
services to Medicare and other public health insurance agencies. This study
compares risk selection and quality of care between Fee-for-Service (FFS) and
managed care organizations (MCO) in a mature managed health care market. A
random sample, population based instrument—the 1997 Los Angeles County Health
Survey—was used to explore these issues.
Risk selection was investigated using self-reported health status, chronic
conditions, hospital admission, and regular source of care. Quality of care was
investigated using access to services, waiting time for appointments, number of
doctors’ visits, and patient satisfaction.
Previous research had shown significant differences between FFS and MCO
for both risk factors and quality of care issues. The present study predicts no
significant difference for risk factors, quality of care, or access variables. This
prediction is based on the Welch’s (1984) theory of regression toward the mean. This
assumes that, within a mature managed care market, Medicare beneficiaries will
have aged into, as well as aged within the managed care system, thereby washing out
any initial effects of favorable risk selection from the 1980’s and early 1990’s.
Results of this study support the hypothesis of no significant differences
between MCO and FFS for health risk factors. No significant difference was found
for self reported health status, chronic conditions, hospital admissions and regular
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
source of care. Regarding the quality of care variables findings for only one variable,
patient satisfaction, showed significant differences between FFS and MCOs.
This study’s findings of few significant differences between FFS and MCOs
within the Medicare population for health risk and healthcare issues indicate that
Medicare’s foray into managed healthcare may be yielding the quality of care,
beneficiary access and cost savings that was intended. Significant issues for both
MCO and FFS systems regarding prescription medications and routine cancer
screening did emerge in the analysis of the findings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
CHAPTER 1
INTRODUCTION
Managed health care organizations have become increasingly important in
the provision of health care services to Medicare and other public health insurance
agencies. Concerns exist as to whether or not managed care organizations continue to
experience favorable selection in the health care characteristics of the populations
they insure. There are also concerns as to whether or not managed care organizations
are providing adequate levels of care and access to their Medicare beneficiaries. This
dissertation will investigate whether risk selection prevails in a mature managed
health care market. It will also address whether, within that market, managed care
organizations are providing levels and quality of health care services comparable to
those provided in the more traditional fee-for-service sector.
Nationally, managed care accounts for approximately 14.5% of the Medicare
population (Centers for Medicare & Medicaid Services, 2001). However, in Los
Angeles County managed care has been an important factor in the healthcare market
for many years, and covers approximately 38% of the Medicare beneficiaries in Los
Angeles County (Health Care Financing Administration, 2000). Therefore, Los
Angeles County presents a natural laboratory for investigating the relative perfor
mance of managed healthcare organizations, and the more traditional fee-for-service
healthcare system that still predominates in the Medicare market in much of the
United States. The 1997 Los Angeles County Health Survey provides a valuable and
effective tool for use in such an investigation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
The American health care system has been in a veritable revolution over the
past 2 decades. Advances in medical science have been staggering with the
progression of heart and other organ transplants, laser surgery, advances in pharma
cology leading to the practical eradication of infectious disease in the Western world.
However, the most profound change to come over the medical world in the United
States may not have been scientific, but organizational: the rise of the managed
health care organization. This form of health care delivery has been decidedly
marginal in presence and has often been considered second rate for almost 50 years.
Managed health care exploded in importance in the 1980s and grew to dominate the
American health care system in the 1990s. The mechanism that American industry
resorted to, to help stem the tide of rapidly rising employee health care benefit costs,
turned out to be a flood that is swamping the American health care system.
Managed health care organizations, health maintenance organizations, or
HMOs as they were commonly called for many years, comprised a small part of the
medical market place for generations. Until the 1980s, HMOs accounted for only
about 4% of the national health care insurance coverage (HCFA, 2000). Even in
California, where HMOs have had a relatively strong presence, they accounted for
only 17% of the health insurance market (HCFA, 2000). In response to persistent
double-digit inflation in employee health care benefit costs, American industry
turned in desperation to managed care as a mechanism to help control health
insurance inflation. Managed care began to grow rapidly so that, by the end of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1990s, managed care comprised more that 40% of all health insurance enrollees
nationally and more than 80% in California (HCFA, 1998).
Just as private enterprise turned to managed care to help control rising health
insurance expenses, public agencies are also turning to contracting with managed
care organizations (MCOs) to help control rising health care costs. An increasing
number of states are exploring ways to contract with MCOs to provide the health
care coverage for their Medicaid beneficiaries. Medicare and its administrative
branch, the Health Care Financing Administration (HCFA, 2000), have also sought
to access managed health care organizations as a means of controlling costs.
Congress authorized Medicare risk contracting with the Tax Equity and
Fiscal Responsibility Act (TEFRA) in 1982 with actual implementation of risk
contracting, beyond the demonstration phase, beginning in 1985. Although managed
care had been growing rapidly in the commercial sector through the late 1980s and
1990s, in the Medicare sector, it experienced relatively constrained growth until the
mid-1990s. At that time, managed care enrollment of Medicare beneficiaries began
to increase at a rate greater than 20% per year. As occurred when managed care
began to emerge in the commercial market, concerns arose about the ability of
HMOs to maintain the level of quality and services that generally had been expected
in the fee-for-service sector (Brown, Clement, Hill, Retchin, & Bergeron, 1993;
Rossiter, Langwell, Jan, & Rivnyak, 1989).
Besides certain trepidation that the overall quality of health care services
might be compromised under managed care, there exists a concern that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chronically ill and most vulnerable of the population, might not be well served in an
HMO environment (Miller & Luft, 1997). Some observers felt that the organizational
arrangements and cost benefits of HMOs were particularly suited to caring for the
chronically ill and financially vulnerable populations (Kane, 1998; Kramer, Fox, &
Morgenstem, 1992). Others were concerned that HMOs’ emphasis on cost
containment, reduced access to specialists, and restrictive bureaucratic structures,
might impair the access to needed services of the sickest or most vulnerable enrollees
(Ware, Bayliss, Rogers, Kosinski, & Tarlov, 1996).
Medicare
Medicare is the health insurance program of the American Social Security
system and is administered by Centers for Medicare & Medicaid Services (CMS)
formerly known as the Health Care Financial Administration (HCFA). The Medicare
program provides health insurance coverage for 39 million Americans aged 65 years
and older (HCFA, 2000). In 1965, when the Medicare program was enacted, only
half of the elderly population in the United States was covered by health insurance,
whereas, today 98% of the nation’s elderly are enrolled (HCFA, 1998).
Medicare is made up of two separate, complementary insurance programs.
Hospital Insurance, called Part A, covers inpatient hospital services, skilled nursing
facility care, home heath care, and hospice care for terminally ill beneficiaries.
Persons age 65 years and older who are eligible for any type of Social Security
benefits are automatically entitled to Part A, as are some disabled persons and
persons with end-stage renal disease. Most people do not pay a monthly Part A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
premium because they, or a spouse, has 40 or more quarters of Medicare covered
employment (CMS, 2002). Supplementary Medical insurance, called Part B, covers
physician services, durable medical equipment, outpatient medical services such as
lab tests, physical and occupational therapy, and ambulance transportation. During
the 1990s, Part B coverage was extended to certain preventive services, including
hepatitis B, flu, and pneumococcal pneumonia vaccines; Pap smears, and screening
mammography.
Part B, Medicare insurance is voluntary and open to all Part A enrollees and
most Americans aged 65 and older. Part B beneficiaries pay a monthly premium of
$54 with an annual deductible of $100 (CMS, 2002). Part B, Copayments are 20% of
allowed charges. Eighty-nine percent of beneficiaries have some form of supple
mental coverage that pays for part or all of Medicare copayments and other health
expenses. This coverage may be Medigap insurance— either self-purchased or
employer-sponsored— Medicaid or other governmental programs (Davis & Burner,
1995).
The Balanced Budget Act of 1997, among other changes in the program,
provided for an expanded set of options for the delivery of health care under
Medicare termed Medicare+Choice. Most beneficiaries enrolled in Medicare Parts A
and B can choose to receive the Medicare benefits through one of the
Medicare+Choice plans:
1. Coordinated care plans (such as health maintenance organizations,
provider-sponsor organizations, and preferred provider organizations);
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2. Medical Savings Account/High Deductible plans (through a
demonstration available to 390,000 beneficiaries); or
3. Private fee-for-service plans (HCFA, 1999).
Managed Care
HMOs and other managed health care organizations, are organized health
care systems that are responsible for both the financing and delivery of health care
service to an enrolled population for a prepaid, fixed fee (Wagner, 1989). The health
plans are “at risk” for providing the contracted services for the contracted premium.
To the extent the health plan keeps costs and expenses within the premiums received,
the health plan will be profitable. If costs exceed premiums received, the health plan
would lose money (Zarabozo, 1989). Therefore, an HMO can be viewed as a
combination of a health insurer and a health care delivery system. In a traditional
health insurance system, the insurance companies are responsible for reimbursing
covered individuals for their health care expenses. In the managed care system,
HMOs are responsible for providing the healthcare services to their members
through their affiliated providers (Wagner, 1989). Peter Kongstvedt (1989) in “The
Managed Health Care Handbook,” provides a succinct description of the managed
care system: managed care refers to any system that manages the delivery of health
care in such a way that the cost is controlled.
Managed care organizations (MCO) have grown in diversity of product
offerings and organizational forms as total enrollment and market share have
increased. The different forms affect (a) methods of reimbursement to providers
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(salary, capitation, or fee-for-service); (b) how financial incentives are used to
control provider behavior (such as risk pools for inpatient and specialty care); (c) use
of utilization management practices; the primary care providers practice settings
(solo or small group practice versus large group practice); (d) whether providers see
exclusively MCOs patients in their practice; and (e) whether the MCOs contracts
directly with the physician or contracts with an intermediary organization that
reimburses the physicians (Shortell & Hull, 1996).
Managed health care organizations vary widely from company-to-company.
The five most common models of managed care organizations are: (a) staff, (b)
group practice, (c) individual practice association (IPA), (d) network, and (e) direct
contract (Wagner, 1989). They can vary in organization by model type, chain
affiliation, enrollment size, for profit status, market areas, and other characteristics
(Shortell & Hull, 1996).
Originally, managed care organizations were predominately staff model
organizations. In a staff model HMO, most of the physicians and other personnel are
direct employees of the company. Physicians in staff model plans receive a salary
along with various performance incentives. Currently, the predominate MCO model
tends to be the independent practice associations (IPAs) and group model plans. An
IPA is an association of either individual physicians or small groups of physicians
who have associated their practices for the purpose of contracting with one or more
managed health care organizations (Wagner, 1989). Group model MCOs, are those
that contract with various physician groups to provide services to enrollees. IPA and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
group model plans typically have various financial incentives, based on utilization
rate, to encourage physicians to practice cost-effective care.
As markets have expanded and competition increased, there has also been a
blurring of distinctions among plan types. Shortell and Hull (1996) point out that
HMO categories such as staff model, group model, IPA, network model and mixed
model, continue to be used although they do not usefully delineate differences
among plans. In general, most forms of managed care combine at least some of the
following elements:
1. Attempts to influence or “manage” providers’ decisions about patient
care by various mechanisms.
2. Prepayment for care, whereby purchasers make capitated payments to
managed care firms and enrollees face very limited out-of-pocket costs.
3. Networks of preferred or required providers, which impose some
limitation on freedom of choice for the enrollee.
4. Payment terms with providers other than traditional fee-for-service
(FFS) payment such as discounted fees, capitation, salary, bonuses and withholds,
and other payment terms.
5. Primary care “gatekeepers” who coordinate care and control referrals
to specialty care (Shortell & Hull, 1996).
Although risk plans are located in all areas of the country, they are most
prevalent in the Pacific and Midwest regions. Distribution of enrollments is very
skewed—one-fourth of plans are located on the West Coast, and account for over
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
45% of total enrollments, reflecting large average size of these plans. The Medicare
risk program is dominated by HMOs in two market areas that together have 12 plans
and account for over half of total risk program enrollments.
Numerous attempts have been made, over the past 2 decades, to evaluate the
success of the Medicare risk program in reducing the costs of care for elderly
Americans. The findings of these studies are persistently clouded by the issue of
favorable selection of Medicare beneficiaries into the risk HMOs (Mello, 1999).
Medicare beneficiaries are allowed to choose whether or not to enroll in an HMO.
Beneficiaries who choose to join an HMO, may differ significantly in their health
status, health care needs, or propensity to consume care, from those who remain in a
traditional plan. If HMOs attract a disproportionate share of relatively healthy
Medicare population, then favorable selection will exist for the HMO. If the
beneficiaries who enroll, or remain enrolled in an HMO are on average, less healthy
then the HMO will experience adverse selection (Wilensky & Rossiter, 1986).
Purpose o f the Study
The aging of the population and contemporary changes in the present health
care system from traditional FFS to managed care raise issues related to access,
quality of care, and consumer satisfaction for the service populations. This is of
special concern for Medicare beneficiaries, who represent a large portion of the
senior population, especially since the Health Care Financing Administration has
made it a prime objective to transfer a substantial portion of the Medicare population
to such managed care organizations (HCFA, 1996).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
The specific aim of the present study is to utilize a random sample,
population-based instrument-the 1997 Los Angeles County Health Survey-to
explore these issues. Adequacy of service issues such as quality of care, access, and
consumer satisfaction for FFS systems and managed care organizations in a county-
wide population of Medicare beneficiaries will be compared and contrasted. Also to
be explored, are the relationships between a variety of sociodemographic variables
and such issues within the FFS and managed care systems. The objectives of the
present study are:
1. To determine if differences in risk selection persists between the FFS
and MCO Medicare populations of Los Angeles County.
2. To determine if the quality of care and/or satisfaction within the
Medicare population are comparable between the FFS and managed care systems.
3. To determine if the quality of care of the high risk/vulnerable
Medicare populations is comparable between the FFS and managed care systems.
With regard to the first question: It is hypothesized that the MCO and FFS
populations will not differ significantly on several health status indicators such as
self-perceived health status, presence of chronic conditions, and health care services
utilization.
Research Question A
Are the health risk factors similar between the Senior FFS and MCO
populations in Los Angeles County? It is hypothesized that:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
Hypothesis I. The rate of health risk factors is the same for the Senior FFS
and MCO populations in Los Angeles County.
a. It is hypothesized that the levels of self-reported health status are the
same for the Senior FFS and MCO populations in Los Angeles County.
b. It is hypothesized that the prevalence of chronic conditions are the
same for the Senior FFS and MCO populations in Los Angeles County.
c. It is hypothesized that the rates of hospital admissions are the same
for the Senior FFS and MCO populations in Los Angeles County.
d. It is hypothesized that the rates of having a regular source of care are
the same for the Senior FFS and MCO populations in Los Angeles County.
The rationale for this hypothesis is that the Medicare HMO population of Los
Angeles County will have participated in managed health care for extended periods,
and they will have more options in selecting acceptable physicians and programs
within the managed care system. Higher Medicare HMO penetration is generally
found in markets where HMOs have been operating for longer periods. Hence, the
Medicare HMO enrollees will have aged within in the system. In addition, members
of commercial HMOs who have reached retirement age within an HMO could
continue their care within the HMO— especially if they have serious or chronic health
conditions. Therefore, unless there is selective disenrollment, any initial favorable
selection in enrollment into HMOs will have dissipated.
In areas of high HMO penetration, the rate of physician contracting with
HMOs is greater than in low penetration areas. In these areas, beneficiaries are also
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
more likely to have several HMOs to choose from including open-panel HMOs.
Open-panel HMOs allow members to retain established relationships with
physicians. For these reasons, patients in high penetration areas are less likely to be
forced to sever their relationships with their existing physician in order to join, and
therefore, an important obstacle to Medicare beneficiaries in relatively poor health
joining HMOs is reduced (Mello, 1999).
The second objective, regarding quality of care and satisfaction will be
addressed on two fronts. The first is overall satisfaction of Medicare beneficiaries
with their health plans. It is hypothesized that in a mature managed care market such
as Los Angeles County, overall satisfaction between HMO and FFS enrollees will be
comparable. In many of the early studies, FFS beneficiaries consistently rated their
level of satisfaction higher than managed care enrollees (Rossiter et al., 1989).
However, the majority of these studies covered beneficiaries who had recently
converted from FFS to HMO status. Patients who had been accustomed to relatively
unlimited access to their physicians, specialists and hospital care, might consider any
control or restrictions by “gatekeepers” or HMO “bureaucrats” as less satisfactory
regardless of the competency of the treatment provided. After enrollees have been
part of the HMO system for longer periods, they would “learn the ropes” and be able
to negotiate the system much more effectively and with greater satisfaction.
Next, the question of quality of care will be addressed by assessing whether
plan enrollees have comparable access to various aspects of health care services.
These include factors such as doctor visits in the past year, a physical examination
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
and various screening tests, as well as access to certain health care services such as
eyeglasses, prescription medications, and immunizations.
There have been concerns that more populations that are vulnerable or those
suffering from chronic conditions, may not be adequately served in the managed care
system (Miller & Luft, 1997). On the other hand, there are those who contend that
the more organized system of managed care and the “gatekeeper” function of the
primary care physician, are better suited to provide more comprehensive care to
people suffering from difficult, multiple or chronic health care problems (Brown et
al., 1993; Kramer et al., 1992). Finally, there are observers who feel that both
systems fail to adequately care for those individuals who are least able to contend
with the health care system due to their health, economic or sociological conditions
(Light, 1998). The third phase of this study will investigate whether the services
provided to the high risk/vulnerable population by each system are comparable.
Research Question B
Is the quality of care comparable between the FFS and MCO systems in Los
Angeles County? It is hypothesized that:
Hypothesis II. The quality of care is the same for Senior FFS and MCO
populations in Los Angles County.
a. It is hypothesized that the rate of people who report having difficulty
with access to health care services is the same for the Senior FFS and MCO
populations in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
b. It is hypothesized that the waiting times are the same for the Senior
FFS and MCO populations in Los Angeles County.
i. “Wait times” for an appointment are the same for the Senior
FFS and MCO populations in Los Angeles County.
ii. “Office wait” times are the same for the Senior FFS and MCO
populations in Los Angeles County.
c. It is hypothesized that the rate of seniors who have seen a doctor or
other health care provider in the past year is the same for FFS and MCO populations
in Los Angeles County.
d. It is hypothesized that the level of satisfaction with the care received
from their health care providers is the same for the FFS and MCO populations in Los
Angeles County.
e. It is hypothesized that the rate that people report that they could not
afford certain medications or treatments is the same for the Senior FFS and MCO
populations in Los Angeles County.
i. The rate that people report that they could not afford
prescription medication is the same for the Senior FFS and MCO populations
in Los Angeles County.
ii. The rate that people report that they could not afford mental
health care or counseling is the same for the Senior FFS and MCO
populations in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii. The rate that people report that they could not afford dental
care is the same for the Senior FFS and MCO populations in Los Angeles
County.
iv. The rate that people report that they could not afford
eyeglasses is the same for the Senior FFS and MCO populations in Los
Angeles County.
The rate that people report that they could not afford seeing a doctor is the
same for the Senior FFS and MCO populations in Los Angeles County.
f. It is hypothesized that the use of preventive services is the same for
the Senior FFS and MCO populations in Los Angeles County.
i. The rate for receiving a colo-rectal exam is the same for the
Senior FFS and MCO populations in Los Angeles County.
ii. The rate for receiving a blood test for HIV/AIDS is the same
for the Senior FFS and MCO populations in Los Angeles County.
iii. The rate among male senior respondents for receiving a
testicular exam is the same for the Senior FFS and MCO populations in Los
Angeles County.
iv. The rate among male Senior respondents for receiving a
prostate exam is the same for the Senior FFS and MCO populations in Los
Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
v. The rate among female senior respondents for receiving a pap
smear is the same for the Senior FFS and MCO populations in Los Angeles
County.
vi. The rate among female senior respondents for receiving a
physical breast exam is the same for the Senior FFS and MCO populations in
Los Angeles County.
vii. The rate among female senior respondents for receiving a
mammogram is the same for the Senior FFS and MCO populations in Los
> Angeles County.
These hypotheses are investigated using data from the 1997 Los Angeles
County Health Survey (LACHS). The LACHS is a randomized survey of more than
8,000 residents of Los Angeles County regarding their health care and health
insurance issues.
Chapter 2 of this dissertation presents the research and historical context of
the present study. It describes the Medicare risk program, the growth of commercial
and public agency risk contracting, the research investigating these programs, the
issues of risk selection and regression toward the mean in health insurance.
Chapter 3 will describe the methodology used in the present study. The 1997
Los Angeles County Health Survey will be described. The strengths and weakness of
the survey will be considered. The statistical methods used to evaluate the survey
data will be described.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 4 will describe the findings of the study that pertain to comparison of
Medicare FFS and HMO health care organizations.
Chapter 5 will present the conclusions and policy implications of the present
findings. Limitations of the study and suggestions for future research will also be
presented in Chapter 5.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
LITERATURE REVIEW
The emergence of managed care and market-based contracting have
generated rapid, substantial changes in the American health care system. Private
health care purchasers who were seeking means to control spiraling health care costs
were the driving force behind the latest reforms in the nation’s health care delivery
system. In addition, the two largest public health insurance programs, Medicare and
Medicaid, are also turning to managed care from traditional FFS systems to help
curtail their health care expenditures (Brown et al., 1993; Lamphere, Neuman,
Langwell, & Sherman, 1997).
Early Literature
Much of the early literature regarding managed health care is based on
samples from areas or populations, which had very limited managed care
participation in the health care market at the time of the studies. Many of the studies
were in areas or at times of transition in the health care market. Indeed, some of the
studies were attempts to compare costs or performance from just before and just after
the implementation of risk plans within a given region (Edgers & Prihoda, 1982;
McCombs, Kasper, & Riley, 1990). Some of the most important studies were from
HMO Demonstration projects in which the managed care plans had been by
definition limited in time and scope (Retchin, Clement, Rossiter, Brown, & Nelson,
1992; Riley, Rabey, & Kasper, 1989).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Much of that written during early research regarding HMOs, came from a
time when managed care was an emerging phenomenon. The tenor of the earlier
research was a judgmental approach exploring concerns about the quality of care
provided by HMOs (Ware, Rogers, Davies, Goldberg, Brook, & Keeler, 1986;
Rossiter et al., 1989). Considering that managed care is now the predominate health
care delivery system in many commercial markets and has a broad representation in
a wide range of Medicare markets (CMS, 2001), there is a need for research which
takes these factors into consideration.
Research based on more recent studies has, indeed, presented a more
ecumenical approach to the comparison of FFS and MCO providers, seeking to
determine which provides the most appropriate care based on effectiveness,
satisfaction, or outcomes (Carrasquillo, Lantigua, & Shea, 2001; Leach, Yip, Myrtle,
& Wilber, 2001, Wilber & Yip, 2000). This study will endeavor to extend the
literature into a consideration of the relative performances of FFS and MCO
paradigms in a mature, relatively broadly based, managed care market, using a local,
population based survey.
The Tax Equity and Fiscal Responsibility Act: In 1982, the Tax Equity and
Fiscal Responsibility Act (TEFRA) was passed. The primary goal of TEFRA, which
established Medicare risk contracting, was to reduce costs of health care for
Medicare beneficiaries while maintaining the quality of services received. At the
time of the implementation of TEFRA, experience with and research into the senior
population, those over 65 years of age, in the managed care market was very limited.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
As participation in managed care began to expand, research into several areas of
particular concern to policymakers and participants began to appear (Lichtenstein,
Thomas, Adams-Watson, Lepokowski, & Simone, 1991). Researchers, policy
makers, and provider organizations raised questions about quality of care, access to
care, and costs. Over the next 15 years, numerous articles addressing these issues
were published. However, even as substantial research was being presented, the
whole health care system in the United States has been in flux. Managed healthcare
organizations have grown from a marginal segment of the American system in the
1980s to the predominate mode of health insurance by the end of the century.
Although the transition by the senior population to the managed care paradigm has
lagged behind the general commercial population, the changes once they began, were
dramatic. These changes in the relationship between the senior population and the
health care organizations that serve them, offer new challenges and opportunities for
research in the appropriateness and effectiveness of those arrangements and their
outcomes.
Since July of 2001, Medicare has been administered by the federal
government through The Centers for Medicare & Medicaid Services (CMS) formerly
the Health Care Financing Administration (HCFA). The long-term goal in
establishing the Medicare risk program was to shift more Medicare beneficiaries into
managed care plans, which might be capable of restraining the growth of Medicare
costs because they could restrict use of unnecessary medical care (Brown et al.,
1993; Ellis, Pope, Iezzoni, Ayanian, Bates & Burstein, 1996; McCombs, Kasper, &
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Riley 1990; Pai & Clement, 1999). These expectations could not be realized without
broad MCO participation (Pai & Clement, 1999).
Although section 1876 of the Social Security Amendments of 1972
empowered Medicare to sign risk-based contracts with HMOs, only one HMO, and
Group Health Cooperative of Puget Sound had a risk based contract with HCFA
through 1980. In the early 1980s, HCFA initiated several demonstration projects in
order to encourage HMOs to participate in Medicare risk-contracting. However, it
took the Tax Equity and Fiscal Responsibility Act (TEFRA) of 1982 to make it more
attractive for HMOs to participate in the Medicare risk-contracting program. In April
1984, the Medicare risk-contracting program that was legislated under TEFRA
became operational and had a dramatic effect on the growth of enrollment in
Medicare risk HMOs (Hellinger, 1987). The primary goal of Medicare risk
contracting under TEFRA, was to reduce costs of health care for Medicare
beneficiaries while maintaining the quality of services received (Brown et al., 1993).
Under TEFRA, a participating health maintenance organization (HMO) or
competing medical plan (CMP), receives a capitation payment amounting to 95% of
the adjusted average per capita cost (AAPCC) of serving FFS beneficiaries in the
same market counties. If the CMP earns a higher profit on Medicare beneficiaries
than on its commercial enrollees, the CMP has to either supplement benefits for
enrollees or return the excess profits to Medicare. All CMPs that show profits have
elected to supplement benefits rather than return funds to Medicare (CMS, 2002).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Commercial HMOs
Since HCFA began risk contracting, the health care marketplace has changed
substantially. Organizational change has come about due to the creation and
transformation of many new provider networks and organizations; the rapid growth
of network and mixed-model managed care organizations, including individual
practice associations (IPAs), preferred provider organizations (PPOs) and point-of-
service plans. Financial pressures on health care providers have increased due to
reduction in indemnity plan coverage; changes in Medicare payment methods; and
generally, a shift form inpatient hospital to outpatient ambulatory care settings and
increased market competition among health plans and among providers (Miller &
Luft, 1994).
Concern about the quality of care that patients, especially the elderly
population, would receive under managed care has been an important focus of health
care research. There are concerns that HMOs, in responding to financial incentives
rewarding more efficient health care, might provide overly restrictive services
leading to lower quality of care. Much of the early research on HMOs addressed this
issue (Miller & Luft, 1997).
The Rand Health Insurance Experiment (HIE) is a seminal study of
alternative insurance plans. The Rand HIE was a randomized field trial spanning
1971 to 1982, with data collection and analysis continuing through the 1980s
(Newhouse, 1993). The HIE included a comparison between people under age 65
years enrolled in FFS plans with various copayments and those enrolled in an HMO.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
The HIE provides strong evidence that, in the absence of selection bias, HMOs can
achieve cost savings (Manning, Leibowitz, Goldberg, Rogers, & Newhouse, 1984).
However, there are several factors that constrain the findings of the experiment: (a)
the study included only one HMO at one site. Cost savings may not be generalizable
to other HMOs; and (b) of the people contacted for the study, 29% refused to join the
experiment. It was determined that the nonparticipants were similar to the
participants. Yet, if the two groups differed systematically, in a manner undetected
by the experimenters, then the results could be biased (Hellinger, 1987).
Hospital Utilization
Utilization of hospital services has changed significantly in both the managed
care and FFS sectors, especially since HCFA implemented reimbursement by
diagnosis related groups (DRGs) in 1985 (Miller & Luft, 1994).
In an early literature analysis of comparisons between HMO and FFS hospital
utilization, Luft (1981) found consistently lower (15%-40%) HMO admission rates,
no consistent differences in hospital length of stay (LOS), and many fewer hospital
days per HMO enrollee.
In a 1994 review of 54 studies of managed care and indemnity plan
performance, Miller and Luft (1994) found that HMO plans generally had lower
hospital admission rates. The Medical Outcomes Study (MOS) indicated that 26% to
37% fewer HMO enrollees were hospitalized compared to indemnity plan patients
(Miller & Luft, 1994). Other studies, including the Medicare TEFRA Evaluation,
showed small and statistically insignificant difference in admission rates between
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
HMO and indemnity plans (Miller & Luft, 1994). The Luft review indicated that
HMO plans generally had from 1% to 20% shorter hospital lengths of stay than
indemnity plans. However, the HMO plans had widely varying reduction in hospital
length of stay, and in hospital days per enrollee compared with indemnity plans.
Physician Office Visits
Miller and Luft (1994) note that although earlier studies tended to show
lower physician service use in HMO compared to FFS, 9 out of 10 of the more recent
studies showed higher rates or little difference in HMO plan office visits per enrollee
compared with indemnity plans. Miller and Luft note that both MOS and Medicare
TEFRA, results showed that higher numbers of physician office visits generally
accompanied lower hospital use. Other, more recent, studies regarding mental health
services showed significantly fewer mental health visits in HMOs compared with
indemnity plans (Ware et al., 1996).
Prevention and Health Promotion
Miller and Luft (1996) found that managed care patients consistently
received more preventive tests, procedures, and examinations (such as cancer and
hypertension screening tests and breast, pelvic, rectal, and general physical
examinations) or health promotion activities than did indemnity plan enrollees.
The findings of Miller and Luft (1996) are supported by other findings.
Potosky, Breen, Graubard and Parsons (1998), evaluated data on approximately
9,400 people using the 1992 National Health Interview Survey to estimate the
proportions of persons screened according to type and extent of coverage adjusted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
for socioeconomic, demographic, and health status characteristics. They found that
for all six screening tests evaluated, managed care enrollees were approximately
10% more likely to be screened than persons enrolled in private fee-for-service plans
(Potosky et al., 1998).
Quality o f Care
The preponderance of studies reviewed by Miller and Luft (1997)-14 of 17-
indicated that HMO and indemnity plans provided enrollees with roughly
comparable quality of care, according to process or outcomes measures. This series
included six National Medicare Competition Evaluation studies that determined that
quality of care, was approximately equal for FFS and HMO beneficiaries regarding
treatment processes and outcomes for patients being treated for heart failure, colo
rectal cancer, diabetes, and hypertension. Both FFS and HMO patients exhibited
similar outcomes and treatment processes.
Research on outcomes among more vulnerable populations indicates some
potential problems for HMOs. Lurie, Moscovice, Christianson, and Popkin (1992),
found that, among patients with chronic mental illness, health status declined more
over time in MCOs than in FFS. Among the elderly with chronic illnesses, Ware and
others (1996) reported a greater decline in health status in MCOs than in FFS.
In a 4-year longitudinal study of approximately 2,200 patients with chronic
conditions such as hypertension, diabetes, recent acute myocardial infarction,
congestive heart failure or depression, it was found that physical health outcomes
were more favorable for nonelderly patients than elderly or chronically ill patients in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
HMOs (Ware et al., 1996). However, the elderly treated in HMOs were nearly twice
as likely to decline in physical health than those treated in FFS. In the Ware and
others’ (1996) study, poor patients with poor initial health were more likely to
experience a decline in health in HMOs than in FFS. Ware and others (1996)
conclude that although on average patients fare as well or better in HMOs as FFS
systems, the most vulnerable populations-those in poverty or the elderly with certain
chronic conditions— may not fare as well in HMOs as in FFS.
Some studies report that HMOs exhibit more attentiveness to process of care
such as appropriate use of procedures than FFS settings (Preston & Retchin, 1991;
Retchin & Brown, 1990). A number of studies report no difference between MCOs
and FFS in process of care (Wholey, Bums, & Lavizzo-Mourey, 1998).
Wholey and others (1998) concluded that, in general, access to primary care
providers (PCPs) is better in MCOs than in FFS, perhaps due to better coordination
within the managed care system. Although access to specialists and to hospitals is
lower and more difficult in MCOs than FFS, Wholey and others did not find process
of care, given access, to be different between MCOs and FFS. In general, MCOs
enrollees are more satisfied with the financial aspects of a health plan, and are less
satisfied with other aspects of health plan organization (Wholey et al., 1998).
In a study of rates of death, hospital readmission, and post-admission
complications among HMO and FFS, no differences in outcomes were noted (Brown
et al., 1993). In a study of stroke patients, it was found that HMOs provided quality
of care equal to that furnished in FFS settings (Retchin, Clement, & Brown, 1994).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
In a more recent study Potosky and others (1999), using multivariate regression
models, compared treatment patterns and the 10-year survival rate of prostate cancer
patients in two large, nonprofit group/staff HMOs to FFS settings. They found that
the 10-year overall survival for the HMO patients was equivalent to the survival
among patients receiving care in the FFS setting, even though there were marked
treatment differences for clinically localized prostate cancer.
There may have been some concerns into the early 1990s regarding the
quality of care provided by MCOs. However, the research outlined here demon
strates that, with few exceptions, MCOs provided care comparable, if not in some
cases superior, to that found in the traditional FFS system.
Enrollee Satisfaction
The findings on enrollee satisfaction are somewhat mixed. HMO enrollees
did not rate the perceived quality of care and patient-physician interactions as highly
as did FFS enrollees. Patients in MCOs are more satisfied with financial
arrangements of their health care although they are less satisfied with access to
specialist, hospitals, and prescriptions; with obtaining information; and with quality
of care (Clement, Retchin, Brown, & Stegall, 1994). However, in 4 out of 5 studies
reviewed by Miller and Luft (1994), HMO plan enrollees were very satisfied, or
rated highly, the financial aspects of their health plans. In addition to these findings,
in 4 out of 5 of these same studies, HMO plan enrollees were very satisfied with, or
rated highly, most aspects of their care.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Costs
In addition to reducing hospital utilization and access, it has been noted that
HMOs reduce cost by limiting the use of expensive services that may have less
costly alternatives. In 18 of 20 comparisons from nine studies, HMO plans used an
average of 22% fewer procedures, tests, or treatments that were expensive and/or had
less costly alternative interventions (Miller & Luft, 1997). Because HMOs also
provided more comprehensive coverage than did indemnity plans, the findings
suggest that HMOs provide care at lower cost than do indemnity plans (Miller &
Luft, 1997).
Medicare HMOs
Across the United States, managed health care plans have increased as the
health insurance plan for many employee benefit packages. Managed care plans have
been making significant, though more limited inroads into the Medicare system. The
Balanced Budget Act (BBA) of 1997 established the Medicare+Choice plan, which
allows Medicare beneficiaries, the option of enrolling in a variety of private plans in
addition to HMOs. Medicare MCOs enrollment grew rapidly during the period from
1993 to 1999. The number of Medicare beneficiaries enrolled in managed care
increased from 1.8 million to 6.3 million from 1993 to 1999 (Cook & McCoy, 2001).
In 1999, 16% of Medicare beneficiaries were enrolled in managed care plans.
However, beginning in 1999, enrollment in MCOs began to decline. By 2001,
managed care enrollment had declined to 5.6 million or 14% of Medicare
beneficiaries (Cook & McCoy, 2001). This drop in enrollment is due largely to plan
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
withdrawals and service area reduction in nonurban areas of the country that have
occurred over the past 3 years (Gluck & Hanson, 2001).
However, in some areas such as Southern California, MCOs have garnered
and maintained a much more substantial portion of the market. Enrollment in
managed care risk plans has remained relatively stable at over the past several years.
In 2001, 35% of Medicare beneficiaries were enrolled in Medicare+Choice risk plans
(Cook & McCoy, 2001). In Southern California, MCOs accounted for more than
35% of the Medicare sector at the time of the LACHS (HCFA, 1996).
Some questions arise as to the future performance of managed care organizations,
particularly as competition between the plans increases, and FFS programs adjust
their marketing strategies. Concerns regarding quality of care and access to specialty
services increase as HMOs come under increasing cost pressure, as reimbursement
prices decrease and membership enrollment becomes less selective (Oberlander,
1997).
Growth o f Risk Contracts
An important factor in the growth of HMOs in Medicare, is the richness of the
benefits packages and reduction of costs that the HMO can offer beneficiaries.
Medicare HMOs tend to provide richer benefits and charge lower costs than
traditional Medicare. Many plans offer additional benefits such as prescription drugs,
eyeglasses, preventive health care services, and often, no additional charges beyond
the monthly Medicare Part B premium (HCFA, 1996; Lamphere et al., 1997).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Although the concept of managed care was not well known when Medicare
was enacted, prepaid health plans were included in the reimbursement structure.
Managed care growth in the private sector began in the 1970s, while growth in the
Medicare managed care lagged (Davis & Burner, 1995).
It was in 1982 when TEFRA was implemented, that Medicare developed a
full-risk managed care capitated payment option based on a prospective payment
methodology, and Medicare health maintenance organization (HMO) enrollment
began to increase (Davis & Burner, 1995; HCFA, 1996).
From its inception in 1984 until the mid-1990s, the response by Medicare
beneficiaries to Medicare HMOs was rather modest. Enrollment as percentage of
eligible beneficiaries increased from 2.6% in 1987 to 8.2% in 1995, reaching 10%
for the first time in 1996 (Lamphere et al., 1997). In 1994, enrollment in Medicare.
HMOs began to surge ahead. From 1992 to 1994 enrollment increased 22%. Then
between 1994 and 1996 enrollment increased more than 40% (Lamphere et al.,
1997). These growth rates represent enrollment of 440,000 Medicare beneficiaries in
1985 to 2.2 million in 1994 and with a surge to 5.2 million beneficiaries by 1997
(HCFA, 1998).
The number of risk contracting HMOs fluctuated in the 1980s, rising from a
few demonstration projects to 161 contracts in 1987 and dropping to 96 in 1990. The
number of HMOs available in the 1990s grew rapidly. Available plans increased
from 96 plans in 1990 to 346 in 1998. However, beginning in 1999 and continuing
for each of the following 2 years, there was a decline in the number of plans
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
participating in Medicare risk contracting. By the fall of 2001, there were only 179
MCOs participating in Medicare risk plans (Gluck & Hanson, 2001).
Medicare risk contracting is not distributed evenly across the country. As of
December 1997, California, Florida, and Pennsylvania had 49% of total risk
enrollment, while only 23% of all Medicare beneficiaries live in these states (Figure
1).
In 1997, nearly 40% of all Medicare HMO risk-enrollees were concentrated
in 15 counties (HFCA, 1998). In 1993, there were 24 states with no HMO risk plan
available to Medicare beneficiaries. By 1997, only seven states had no HMO risk-
plans available (HCFA, 1998). In 1995, only 55% of beneficiaries had HMO
available to them, but by 1997, 67% of Medicare beneficiaries resided in areas with
Medicare HMOs available (HCFA, 1998) (Figure 2). By 1999 up to 72% of
Medicare, beneficiaries had a Medicare HMO available in their area. However, from
1999 to 2001 the share of Medicare beneficiaries with an HMO option declined to
64% (Gluck, 2001).
An important factor influencing Medicare HMO enrollment in a county or
state is a history of strong HMO penetration in that area’s general health care market.
(Lamphere et al., 1997; Welch, 1996). According to a 1995 General Accounting
Office report to Congress, the states with the highest concentration of residents
participating in HMOs in the general population also have the highest concentrations
of Medicare beneficiaries participating in Medicare managed care programs,
particularly in Arizona, Oregon and California (Inglehart, 1996). In 1997, California
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
33 States with <1%
7%
12 Other States (+DC)
27%
New Yoik
7%
Texas
5%
( alifomia
27%
Florida
13%
Arizona
5%
Pennsylvania
9%
Figure 1. Distribution of Risk HMO Enrollment by State, 1997
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
45%
H June 1995 ■ June 1996 □ June 1997
No Plans Available O ne Plan Two to Four Five or More Plans Available
Figure 2. Availability of Medicare Plans by Percent of Population with Plans Available, 1995-1997.
U)
OJ
34
led the nation in HMO penetration of the commercial health care market at 77%, and
the Medicare market at 38% HMO penetration (HCFA, 1998) (Figure 3).
The Congressional Budget Office (CBO) had predicted that growth in the
Medicare HMO market would be at least 20% per year (as cited in Lamphere et al.,
1997). Medicare managed care enrollment was projected to be 21% of Medicare
beneficiaries by 2001 and 34% by the year 2007 (CBO, 1995) (Figure 4). However, in
1999 MCO enrollment leveled off and actually declined to 14% of Medicare
beneficiaries by the year 2001 (Gluck & Hanson, 2001).
Early concerns raised questions about managed care organization’s ability to
provide satisfactory care or adequate levels of service to the Medicare beneficiaries
(Rossiter et al., 1989). Much of early research comparing managed care organizations
with fee-for-service insurance programs found no distinct over-all advantage of one
program over the other. Most studies found mixed results for each program (Miller &
Luft, 1997; Newhouse, 1993; Retchin et al., 1992).
The Behavioral Model o f Health Services Utilization
The Behavioral Model of Health Services Utilization developed by Andersen
and others (Aday & Andersen, 1974; Andersen & Newman, 1973; Aday, Sellers &
Andersen, 1981) has been widely used to evaluate access, utilization, and satisfaction
of health care services (Burnette & Mui, 1999). Wolinsky (1994) describes several
advantages to using the behavioral model for examining use of health services by
older adults: (a) it is the prevailing conceptual framework for investigating service use,
especially for determining whether or not access to and use of health service are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
■ Medicare Risk
1 1 Commercial
1 51 %
Percent Market Penetration
Figure 3. Medicare and Commercial HMO Penetration in Selected States, 1996
Note. Commercial HMO penetration exceeds Medicare HMO penetration.
U >
LO
36
C M
CO
C M
o >
C M
o
h -
C M
00
M -
C M
0 >
C M
C M
O
C M
O
h -
T"
o
t J-
o
C M
00
O
h - C O O O x - C M C O T f l D C D h - C O C T )
00 00 CO O) O C j ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 )
0 ) 0 0 ) 0 ) 0 ( 7 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 )
o
T—
CM CO M ’ i n CO N -
o o O O O o o O
o o O O O o o O
CM CM CM CM CM CM CM CM
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4 . Medicare risk p lan enrollment a s percentage o f beneficiaries, 1987-2007.
37
equitable; (b) it is eclectic in its interdisciplinary approach; and (c) it is amenable for
framing secondary analysis.
System
The delivery system has two major components— resources and organization.
Resources are the labor and capital devoted to health care. Organization refers to how
the resources are utilized within the health care system. How the population gains
entry into the healthcare system, the subsequent treatment process, and how satisfied
the population is, are each functions of the organization (Aday & Andersen, 1974).
Population
Aday and Andersen (1974) present service use by a given population as a
function of three sets of factors: (a) predisposing, (b) enabling, and (c) need.
Predisposing factors are the sociodemographic characteristics and health-related
attitudes that exist before a given illness episode. Predisposing factors include (a) age,
(b) race, (c) gender, and (d) genetic factors. Enabling factors are social and economic
resources that facilitate or impede service use. The enabling factors include income,
health insurance, source of care and geographical factors. Need factors are those
illness and/or impairment-related conditions for which services are sought. The need
may be self-perceived health status, acute illness or chronic conditions. The need can
be perceived by the consumer or by the health care professionals (Aday & Andersen,
1974). (Table 1).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
Table 1
Process and Outcome Variables for the Study o f Health Care Systems
Process Outcome
System Population Utilization Satisfaction
Access Health status Hospital admit Afford
medication
Regular
source of
care
Chronic conditions Last doctor visit Satisfaction
Appointment Age/gender
Race/ethnicity
Routine care
Adapted from Aday, L. A. & Andersen, R., A. (1974, Fall). Framework for the study
of access to medical care, Health Services Research.
Utilization
Measures of utilization entail the type and purpose of service and the locale of
services received within the system for an individual in the course of treatment or an
illness. Type of service refers to services such as hospital, physician, dentist, or
emergency care as well as the site the care was rendered, including facilities such as
clinic, inpatient hospital, or medical office (Aday & Andersen, 1974) (Table 1).
Satisfaction
Consumer satisfaction entails such variables as the percentage of the study
population who were satisfied or dissatisfied with convenience, cost, coordination
courtesy, medial information, and overall quality of care. It can also include the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
percentages of patients who wanted medical care but did not get it (Aday & Andersen,
1974) (Table 1).
Access and Managed Care
Because of the increased bureaucracy involved in using an HMO and the
emphasis on controlling costs, whether or not elderly patients have adequate access to
services is an important consideration. Most studies find that access to necessary
services is comparable between FFS and HMOs for Medicare patients, and in some
cases, the HMO patients avail themselves of more routine and preventative services
than do the FFS beneficiaries (Potosky et al., 1998). Brown and his colleagues found
that there were no consistent patterns of differences between HMO and FFS patients in
the likelihood of receiving medical attention for three common, chronic problems of
elderly people— joint pain, urinary incontinence, and chest pain (Brown et al., 1993).
Riley, Potosky, Klabunde, Warren, and Callard-Barbash and others (1999),
conducted a study that linked Medicare enrollment records to the National Cancer
Institute’s Surveillance, Epidemiology, and End Results (SEER) program records. The
study evaluated women aged 65 years or older, residing in 11 geographic areas, newly
diagnosed with breast cancer between 1988 and 1993. Riley and others found that,
although results varied widely by region and between HMOs, the HMO enrollees were
less likely to have breast cancer diagnosed at late stages than the FFS patients. Among
those diagnosed as early-stage cases, the percentages undergoing breast conservation
surgery were similar in HMO and FFS settings overall. In addition, Riley and others
concluded that elderly women enrolled in HMOs, did not have systematic access
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
problems receiving radiation therapy following breast surgery compared to women in
the FFS settings.
A more recent study, using 1996 data from the Medicare Beneficiary Survey,
found that HMO membership slightly increased the probability of a Medicare
beneficiary using routine physician services (Mello, 1999).
Satisfaction and Managed Care
Another important consideration in the emergence of Medicare HMOs is
whether Medicare beneficiaries were satisfied with risk plans (Rossiter et al., 1989). It
was found that although HMO enrollees were significantly less likely than FFS
patients to rate their care as excellent, more than 90% of both HMO enrollees and FFS
beneficiaries rated various dimensions of their care as good or excellent (Brown et al.,
1993).
Clement and others (1994) concluded that overall satisfaction with services
provided through HMOs was high. They found that HMO enrollees were not less
satisfied than nonenrollees with many aspects of the quality of care and medical
service they received.
Historically, most Americans have expressed satisfaction with their health
insurance plans, with Medicare enrollees demonstrating the highest levels of
satisfaction (Adler, 1995). There are indications that the disadvantaged populations,
the ones most in need, may be less likely to rate their satisfaction highly (Lee &
Kasper, 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
Outcomes and Managed Care
Retchin, Brown, Yeh, Chu, and Moreno (1997), compared treatment and
outcomes of more than 800 Medicare stroke patients treated in HMOs and FFS. They
found similar survival patterns between comparable patients in FFS and HMO
settings. They note that patients in Medicare HMOs were more likely to be discharged
to nursing homes, and less likely to go to rehabilitation facilities following their
hospitalization for stroke.
Wholey and others (1998), in a survey review of more than 20 studies
presented in the 1990s, examined the performance of managed care organizations in
delivering health care, particularly primary health care, to the elderly and the
chronically ill. They found that the research on access to care shows that MCOs have
higher rates of physician office visits and lower rates of hospitalization, drug
prescriptions, and referral to specialists than FFS.
Clement and others (1994) reported on a 1990 telephone survey of Medicare
beneficiaries, reported joint pain or chest pain during the previous 12 months. Clement
and others found, after controlling for demographic factors, that although there were
differences in treatment patterns and use of specialty services, there were no
differences in complete elimination of symptoms between those enrolled in an HMO
versus FFS.
Yelin, Criswell and Feigenbaum (1996), in an 11 year longitudinal study of
patients with rheumatoid arthritis, found that HMO and FFS patients did not differ on
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
quantities of health care services received or outcomes experienced on either an
annual or long-term basis.
Brown and others (1993) compared the quality of care received by Medicare
beneficiaries in HMO versus FFS settings by comparing both outpatient services for
three chronic conditions (joint pain, urinary incontinence, and recurring chest pain)
and the services received by patients hospitalized for colon cancer or stroke. Brown
and others (1993) found no difference in outcomes between HMOs and FFS on
hospital readmission, post admission complications, or the rates of death.
Two Medicare TEFRA Evaluation studies on patients with colon cancer,
cerebrovascular accident, or chronic problems (joint pain, urinary incontinence, and
chest pain), also indicated comparable quality of care, although a few differences
suggest that HMO plan enrollees may receive less adequate care in some situations
(Miller & Luft, 1994).
Miller and Luft’s (1994) review of 54 studies report two Medical Outcomes
Study (MOS) observations on beneficiaries with mental health problems produced the
only results that were significantly unfavorable for managed care plans (Wells et al.,
1989; Miller & Luft, 1994; Rogers, Wells, Meredith, Sturm, & Bumam, 1993). In
similar findings in another MOS study, Ware and others (1996) found that outcomes
were less favorable for the elderly and those in poverty in HMO systems than in FFS
systems.
In a study of older patients with acute myocardial infarction, HMO patients
had similar mortality after adjustment of differences in level of sickness at admission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
It was also noted that HMO patients had greater compliance with process of care
criteria for Medicare patients compared to a representative group of FFS Medicare
patients treated in urban hospitals (Carlisle, Siu, & Keller, 1992).
A more recent study by Kramer and others (2000), indicates that there might
be some problems with chronic care provided in HMO delivery systems. In a cohort
study of Medicare patients that had suffered a stroke and received rehabilitation
treatment, Kramer and others report that FFS patients had a higher likelihood of
community residence, and a lower likelihood of nursing home residence than HMO
patients. The study analyzed post-stroke treatment and rehabilitation of 429 stroke
patients from 11 health plans (6 HMOs and 5 FFS delivery systems) from across the
country. After adjusting for co-morbidities and several risk factors for nursing home
placement, Kramer and others found that the likelihood of nursing home placement
was 2.10 times greater in HMO setting at 6 months and 2.4 times greater at 12 months.
Kramer posits that longer hospitalization and increased use of post-hospital skilled
nursing facilities, along with more physician and specialists visits provided in the FFS
delivery system, may have contributed to the increased rate of FFS patients residing
the community 1 year after being hospitalized for stroke ( Kramer et al., 2000).
Following Kramer, findings were presented by Leach and others (2001)
showing that MCOs achieved better results than FFS in providing certain rehabilitative
services. In a study with relatively small sample size (n = 37), Leach and others found
that elderly orthopedic managed care patients performed better on the SF-36 vitality
scale and the Sickness Impact (SIP) mobility dimensions than the FFS patients. Leach
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
and others report that although the FFS patients received more that twice the physical
and occupational therapy and had significantly longer rehabilitation stays than the
MCO patients, they did not perform as well at 4 months post discharge follow up.
Leach and others concluded that MCOs effectively provided appropriate care to the
elderly population in a rehabilitative environment.
Costs and Managed Care
Although Medicare HMOs have been found to be comparable to FFS in quality
and access, the fundamental motivation for implementing the Medicare risk-
contracting program was to curtail the increase in health care expenses for HCFA
(Brown et al., 1993; Clement et al., 1994; McCombs et al., 1990). Proponents of
moving Medicare beneficiaries into managed care programs contend HMOs can
improve the quality of medical care and expand the range of covered benefits, while
reducing the cost to the Federal government— especially for chronically ill or geriatric
patients (Kane, 1998; Kramer, Fox, & Morgenstem, 1992).
Brown and others (1993) conducted a study drawing from a 1990 survey of
more than 6,400 randomly selected HMO enrollees from 75 risk plans and a
comparable number of FFS beneficiaries. In a segment of their study evaluating cost
savings, controlling for demographic variables, attitudes toward health care, as well as
health status, they found that HMOs reduce hospital average length of stay by almost
17%. By modifying intensity and duration of hospital stays, physician visits and
skilled nursing facility (SNF) use, Brown and others (1993) found that HMOs may
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
have spent about 10.5% less than Medicare would have spent in reimbursements for
all Medicare covered services compared to FFS providers.
In a study analyzing findings from early HMO experience in two HMO
demonstration projects, McCombs and others (1990) endeavored to determine if
prepaid plans achieved cost savings. McCombs and others conducted a longitudinal
study in order to analyze charge data before and after HMO enrollment. They used the
data from the HMO Capitation Demonstration projects at the Fallon Community
Health Plan and the Greater Mansfield Community Health Plan. Because the available
data include both pre- and post-enrollment charges, they provide an opportunity to
overcome some of the data limitations in previous research. Medicare covered both
groups in the predemonstration period. Therefore, any differences in prior use were
not the result of differences in coverage. The comparison group sample was selected
to approximate the enrollee population with regard to age, gender, and geographic
distribution. In addition, the HMO demonstration sites provided charge data for the
services consumed by their enrollees.
McCombs and others (1990) found that some HMOs may, in fact, significantly
reduce the rate of increase in health care costs, even for a high-risk population such as
the elderly. Consistent with previous studies of HMO performance, they found that
hospital charges appear to have been reduced relatively more significantly than
charges for outpatient services.
Fallon Community Health Plan of Massachusetts did not reduce total charges
significantly for survivors in their first year post-enrollment, yet the plan enjoyed
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
reduction in total charges per month after the first year of nearly 39%. Greater
Mansfield Community Health Plan of Wisconsin was not successful in controlling
charges during the demonstration period. Mansfield incurred losses in the first post-
enrollment year for survivors due to a 38% increase in total charges per month. In the
second year, the Mansfield plan was able to reduce losses for survivors to roughly
11%. McCombs and others (1990) conclude that whether or not the potential of an
HMO to save money is realized, may depend on other factors working in conjunction
with prepayment such as financial incentives and organizational structure which might
contribute to an adverse selection of medical risks.
Medicare’ s Capitation Rate
In order to increase enrollment of the elderly into HMOs, TEFRA provided
financial incentives for prepaid groups to enroll Medicare beneficiaries. TEFRA
regulations stipulated that the capitation rate for HMO enrollees would equal 95% of
the adjusted average per capita cost (AAPCC). The AAPCC is defined in the enabling
legislation as the average per capita cost of providing services to the enrolled group of
beneficiaries if the beneficiaries had been receiving services in the FFS sector of the
health care system. The legislation specifies that the average cost is to be adjusted to
account for actuarial differences in risk between the enrolled Medicare group and the
Medicare population in the same geographic area. The adjustment is based on four
factors: (a) age (65-69 years, 70-74 years, 75-79 years, 80-84 years, and 85 years or
over); (b) gender; institutional status (living in the community or not); and (c)
eligibility for Medicaid. Disability is accounted for by calculating separate AAPCC
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
rates of aged and disabled beneficiary enrollment groups (Eggers & Prihoda, 1982;
Epstein & Cumella, 1988).
Since Medicare began contracting with HMOs for the purpose of enrolling its
elderly beneficiaries, policy analysts have criticized the payment methodology used by
HCFA. Numerous studies have demonstrated that the AAPCC explains only about 1%
of total variability in annual costs across Medicare beneficiaries (Epstein & Cumella,
1988; Ellis et al., 1996). When the four risk factors of age, gender, institutional status,
and welfare status do not control for major expenditure differences, the AAPCC
payment could be either high or too low for a given enrollment group. Adverse
selection (from the HMO’s point of view) would occur when the risk of incurring
medical expenses by an enrollment group is greater than that predicted by the
AAPCC. Favorable selection results when the risk is less than predicted by the
AAPCC. When HMOs experience favorable selection, AAPCC based payments could
be high relative to risk (Manton & Stallard, 1992). This favorable selection causes
Medicare’s costs to be higher because Medicare could be paying higher premiums for
beneficiaries who are costing the MCOs less than average to treat (Eggers & Prihoda,
1982). Another part of the problem is that, without adequate payment adjustments for
the medical risk posed by individual beneficiaries, health plans have a financial
incentive to enroll the healthy and avoid the sick or chronically ill (Luft, 1978).
Risk Selection
Patient self-selection is said to be biased when premium payments for health
care do not equal actual costs. This might occur because some factor about the insured
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
population that influences health care costs is not factored into the calculation of the
payment. Bias in patient self-selection is said to be adverse when higher than average
expected risks are enrolled for a prospective capitated payment. Favorable selection is
said to occur when lower than average expected risks enroll (Wilensky & Rossiter,
1986).
This heterogeneity in patient populations has significant consequences. An
HMO could enroll a Medicare population that, by chance alone, differed from the
average Medicare population. In this situation, the HMO would receive
inappropriately high or low payment. This random favorable or adverse selection is
more likely to occur when an HMO is enrolling only small number of Medicare
beneficiaries, and so does not benefit from the “law of large numbers.” In this
situation, risk aversive HMOs would be disinclined to participate in the Medicare risk-
contracting program (Epstein, 1988).
A more significant problem for HCFA, is the incentive for providers to engage
in biased selection by skimming off healthy beneficiaries by making the plans more
attractive to healthy enrollees or discouraging less healthy enrollees. HMOs can
structure advertisements, presentations to seniors, and even benefit packages, that
appeal more to healthy individuals and less to the chronically ill or disabled (Epstein,
1988; Neuman, Maibach, Dusenbury, Kitchman, & Zupp, 1998). Indeed, the
Mathematica Policy Research Institute, in a study of enrollment of 98 health plans for
1987 and 1988, found that all 98 HMOs experienced favorable selection (Ellis et al.,
1996; Hill & Brown, 1990).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
Risk selection could thus lead to increased cost to HCFA by requiring
Medicare to overpay for the healthy patients who have enrolled in HMOs, and leaving
Medicare fully responsible for the sicker beneficiaries who remain in the FFS system.
Importance o f Risk Selection
Medical expenditures are so skewed that the sickest 2% of a large pool are
estimated to use 41% of the total costs, and the sickest 10% consume 72% (Berk &
Monheit, 1992). Competing insurers can make large profits much more easily by
taking in fewer than their proportionate share of these very sick people, or by inducing
them to disenroll, than by producing the fruits of competition for society by becoming
more efficient or providing better service (Light, 1998).
Risk selection becomes a market problem when some insurers are able to
attract favorable risks and still receive the average premium. The insurers with lower
fisks would gain, while those with higher risks still receiving the average premium,
would lose. Several rounds of favorable selection among some health plans would
leave those plans with adverse selection in poor financial condition, serving an
increasingly higher risk population. This could lead to a deterioration of the health
insurance market. Increasing adverse selection could lead to a premium spiral.
Increasing higher premiums would be paid by the least healthy segment of the
population, defeating the purpose of insurance, which is to pool risk (Wilensky &
Rossiter, 1986). With risk selection, competition between health insurers obeys the
inverse coverage law: the more that people need insurance, the less coverage they will
get, and the more they will pay for what they get (Light, 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Extensive research designed to identify risk adjusters that would avoid adverse
selection in competition between health insurers has failed. The maximum explainable
variance in annual acute health care expenditures per individual is around 15%,
leaving at least 85% open to cream skimming (Weiner et al., 1996).
Risk Selection and Nonelderly
In an early review of selection in HMOs, Luft (1981) found few systematic
differences between the health status of HMO enrollees and nonenrollees, although
people who joined HMOs, were lower users of hospital care than those who chose FFS
plans. Hellinger (1987) notes that the studies reviewed by Luft (1981) were
descriptive, containing only simple comparisons of the health status characteristics of
people who joined HMOs or remained in the FFS sector. Hellinger compiled a review
of more recent studies that used multivariate statistical techniques to compare health
status, and demographic characteristics of people who enroll in HMOs with those who
do not. Although there are some exceptions (Dowd & Feldman, 1985; Grazier,
Richardson, Martin, & Diehr, 1986), Hellinger found that Luft’s findings were
supported by the more recent, more sophisticated studies. It was found that enrollees
experienced lower health expenditures during the period before joining an HMO,
although the health status of HMO enrollees and non-enrollees is about the same
(Hellinger, 1987).
Other studies have demonstrated biased selection among HMOs. Jackson-
Beeck and Kleinman (1983), using unadjusted means to report evidence of biased
selection in an employed population, find that people who enrolled in HMOs, used
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
fewer health care services than people who remained in FFS. Price and May (1985)
found similar effects among members of the Federal Employees Health Benefits
Program. The high option Blue Cross and Blue Shield plans attracted high-risk
enrollees, and the HMOs attracted the lower-risk enrollees.
Wilensky and Rossiter (1986) present information from a literature review and
findings presented at a conference on “Biased Selection in Health Care Markets” held
in Berkeley, California in 1985. In this comprehensive review of self-selection in
HMOs, Wilensky and Rossiter found that evidence existed for biased selection in
HMOs. However, they caution that there have been mixed findings on the subject of
biased selection in the U. S. health care system. Indeed one of the more “useful and
sophisticated” studies of a nationally representative sample on self-selection reviewed,
found little evidence for biased selection in U. S. health insurance in the late 1970s
(Farley & Monheit, 1985). Wilensky and Rossiter (1986) conclude that the causes of
self-selection involved two decision-making processes. First, the decisions of
employees concerning their choice of health plan, and second, the decisions of
employers concerning the plans to be offered, with a significant, unpredictable random
component.
Hellinger (1987), in a review and analysis of research regarding selection bias
in HMOs, concluded that prepaid group practice HMOs experience favorable selection
for both those under 65 years of age and those 65 years of age or over. Hellinger
reported that it has been demonstrated in numerous studies that prior use of health
services by HMO enrollees is less than prior use of health services by those who
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
remain in the FFS sector, and that there is considerable evidence that shows a
statistically significant positive relationship between prior use and current use.
Hellinger also noted that the health status of those who enroll in HMOs, is not
significantly different from those who choose a FFS plan as measured by self-
assessments and the number of chronic conditions. This indicates to Hellinger, that
HMO enrollees are more conservative users of medical services, for a given health
status, than are FFS patients (Hellinger, 1987).
Risk Selection and Medicare
Early studies on risk selection in the Medicare demonstration projects or early
in TEFRA contracting, consistently found evidence of favorable selection for
Medicare HMOs. Although an amelioration of this effect was anticipated and
theoretically predicted by some, a reduction in favorable selection into Medicare
HMOs has only recently began to become evident.
In an early study of risk selection, Eggers and Prihoda (1982) examined the
pre-enrollment reimbursement experience of Medicare beneficiaries who enrolled in
three demonstration HMOs to determine whether or not a nonrandom selection
process took place. In two of the HMOs, there was evidence of a selection process
among the HMO enrollees. Enrollees in Fallon and Kaiser health plans were found to
have had, on average, 20% lower Medicare reimbursements than their respective
comparison groups in the four years prior to enrollment. This effect was strongest for
inpatient services, but a significant difference also existed for use of physician and
outpatient services. In the third HMO (Mansfield), there was no statistically significant
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
difference in hospital costs between the enrollee and comparison groups. However,
outpatient and physician reimbursements were significantly higher (22%) among the
enrollee group (Eggers & Prihoda, 1982). It was hypothesized that because Mansfield
employed almost all physicians in its area, it did not realize the same degree of
favorable selection found in more traditional group practice HMOs (Hellinger, 1987).
Hill and Brown (1990) drew a random sample of over 100,000 elderly
Medicare enrollees from 98 HMOs, and a similar-sized control group from traditional
FFS Medicare from the same market areas. They adjusted all comparisons for
differences between the two groups in age, gender, institutional status, and welfare
status. The study found that HMO enrollees used fewer Medicare covered services the
period before they joined the HMO, compared with those in the same market area who
stayed in FFS. Over the 98 plans the HMO enrollees, when they were enrolled in the
FFS, cost 23% less than the nonenrollees (Hill & Brown, 1990).
In the same study, Hill and Brown (1990) examined a second measure of the
risk mix among HMO and non-HMO enrollees— the rate of hospitalization for
conditions that were relatively nondiscretionary and known from other research to
predict future expenditure. There were 25% fewer nondiscretionary admissions among
HMO enrollees in the period before enrollment.
Hill and Brown (1990) also examined mortality rates in the post-enrollment
period among enrollee and nonenrollees. The results, when adjusted for demographic
characteristics showed that for all 83 plans studied, the HMO enrollees had an average
of 25% lower mortality than nonenrollees. These data suggest not only that healthier
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
beneficiaries enrolled in the HMOs, but also that the HMOs were able to avoid the high
costs associated with the last few months of life.
In another study examining prior reimbursement for a sample of nearly 100,000
new Medicare HMO enrollees compared to nonenrollees in 1987 and 1988, Brown and
others (1993) found the risk adjusted reimbursement about 20% less for the enrollees.
Brown estimated that the HCFA expenditures were approximately 5.7% higher than
they would have been for in a strictly FFS system. Brown and others (1993) conclude
that this increased expenditure is primarily a result of favorable selection into Medicare
risk plans.
Using a detailed telephone survey of almost 13,000 Medicare beneficiaries,
more than half of who had been enrolled in an HMO for at least 3 years, Brown (1994)
again found significant differences in the health care characteristics of the two
populations. Brown found that enrollees are much healthier on several measures of
health status (including number of impairments on activities of daily living and number
of impairments on instrumental activities of daily living) that indicate a lower
propensity to seek care.
In a study involving several hundred thousand FFS and HMO beneficiaries in
southern Florida from 1990-1993, Morgan, Vimig, DeVito and Persily (1997),
demonstrated that the effectiveness of the Medicare managed care system is
undermined by a marked selection bias with respect to HMO enrollment and
disenrollment. Morgan determined that significant numbers of Medicare beneficiaries
are switching in and out of HMOs and FFS in order to reduce their expenses or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
increase their access to specialist or other needed services. This maneuvering relieves
the HMOs and their providers of the cost and pressure of providing more expensive
services, while shifting the expenses back onto to Medicare (Morgan et al., 1997).
Riley, Tudor, Chan, and Ingber (1996) compared the health status of 863 HMO
enrollees with that of 4,576 nonenrollees, controlling for demographics and area of
residence, using 1994 data from the Medicare Current Beneficiary Survey (MCBS).
They found HMO respondents were less likely to report fair or poor health, functional
impairment, or heart disease. Average predicted costs based on various health status
measures were substantially lower for HMO than FFS respondents. On the other hand,
Dowd, Moscovice, Feldman, Finch, and Wisner (1994), studying three HMOs in an
established HMO market, found no evidence of selection among Medicare HMO
enrollees in the Twin Cities, using both survey and Medicare claims data.
Regression Toward the Mean
Several empirical studies (Haan, Selby, Quesenberry, Schmittidiel, Fireman, &
Rice, 1997; Mello, 1999; Riley et al., 1989), as well as early theoretical work (Welch,
1985a, b), indicate that the effects of risk selection, either favorable or adverse, may
not persist indefinitely.
Riley and others (1996) in discussing biased selection, note that their data from
the MCBS are taken from a time of very rapid growth in the Medicare risk-contracting
program. Only 65% of their HMO respondents were enrolled under risk contracts for
all of 1993 and 1994. They consider that their findings may reflect in part the effect of
having large numbers of recent HMO enrollees in the Medicare program. Riley and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
others (1996) postulate that when growth under Medicare risk-contracting programs
stabilizes, health status difference between the HMO and FFS sector may begin to
diminish.
Whenever consumers have the choice of several insurance plans, there may be
biased selection: that is, the enrollees of one plan may be healthier than the enrollees
of another plan. This possibility has at least two implications. First, although prepaid
group practices, in general, have lower costs than conventional insurance, this may
simply be the result of biased selection. Second, even if prepaid group practices truly
have lower costs, competition among insurance plans may involve competition for
desirable patients instead of cost cutting competition (Welch, 1985a).
Biased selection can be the result of either enrollee behavior or HMO behavior.
The incentive for a person to enroll is the result of two conflicting tendencies. On the
one hand, HMOs tend to offer more comprehensive coverage than does conventional
insurance such as Medigap, giving the sick greater incentive to enroll than the well.
On the other hand, joining an HMO usually entails switching physicians. People with
high health care utilization in the past may have stronger relationships with their
physicians. This would discourage the sick from enrolling. Given the role of
attachment to ongoing providers, most observers feel that the selection process attracts
low utilizers to the HMOs and leaves high utilizers in the FFS sector (Luft, 1978).
The incentive for biased selection in disenrollment is substantially different.
After enrollment, the high utilizers will have stronger relationships with HMO
physicians. In this situation, both the physician relationship and the more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
comprehensive coverage give the sick a greater reason to remain enrolled in the HMO
(Welch, 1985b).
The incentive for the HMO is to select healthy enrollees. Because of the wide
disparity in enrollee health care needs, an insurer’s (or provider’s) success in the
market depends in part on the mix of enrollees attracted. This gives insurers a strong
incentive to actively cream-skim the potential enrollees. Competition may be primarily
competition for healthy people instead of more economically efficient cost-cutting
competition (Newhouse, Buntin, & Chapman, 1997).
However, measuring the magnitude of the supposed differences can be very
difficult. Actual FFS costs generated before enrollment could stem from a short-term
difference in health care needs, rather than from a major inherent enrollee-
nonenrollees difference in health status. Another issue is that Medicare HMO
enrollees tend to be lower income than nonenrollees are and less likely to have
Medigap insurance. Therefore, they may have faced barriers to adequate care in the
FFS sector. The prior service use observed may actually indicate the lower level of
access experienced before joining an HMO (Rossiter, 1988). By giving a structured
source and perhaps enhanced access to care, HMOs would experience increased costs
by providing a different, more accessible product compared to FFS. Moreover, if
enrollee costs regress toward the mean over time, as others have demonstrated (Beebe,
1988), the pre-enrollment differences are overstated (Luft, 1981).
Welch (1985b) argues that many of the estimates of biased selection overstate
the effect. They make no allowance for regression toward the mean, the tendency of a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
variable that is different from its mean to move toward the mean over time. This
tendency was originally noted in comparing the heights of children and their parents:
tall parents tend to have offspring who are tall, but not as far above the average as they
are, and conversely for short parents. Welch demonstrates that, consistent with this
phenomenon, health care costs regress toward the mean over time.
Many studies have estimated the extent of biased selection using the health
care expenditures or utilization rates of individuals before their enrollment. Some
studies compared the pre-enrollment expenditure rates of individual who joined
HMOs with those of individuals who did not join (Hill & Brown, 1990). Still other
studies compared the utilization of enrollees who stayed and those who subsequently
disenrolled. The methodology of these studies assumes not only that high utilizers in 1
year remain high utilizers in subsequent years, but also that their utilization on average
does not fall. In contrast, the evidence reviewed by Welch indicates there is substantial
regression toward the mean both for individuals above and below age 65. In sum, most
of an individual’s cost does not persist over time: It is year specific” (Welch 1985a,
p. 1240).
In discussing the numerous studies that he reviewed, Luft (1994) noted that all
of the studies have relied on a one-time snapshot of the biased selection issue. Usually
indications of prior use of services before joining an HMO were used to imply biased
selection after enrollment. However, the problem with using prior service use and
costs as measures of biased selection on only new enrollees is that the market
experience for all enrollees may be quite different (Miller & Luft, 1994).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
Besides the numerous studies examining or reporting biased selection
discussed above, there has been a limited number of studies which demonstrate,
empirically, the effects of regression toward the mean in managed health care
organizations.
An early Kaiser Medicare demonstration case study demonstrates how
preferential selection dissipates. The health plan’s enrollees had pre-enrollment
expenditures below nonenrollees. Following enrollment in 1980, enrollees had
hospital days of 1,667 per thousand in 1981. The rate increased to 2,116 admissions
per thousand in 1984. Half of this increase was attributed to the aging of the cohort
and half appears to be due to regression toward the mean and adverse disenrollment
(Welch, 1985b).
Riley and others (1989), in a longitudinal study, tracked mortality of three
HMO and FFS demonstration cohorts for 6 years after enrollment. Their data suggest
an initial strong regression toward the mean in enrollee health status in the first 2
years, followed by further gradual regression toward the mean over time. By year 6,
mortality was the same for the HMO enrollees as the FFS beneficiaries, after
adjustment.
A more recent Kaiser study evaluated health care service utilization of two
cohorts of approximately 3,000 members aged 65 or older of the Northern California
Kaiser Permanente Medical Care Program in 1971 and then 1980. The subjects were
compared for hospital and outpatient utilization during 9-year follow-up periods
(1971-1979 and 1980-1988). The cohorts were grouped as less than 10 years, 10 to 20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
years, and more than 20 years enrollment. As length of enrollment increases, the
interval since selection effects such as healthy worker, effect and enrollment screening
practices lengthens. The impact of those selection factors diminishes with time.
Hospitalization rates were 17% lower for persons enrolled less than 10 years
compared with those enrolled for longer periods after adjustment for age, gender, and
cohorts. There was no difference in rates between the other two longer strata. Thus,
any selection effects of these enrollees appear to wash out at about 10 years of
enrollment (Haan et al., 1997). In a recent dissertation study of health services
utilization and selection bias based on 1992-1996 Medicare Beneficiary Survey data,
Mello (1999) found little evidence of persistent favorable selection into Medicare
HMOs in the1990s.
In summarizing his work on biased selection, Welch (1985 a, b) concluded that
there appears to be little or no bias in prepaid group practice enrollee populations. He
recommends that attempts to increase the sophistication of the AAPCC to compensate
for biased selection should focus on predictors to the permanent component of
expenditures (chronic health conditions) instead of using past expenditures as the
predictor.
Summary o f Literature
TEFRA of 1982 implemented risk contracting with HMOs with the objective
of controlling costs and maintaining quality. Early research comparing HMOs with
FFS on various aspects of quality of care found that HMOs provided comparable
levels of service and quality (Luft, 1980; Manning et al., 1984). Although later work
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
supported the findings on quality of care of the earlier studies (Clement et al., 1994;
Yelin et al., 1996), the question of risk selection began to cloud the issue. Numerous
studies began to report that managed care organizations had been benefiting from
significantly favorable selection in the demographic and health status of new enrollees
(Hill & Brown, 1990). HMOs in general and Medicare HMOs in particular, had been
demonstrated to be enrolling younger and healthier beneficiaries, thus calling into
question the financial benefit to HCFA (Brown 1998; McCombs et al., 1990) and the
performance capabilities of the HMOs (Ware et al., 1996). The findings on risk
selection in HMOs raise the question of whether or not the HMOs are adequately
serving the chronically ill or most vulnerable populations.
Risk selection has been demonstrated in several studies, particularly in new
HMO enrollees. However, early theory (Welch, 1985a, b) and more recent empirical
work (Haan et al., 1997; Mello, 1999; Riley et al., 1989) indicate that favorable
selection for the HMOs, may not persist indefinitely as the enrolled population ages
and as degree of HMO market penetration reduces the opportunities for advantageous
selection in new enrollees.
Both the early research and later studies have found that in general HMOs have
provided quality of care and levels of satisfaction comparable to the FFS sector
(Clement et al., 1994). It has been demonstrated that Medicare beneficiaries, either in
FFS or in managed care, are satisfied with their health care services (Adler, 1995).
However, there have been some concerns raised about the adequacy of service
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
provided by HMOs to beneficiaries with chronic health care needs (Kramer et al.,
2000; Ware, 1996).
However, concerns have been raised about managed care organization’s ability
to provide satisfactory levels of care. Such issues about service care provision are of
special concern with an aging population, since the two largest public health insurance
programs affecting the elderly, Medicare and Medicaid, are increasing their use of
managed care programs (Saaga et al., 1998). The present proposed study will explore
these issues.
Research Questions
Are the health risk factors similar between the senior FFS and MCO
populations in Los Angeles County?
As discussed previously, early theory and some empirical work indicate that
the favorable selection experienced by managed care risk contractors may not persist
indefinitely. Several factors would predict that a more mature HMO market would
yield more similar HMO and FFS beneficiaries’ populations. The population of
enrollees, regardless of favorable selection would age within the HMO. More
beneficiaries who were HMO enrollees while employed could retire into the Medicare
HMO, rather than requiring that beneficiaries switch into the HMO, as was the case in
earlier studies.
California, in general, and Los Angeles County in particular, have been in the
forefront of managed care penetration into the health care marketplace. California
managed care market share exceeded 22% by 1985 and by 1996, the timeframe of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
Los Angels County Health Survey utilized in the present study, had exceeded 75%
(HCFR, 1996). Although enrollment of Medicare beneficiaries in managed care
lagged significantly behind the enrollment of the commercial market as in the rest of
the country, Medicare beneficiaries’ participation in managed care was dramatically
higher, at approximately 35%, than Medicare enrollment rate nationally at
approximately 13% (HCFR, 1996; HCFA, 1998; HCFA, 1999).
Because a significant portion of the population has participated in managed
care for a number of years, and because a significant proportion of the population has
had the opportunity to age into as well as age within Medicare HMOs, the LACHS
affords the opportunity to assess whether or not the regression toward the mean
anticipated by HCFA and predicted by Welch (1985 a, b) manifests itself.
This study presents several advantages over earlier studies. This is a random
sample of the population at a given time. Many of the previous studies that used
HCFA Medicare beneficiary data, imputed health care cost or utilization
characteristics from new enrollment or disenrollment samples to a sample population
(Hellinger, 1987). This study can be used to evaluate the health care characteristics
and vulnerabilities of the population, as they exist at a given time. The evaluation is
not confounded by whether or not beneficiaries have recently enrolled or disenrolled
differences in treatment patterns or attempts to compare costs of care across different
health care modalities.
Is the quality o f care comparable between FFS and MCO populations in Los
Angeles County?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
There has been increasing concern in the research literature on the differential
effect of factors such as health insurance programs, either FFS or managed care, on
delivery of health care services on disadvantaged populations and racial and ethnic
minorities (Vimig, Morgan, Vito, & Persily, 1998). In addition, there are some
indications that responses to financial incentives have not been optimal to the health
care consumers (Newhouse et ah, 1997). The FFS system has been criticized on the
one hand, for over utilizing services, providing inappropriate, ineffective, or
unnecessary treatments. On the other hand, it has been faulted for not providing the
most effective or most advanced services to the most needy, vulnerable, or marginal
members of society (Escarce, Epstein, Colby, & Schwartz, 1993; Lee, Gehlbach,
Hosmer, Reti, & Baker, 1997). Managed care organizations have been questioned on
responding to financial pressures by curtailing needed coverage, denying services,
restricting access to specialists, or reducing quality or services by over-booking
providers (Clement et al., 1994).
There is some concern as well as to the effects or race and ethnicity and other
socioeconomic factors on access to and utilization of types of health insurance
(Blustein & Weiss, 1998) and health care services (Lee et al., 1997; Frieman, 1988).
The importance of income and education have emerged as important predictors of
need for and access to health insurance as well as other health care related services
(Lillard, Rogowski, & Kington, 1997).
Effects of sociodemographic variables and their relationship to health care
delivery system (FFS or HMO), or health care related factors such as insurance
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
coverage require further exploration, especially in a vulnerable population such as
Medicare beneficiaries and other seniors (Miller, 1998; Osmond, Vranizan,
Schillinger, Stewart, & Bindman, 1996).
Quality o f Care, Service Access, and Patient Satisfaction
Issues such as quality of care, access to services and patient satisfaction, are
particularly important in the aging, whose use of health care service begins to increase
dramatically after age 65. Such issues are of special interest in comparing the two
most prominent systems of service provision: fee-for-service and managed care,
especially regarding disadvantaged or chronically ill seniors. Relatively little is known
about these systems, either individually or comparatively, as they address these issues.
The study proposed here will explore these variables.
Quality of care will include such factors as availability and utilization of
services, availability of preventive care such as mammograms, and the provision of
care (Kassirer, 1997). Access to care will address both perceived access and
accomplished access variables (Aday & Andersen, 1974). Access to care will be
defined by variables such as ease of access, location of care, unmet health needs, and
the availability of clinical and preventative services (Bloom, Simpson, & Parsons,
1997).
Geographically Explicit Sampling
To date, there has been limited research addressing aging, diversity, and FFS
versus managed care in a geographically distinct region (Siddharthan, 1991). Such
geographic specificity appears to be necessary to evaluate such effects (Gold &
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Stevens, 1998). This study will examine the interaction effects, at a geographically
specific level, of service delivery model, adequacy of service and sociodemographic
and health care related factors. Numerous previous studies have addressed such
population based issues using large national surveys and national data bases such as
the National Center for Health Statistics National Health Interview Survey, and the
Medical Expenditure Panel Survey of the Agency for Health Care Policy and
Research. Although these are large, sophisticated survey instruments, the effective
sample size is too small to provide even state-level data (Eden, 1998). However,
effects of health plans can vary across populations and across markets such that
geographically specific surveys may be necessary (Gold & Stevens, 1998) to evaluate
the differential effects of the social-demographic and health care related factors both
within the FFS and managed care paradigms and between the two systems as well.
The present study, based on a large population based survey of Los Angeles County,
offers the statistical power to accomplish such analysis.
It has been demonstrated that local level data offer several advantages over
national level survey and census data (Myers, 1992). The use of local (county) level
data allows for the exploration of issues, especially those regarding changes in the
health care market that may be relatively unique to large urban populations such as
Los Angeles County. Furthermore, the geographically explicit area of Los Angeles
County offers an important laboratory for exploring population related variables,
especially among disadvantaged populations, because it offers an extremely diverse
population. Adding to the significance of such study is the lack of research in ethnic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
diversity in elders (Williams, La Vizzomourey, & Warren, 1994), especially Hispanics
and Asians (Cornelius, 1995). Moreover, existing variables are not unlike those to be
found in other large American locales such as Florida, New York, or Chicago (Myrtle,
personal communication, 1998). One major difference making Los Angeles County an
especially appropriate locale for this study is the strong presence of managed care.
Summary
Changes in the health care market, as well as modifications of governmental
policy, are leading to changes in health care services and health insurance coverage for
millions of Americans in both traditional fee-for-service and managed care systems.
Although the majority population of citizens, particularly those aged 65 and over,
appear well served by their available health coverage, it remains to be seen if the
current transitions in the local health care market will prove to be beneficial to all. Of
particular interest, is the effect of the emerging paradigm of managed care on the most
vulnerable and disadvantaged of our citizens? This study will help to further detail,
within a defined geographical region, some of the institutional as well as socio
demographic and health care related factors which impact the quality of care, health
care access, and consumer satisfaction of some of the most needy individuals in the
population.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
METHODOLOGY
This chapter describes the data acquisition and statistical analysis techniques
used in this study. The purpose of the study is to analyze the relative health status and
quality of care between senior citizens enrolled in Medicare FFS versus HMO
healthcare plans. This chapter describes the health care survey instrument utilized in
gathering the data, the telephone methodology utilized and the statistical methods used
to analyze the data.
Data Source and Method
Data to be utilized in this study is derived from the 1997 Los Angeles County
Health Survey (LACHS). The Health Survey was commissioned by the Los Angeles
County Department of Health Services and conducted by Field Research Corporation
(FRC). The main objective of the survey was to examine the health status indicators of
persons living in Los Angeles County. A total of 8,004 Los Angeles County adults,
age 18 or older, including 1,054 individuals age 65 years or older, were included in the
survey. Respondents were interviewed using a random digit dialing telephone survey
(LACHS, 1997).
The Survey Questionnaire
The Los Angeles Department of Health Services and the Field Research
Corporation (FRC) developed the questionnaire. The survey was extensively pre
tested and revised before the actual survey was conducted (FRC, 1997). A well-
constructed questionnaire will maximize the probability that interviewers will
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
administer it in exactly the same fashion to each respondent (Miller, 1991). The
LACHS was programmed onto FRC’s computer-assisted telephone interviewing
(CATI) system. The CATI system is used to control the telephone scripts read to
individual respondents by displaying the appropriate questionnaire items on a
computer screen at each interviewer’s booth. After reading each question aloud to the
respondent from the screen, the interviewer enters the precoded answer using the
keyboard directly to a computer disk. The use of on-line interviewing using CATI
allows for greater consistency in interviewing by controlling skip patterns, branches,
randomization of items in a battery, and other control features during the call
(Nicholls, 1988).
After being programmed into the CATI system, the survey was validated
through pretesting on small samples of adults in Los Angeles. Subsequently, the
survey was revised several times to enhance interviewer competence and respondent
cooperation, and to ensure the meaningfulness and validity of survey responses
(Fowler, 1995). Moreover, much of the questionnaire was based on previous national
surveys. The survey, which is comprised of more than 100 questions, gathered
information on a wide variety of demographic, sociological, lifestyle, and health care
related issues.
After the English language version of the survey was validated, it was
translated into and validated for six other languages prevalent in the Los Angeles area.
These included: Spanish, Cantonese, Mandarin, Korean, Vietnamese, and Tagolog.
During the survey, when a household was contacted they were informed that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
interview could be conducted in any of seven languages. If a non-English language
was preferred, interviewers fluent in that language called the household back to
conduct the survey (FRC, 1997). In total 1,752 interviews were conducted in
languages other than English (Table 2).
Table 2
Number o f Interviews Completed by Language
Total interviews completed 8,004
English 6,252
Spanish 1,466
Mandarin 91
Cantonese 50
Korean 81
Vietnamese 41
Tagolog 23
Because more than 98% of all County residents age 5 years or older, are fluent
in at least one of these seven languages the addition of any single other language to the
survey would not have increased the survey’s coverage by more than X A of 1% (FRC,
1997).
Homeless people, who are actually living in homeless shelters, on the streets,
or other nonhousehold settings, rather than living with family or friends, are missing
from the survey sample. Although this population might be of particular concern to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
public health and department of social service personnel, they are not represented in
this survey.
Survey sponsor identification. Respondents understanding about issues such as
confidentiality, the voluntary nature of a project, and who will be using the informa
tion, can have an effect on answers provided (Fowler, 1993). All survey respondents
were told that the study was being conducted for the Los Angeles County Department
of Health Services and were assured that all responses were confidential.
Contact attempts. Up to six attempts were made at different times and on
different days to increase the probability of finding adults available for the interview.
If necessary, appointments were made for a specified date and time to maximize
cooperation rates. In addition, refusal conversion was attempted at each household
where a spokesperson initially refused to participate in the survey. Of 17,662
households contacted, 8,004 completed the survey (Table 3).
Interview Length and Respondent Incentives: The 1997 Los Angeles County
Health Survey was comprised of two parts. Part I, the Main Questionnaire, focused on
adults. Part II, the Children’s Questionnaire, focused on children under the age of 18
years. This analysis is based on the 106 question Main Questionnaire. The average
length of the telephone interview applicable to the Main Questionnaire was 25-30
minutes. No monetary incentives were offered to respondents to complete the survey.
Survey Data
Of the 8,004 households that responded to the survey, 1,054 of them comprise
households with respondents aged 65 years or older. This population of seniors
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
Table 3
Disposition o f Telephone Contact Attempts
Total RDD listings opened 52,444
Nonresidential (not assigned, not in service, businesses, institutions, 15,944
modems)
No answer/busy/answer machine after all attempts 15,700
Household ineligible (not in county, no adult in household) 1,263
Other incomplete 1,919
Contact made with county households 17,617
Initial refusal prior to respondent selection 7,015
Language barrier (other than languages available for 387
survey)
Designated adult never available for interview 1,688
Terminated by designated adult during interview 487
Designated adult completed survey 8,004
represents the data set for the present analysis. Statistical analysis of this data set will
be completed on several levels related to the sociodemographic and health care
variables as well as how they relate to quality, access and satisfaction issues for both
FFS and m an aged care programs in an elderly population.
Analysis o f Data
The study proposed here will address a variety of issues important to elderly
populations within fee-for-service and managed care contexts. The Los Angeles
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
County Health Survey contains more than 100 questions addressing issues of health
services utilization and forms the data set for the present study. To adequately explore
the system, sociodemographic and health care variables to be addressed, a variety of
statistical methods are required. Statistical methods applied in this study include: (a)
contingency table analysis (CTA) of the variables, (b) t-tests and Wald chi-square tests
of significance, and (c) logistic regression to obtain odds ratios.
The two health care systems, fee-for-service and managed care, will be
compared and contrasted in the present study. The specific variables to be addressed
were determined through review of responses to the more than 100 questions
comprising the survey data. For example, quality of care will include questions
regarding availability of and utilization of routine services, as well as preventive care
such as mammograms or colonoscopy. Access to health care services will include
questions about having a regular source of care (Lambre, Defiese, Carey, Ricketts, &
Biddle, 1996), the reporting of unmet health care needs, and the use of routine
checkups (Cohen, Bloom, Simpson & Parsons, 1997). Consumer satisfaction will be
addressed through use of questions regarding satisfaction and affordability of patient
care.
Healthcare Variables
Variables for hypothesis testing were categorized as “health risk” and “quality
of care” variables. The “health risk” factors in this study correlate to the characteristics
of the “population at risk” in the Aday and Andersen (1974) model. The character
istics of the population are composed of the predisposing, enabling and need factors of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
the population (Aday & Anderson, 1974). In this study, the predisposing component is
represented by the age, gender, and race/ethnicity variables. The enabling component
is represented by income (relative to federal poverty level), insurance coverage, and
having a regular source of care. The need component is represented in this study by
self-perceived health status, chronic conditions, and hospital admission in the past year
(Table 4).
The “quality of care” factors relate to the outcome indicators of Aday and
Anderson, Aday, and Andersen (1974) categorize the outcome indicators as utilization
and satisfaction. For this study, the utilization variables are represented by doctor visit
in the past year and preventive services (routine examinations). The satisfaction
category is measured by the variable access, wait for appointments, afford
medications/treatments and satisfaction (Table 4).
Insurance status. In the LACHS a respondent’s health insurance status was
determined by a series of questions (questions 17-28) that determined whether or not a
person had health insurance and whether the source of that insurance was a public
(government) or private agency.
For respondents age 65 years or older, or disabled, respondents were asked if
Medicare in question 18 currently covers them. If the respondent is covered by
Medicare, question 19 determines if the respondent is in a fee-for-service plan or a
managed care plan. “ Is your current health insurance coverage a fee-for-service plan
that allows you to go to almost any doctor or hospital or is it an HMO, PPO or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
Table 4
Health Risk and Quality o f Care Variables Used in Analysis o f Fee-for-Service
and Managed Care Organizations
Health risk variables Quality of care variables
Age/gender Doctor visit in past year
Race ethnicity Preventive examinations
Income Access
Insurance Waiting times for appointment
Health status Afford medications/treatments
Chronic conditions Satisfaction
Hospital admissions
Regular source of care
another type ofplan that directs you to a list o f doctors and hospitals who are on the
plan?”
Question 23 asks whether or not the respondent was on Medicaid (called Medi-
Cal in California) the federal government’s health insurance program for people on
public assistance. If the response to Q23 is “yes,” the respondent was asked in Q24, if
the respondent was in a managed care plan or if they could go to “any doctor who will
accept Medi-Cal.”
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
Type o f health plan. For the purposes of this study, the responses of those 65
years and older were grouped to yield health insurance status of MCO, FFS or
uninsured.
Self-reported health status. In the LACHS question 60 was used to determine
the respondents overall self perceived health status. “ In general, would you say your
health is excellent, very good, good, fair or poor? ”
Chronic conditions. In the LACHS question 62 was used to determine the
prevalence of chronic health care conditions. The question asked: “ As fa r as you
know, do you have any o f the following medical conditions or problems? (a) arthritis,
(b) diabetes or sugar in the blood, (c) heart disease, (d) cancer, (e) kidney disease, (f)
smoking-related lung diseases, (g) HIV infection or AIDS, (h) high blood pressure or
hypertension. ” After each “yes” response, the respondent was asked in Q63, “ Are you
being treated by a physician fo r __________ ? ”
The overall rate for any chronic condition was determined. In addition, chi-
square analysis was run on individual disease states using dummy variables comparing
FFS and MCO health systems.
Regular source o f care. To determine if the respondents have a regular source
of care they were asked if they have one particular place or health provider to whom
they go most often when they are sick or want advice about their health. The
respondents were then asked a series of follow-up questions (Q32 to Q40) to
determine if the patient’s regular source of care is a doctor’s office or Kaiser facility,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
country or community clinic, hospital or outpatient clinic, emergency room, or other
healer other than an M. D.
Wait time for appointment. In order to further evaluate access to healthcare
services, the respondents were asked how long they had to wait for an appointment to
see their healthcare provider. “Wait time” for an appointment can be evaluated in two
ways: (a) how long the patient must wait to get an appointment with the healthcare
provider, and (b) how long the patient must wait to been seen once the patient arrives
at the place of the appointment.
The first option (A) will be termed “wait time” for an appointment. The
possible responses to Q46 included: no appointment necessary, same day-less that 2
days, 3-6 days, 1-2 weeks, 3-4 weeks, more than 4 weeks.
The second option (B) will be termed “office wait.” In Q49, the respondents
are asked once they got to their respective healthcare facility how long did they
usually have to wait to see the healthcare provider with whom they have the
appointment. The available responses are: 0-30 minutes, 31-60 minutes, between 1 and
3 hours, 3 to 5 hours.
Access. To determine ease of access to their healthcare services the
respondents were asked: “Overall, how easy or difficult is it for you to get medical
care when you need it: Would you say it is very difficult, somewhat difficult, somewhat
easy, or very easy?
Satisfaction. After earlier questions to determine the regular source of care,
Q52 sought to determine the respondent’s level of satisfaction with their healthcare
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
services. The question asks: “Overall, how satisfied are you with the care you are
receiving from this (doctor’ s office) (clinic) (hospital) (place)-extremely satisfied,
very satisfied, some what satisfied, not very satisfied, or not satisfied at all? ”
Last visit. Question 5 sought to determine whether or not the respondent has
seen a health care provider in the past year. Health care providers include doctor,
nurse, or physician’s assistant.
Hospitalization. Question 30 was directed at determining whether or not or
how many times the respondent has been admitted to the hospital in the past year. The
hospital admission data were analyzed to investigate three factors: (a) the over all rate
or hospitalization between FFS and MCOs; (b) the rates of any hospital admission
versus no admission between FFS and MCOs; and (c) the rates of “high” (two or
more) admissions between the two systems.
Preventative health care procedures. In order to assess their use of
preventative healthcare services, respondents were asked in Q54, if during the past 2
years, they had any of a number of tests or exams administered to them. All
respondents were asked if they received a colo-rectal exam or a blood test for
HIV/AIDS. Males were asked if they received a testicular exam or if they received a
prostate exam. Females were asked if they received a pap smear or if they received a
physical breast exam by a doctor, nurse or other health professional. Females were
also asked if they had received a mammogram.
Afford medications/treatment. In order to determine if barriers exist in the
utilization of healthcare services in Q7, the LACHS asks if in the previous 12 months,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
if there was ever a time when the respondent needed any of several medications or
services, but did not get it because they could not afford it. The specific items are:
prescription medicine, mental health care or counseling, dental care (including
checkups), eyeglasses, and seeing a doctor for a health problem.
Specificity o f Locale
The geographic specificity of the present survey study, through its application
within Los Angeles County, aids in the validation and interpretation of the findings.
Los Angeles County offers distinct advantages as a research locale because of the
significant presence of vulnerable populations, especially the elderly population and
Medicare beneficiaries targeted in this study, and the strong presence of managed care
systems within the area.
The Telephone Survey
The original sample of telephone listings developed for the survey was drawn
using a random digit dialing (RDD) technique. The RDD sampling approach is the
most inclusive and representative telephone sampling method available. RDD insures
that all telephone households, including both listed and unlisted, are given an equal
probability of being included in the survey (Massey, 1988). “The development of
telephone interviews and procedures for sampling households by means of random
digit dialing, may be the most important innovation in survey research since the
introduction of multistage-probability sampling” (Miller, 1991, p.162). There has been
widespread use of telephone surveys to conduct population-based research on health
related issues (Anderson, 1998). Telephone surveys offer certain advantages over in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
person interviews. The advantages include substantial cost savings relative to in-
person surveys. In addition, they enhance the ability to rapidly collect data and to
standardize the interviewer-respondent interactions (Groves, Biemer, Lyberg, Massey
& Nichols, 1988, 1988).
In any telephone survey, there are several factors that can introduce response
bias into the sample which are to be addressed in the research design: (a) households
without telephones, (b) unlisted telephone numbers, (c) answering machines, (d)
eligible respondent not at home, (e) non-English speaking household or respondent,
and (f) homeless people.
Although households without telephones are by definition excluded from
telephone surveys, data from the 1990 Census and 1994 National Health Interview
Survey (NHIS) indicate that 95% of households nationally and over 96% of all
households in Los Angeles County have telephones (Anderson, 1998; FRC, 1997).
The issue of unlisted telephone numbers is obviated by the RDD protocol that
calls a random sample of all residential telephone numbers in the county. As noted
above, the RDD sampling approach is the most inclusive and representative telephone
sampling method available insuring that all telephone households, including both
listed and unlisted, are given an equal probability of being included in the survey. The
matter of respondents not at home or encountering an answering machine was
addressed by (a) conducting the survey in the evenings and on weekends, and (b)
recalling the selected telephone number up to six times. In addition, studies by Piazza
(1993) and Oldendick (1993) have demonstrated that widespread use of answering
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
machines in the population have not presented a significant threat to obtaining
representative samples using RDD survey methodology (Anderson, 1998).
In the United States, more than 90% of all households are reachable by
telephone, so that the overall coverage of an area probability sample of households
begins to approach the levels typically obtained by personal interviews (Thomberry &
Massey, 1988). For some subpopulations such as the rural, poor and the elderly,
households not covered by telephones might prove significant (Miller, 1991). In
analyzing 1994 NHIS data, Anderson, Nelson, and Wilson (1998) found that, though
small, there were some systematic differences in telephone coverage among major
population subgroups. They found that coverage was highest among respondents aged
65 years or older at 97%, compared to 95% overall. Among Black and Hispanic
respondents coverage was 90% compared to 97% among Whites. However,
socioeconomic status accounted for the largest difference in telephone coverage.
People living below the poverty level had only 83% telephone coverage.
Anderson and others (1998) found that differences in health related variables
between all respondents to the 1994 NHIS and those with telephones, tended to be
small. They found differences averaging 1.4 percentage points or less for all of the
health related factors covered (Anderson et al., 1998). These findings are consistent
with earlier work by Thomberry and Massey (1988), which found that, because
telephone coverage is so extensive in the United States, the difference between those
who have telephones and the total population is quite small. The Anderson and others’
analysis of the NHIS results indicate that, even for those below the poverty level (who
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
have only 83% telephone coverage), limiting attention to only those with phones does
not seem to affect estimates of health care factors very much (Anderson et al., 1998).
Statistical Software
Database correlation and statistical analysis were carried out using SAS for
Windows 7.1, SAS Enterprises Cray, North Carolina. Percentages for the demographic
and health care utilization variables with chi square tests of significance are reported.
Unstandardized logistic regression analysis is used for developing odd ratios. For the
purposes of this study, a confidence level of .01 is considered highly statistically
significant; a .05 level is considered statistically significant; and a .10 level is
considered suggestive of a trend or to have some probability of a difference.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
FINDINGS
This chapter presents the findings obtained from the 1997 Los Angeles County
Health Survey (LACHS) related to the study population, those respondents 65 years of
age and older enrolled in Medicare. The findings are presented in five sections. The
first section reviews the methodology used to categorize the population and the
variables used. The next section presents the descriptive statistics (demographics) of
the study population. The third section describes the health insurance status of the
target population as it pertains to this study. The fourth section presents the hypotheses
concerning “health risks” and “quality of care” and the variables used to test those
hypotheses. The final section presents post hoc multivariate analysis of the health risk
and quality of care variables controlling for type of supplemental insurance and
selected socio-demographic characteristics.
Definition o f Variables
In this analysis, the system variables are access— having a regular source of
care (RSC) and waiting times for an appointment. The population variables are (a)
age; (b) gender, race and ethnicity; (c) income, (d) self-reported health status; and (e)
chronic conditions (Table 4).
In this study, “quality of care” is categorized as utilization and satisfaction. For
this study, the utilization variables are represented by doctor visits in the past year and
preventive services (routine examinations). The satisfaction category is measured by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
the (a) variables access, (b) wait for appointments, (c) affordability of medications-
treatments, and (d) satisfaction (Table 4).
Demographics/Descriptive Statistics
The demographic findings for the present study are presented in Table 5. In
Table 5, the first column presents the demographic findings for the overall population
of Los Angels Country Health Survey for people age 65 years and over. The second
column presents the characteristic of the respondents who are age 65 years and over
and has Medicare health insurance benefits. The third and forth columns compare the
characteristics of respondents to the LACHS with FFS versus those with MCO health
insurance coverage.
In the LACHS, 56.5% of the respondents were female and 43.2% were male.
Of these 67.4% were White, 14.4% Hispanic, 10.5% African-American, and 7.4%
Asian/Pacific Islander (Table 5).
The age distribution of Los Angeles County is given in Table 5. In Los
Angeles county 60.7% of the over 65 year old population is between 65 and 75 years
of age, and 39.3% are age 75 years and older. This compares to 61.5% and 38.5%
nationally (HCFA, 2000).
Education. In Los Angeles County, 59% of the population age 65 years and
older has some college education or degree, and 41% with a high school education or
less. In the study population, 60% have some college or a degree, with 40% with a
high school education or less (Table 5).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Table 5
Descriptive Statistics 1997 LA County Health Survey: Comparison o f Los
Angeles County Population Aged 65 Years and Over and the Los Angels County
Health Survey Medicare Population Aged 65 Years and Over
L.A. county Study
health survey sample
age 65+ population FFS HMO
% % % %
Gender
Male 36.3
Female 63.6
Age
65-69 31.8
70-74 28.9
75-79 22.3
80-84 10.9
85+ 6.0
Race-ethnicity
White 67.4
Black 10.5
Hispanic 14.4
Asian 7.4
36.5 38.9 34.7
63.5 61.1 65.3
30.9 29.6 31.9
29.6 27.7 31.0
22.7 22.8 22.6
11.3 13.3 9.8
5.6 6.5 4.8
70.2 43.9 56.1
10.6 38.6 61.4
12.3 43.1 56.9
6.7 61.8 38.2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
Table 5 (continued).
L.A. county Study
health survey sample
age 65+ population
% %
FFS
%
Others 0.4 0.2 0.0
Insurance
Private &/or employer 6.1 6.9
Medicare 84.2 83.4 44.4
Medicare only 26.8 28.5 8.1*
Medicare+Supplement 39.3 42.8 25.7*
Medi-Medi 18.2 19.1 10.6
Medicaid/Medi-Cal only 2.4 2.7 1.5
Uninsured 7.2 0.0 0.0
Managed care
HMO 55.6 55.6
FFS 44.4 44.4
Health status
Excellent 15.0 15.2 15.7
Very good 32.2 33.2 33.1
Good 27.9 27.7 24.0
Fair 19.5 18.8 20.1
Poor 5.4 5.2 7.1
HMO
%
0.0
55.6
23.5*
21.7*
10.5
1.3
0.0
14.7
33.3
30.9
17.7
3.5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
Table 5 (continued).
L.A. county Study
health survey sample
age 65+ population FFS HMO
% % % %
Any chronic condition 52.5 53.3 51.6 54.8
Arthritis 53.1 53.9 52.2 55.5
Diabetes 14.3 14.1 14.0 14.4
Cancer 5.4 5.5 5.9 5.2
Heart disease 18.1 18.0 18.9 17.2
Hypertension 39.8 40.1 39.3 40.7
Lung disease 5.2 5.7 4.2* 6.9*
Kidney disease 1.8 1.9 2.9 0.9
HIV/Aids 0.0 0.0 0.0 0.0
Immigration status
U. S. native 82.0 84.9 42.4 57.6
Naturalized citizen 14.0 12.5 54.8 45.2
Noncitizen 3.9 2.6 45.2 40.9
Economic status
100% FPL 6.8 5.6 4.8 6.2
200% FPL 36.1 35.0 31.1* 38.1*
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
Table 5 (continued).
L.A. county
health survey
age 65+
%
Study
sample
population
%
FFS
%
HMO
%
Income
<$10k 14.0 13.1 20.3* 13.2*
S10K-20K 24.2 24.8 23.0* 36.5*
S20K-30K 12.7 12.4 14.1 16.3
$30K-40K 10.0 10.4 12.7 12.9
S40K-50K 5.5 6.1 7.2 7.8
>$5 OK 13.2 13.9 22.7* 13.2*
Doctor visit in past year 92.2 93.7 93.3 94.1
No hospital stay in past year 82.4 82.6 81.2 83.9
Education
Did not graduate HS 16.3 19.4 13.6 13.6
HS graduate 25.6 20.5 25.3 29.2
Some college or degree 58.1 60.0 61.1 57.2
*p < .05, n < 13
Income. In Los Angeles County, 6.8% of the population has income below
100% federal poverty level (FPL) compared to 5.6% of the study population. In
addition, 36.1% of the Los Angles County population has income below 200% of the
FPL compared to 35% for the study population (Table 5).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
Immigration status. Of Los Angeles County’s population age 65 years and
older, 82% are U. S. natives, 14% are naturalized citizens, and 3.9% are noncitizens.
Within the study population, 84.9% are U. S. natives, 12.5% are naturalized citizens,
and 2.6 are noncitizens (Table 5).
Insurance status. The responses to the LACHS questions regarding health
insurance status yielded 38.7% (409) respondents with FFS, 44% (465) with HMO,
7.1% (75) uninsured. In addition, 6% (64) had employer or private insurance and 4.1%
(43) did not know or refused to answer (Table 5).
The uninsured rate among the senior population of Los Angeles County is
remarkably high compared to the national rate. In Los Angeles County more than 7%
of seniors are uninsured whereas, nationally less than 1% of seniors are uninsured
(HCFA, 1998).
Tests o f the Hypotheses
Research Question A
Are the health risk factors similar between the Senior FFS and MCO
populations in Los Angeles County?
In this study, “health risk factors” are assessed using the need factors of Aday
and Anderson (1974): (a) self-reported health status, (b) chronic conditions, (c)
hospital admission, and (d) having a regular source of care. In order to test the
relationship between the health risks factors of the FFS and MCO populations in Los
Angeles County, a hypothesis and subhypotheses were generated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
Immigration status. Of Los Angeles County’s population age 65 years and
older, 82% are U. S. natives, 14% are naturalized citizens, and 3.9% are noncitizens.
Within the study population, 84.9% are U. S. natives, 12.5% are naturalized citizens,
and 2.6 are noncitizens (Table 5).
Insurance status. The responses to the LACHS questions regarding health
insurance status yielded 38.7% (409) respondents with FFS, 44% (465) with HMO,
7.1% (75) uninsured. In addition, 6% (64) had employer or private insurance and 4.1%
(43) did not know or refused to answer (Table 5).
The uninsured rate among the senior population of Los Angeles County is
remarkably high compared to the national rate. In Los Angeles County more than 7%
of seniors are uninsured whereas, nationally less than 1% of seniors are uninsured
(HCFA, 1998).
Tests o f the Hypotheses
Research Question A
Are the health risk factors similar between the Senior FFS and MCO
populations in Los Angeles County?
In this study, “health risk factors” are assessed using the need factors of Aday
and Anderson (1974): (a) self-reported health status, (b) chronic conditions, (c)
hospital admission, and (d) having a regular source of care. In order to test the
relationship between the health risks factors of the FFS and MCO populations in Los
Angeles County, a hypothesis and subhypotheses were generated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Health Risk Variables
Hypothesis I
The rate of health risk factors is the same for the Senior FFS and MCO
populations in Los Angeles County. This hypothesis includes the following
subhypotheses:
a. The rates of self-reported health status are the same for the Senior FFS
and MCO populations in Los Angeles County.
b. The rates of chronic conditions are the same for the Senior FFS and
MCO populations in Los Angeles County.
c. The rates of hospital admissions are the same for the Senior FFS and
MCO populations in Los Angeles County.
d. The rates of people who report having a regular source of care are the
same for the Senior FFS and MCO populations in Los Angeles County.
Self-reported health status. There is no significant difference in overall self-
reported health status (t = 1.05,/? = .29) observed for the study population. For seniors
enrolled in FFS plans 72.8% report their health as good to excellent, with 27.2%
reporting their health as fair to poor. For seniors enrolled in MCOs, 78.8% report their
health status as good to excellent and 21.1% report fair to poor health status (Table 6).
Given that self-reported health status is often considered a valid indicator of a
person’s actual health status (Long & Marshall, 1999), these findings do not support
concerns that a disproportionate number of sicker people remain in the FFS system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
Table 6
Self-reported Health Status o f Los Angeles County Seniors by Health Insurance Type
Health status
2
1
Very
3 4 5
Excellent
good
Good Fair Poor
Insurance
%
%
% % %
type (») (n)
(n)
(»)
(n) Mean SD
FFS (n — 408) 15.7 33.1 24.0 20.1 7.1 2.7 1.2
(64) (135) (98) (82) (29)
MCO (n = 463) 14.7 33.3 30.9 17.7 3.5 2.6 1.0
(68) (154) (143) (82) (16)
Hypothesis I (a), that the levels of self-reported health status are the same for
the Senior FFS and MCO populations in Los Angeles County, is supported.
Chronic conditions. Overall, in the senior population, 52.4% of the
respondents report having at least one of the several chronic conditions listed.
However, no significant differences were noted between the MCO (54.8%), and FFS
(51.6%) populations who report having at least one chronic condition (x -3 .0 \,p =
.08). Of the seniors who report having a chronic condition, three percent (11
respondents), report that they are not being treated for that condition. There is no
significant difference between the number of FFS and MCO populations, who are not
being treated for their chronic conditions (x2 = 1.47,/? = .22 ) (Table 7).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
Table 7
Prevalence o f Chronic Health Conditions: Los Angeles County Seniors
Reporting Having Chronic Health Conditions by Insurance Type
Self-reported chronic condition (n)
FFS
(409)
% (n)
MCO
(465)
% (n) Sig.
Any chronic condition (466) 51.6 211 54.8 255 NS
Untreated chronic condition (11) 4.4 7 1.2 4 NS
Arthritis (466) 52.2 211 55.4 255 NS
Diabetes (122) 14.0 57 14.1 65 NS
leart disease (156)
18.9 77 17.2 79 NS
Cancer (48) 5.9 24 5.2 24 NS
Lung disease (49) 4.2 17 6.9 32
*
Kidney disease (16) 2.9 12 0.9 4
**
Hypertension (349) 39.3 160 40.7 189 NS
% 2,p < .10, * *x2 ,p < .05, * **%2 ,p <.001
These findings support HI (b) that the rates of chronic conditions are the same
for the Senior FFS and MCO populations in Los Angeles County.
Post hoc analysis of the chronic conditions indicate that there is no significant
difference in occurrence for five of the specific conditions investigated: (a) arthritis,
(b) diabetes, (c) heart disease, (d) cancer, and (d) hypertension)— suggestive findings
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
for one condition, lung disease (% 2 = 3.01,/? = .08), and significant differences for
another kidney disease. The prevalence of kidney disease is significantly higher within
the FFS (2.9%) population than in the MCO population (0.9%) (x2 = 5.28,/? = .02).
Although the findings are significant at the .05 level, the results should be interpreted
with caution due to the small number of respondents (n = 16) reporting this condition.
The findings for the specific chronic conditions thus yield further support for
Hypothesis I (b) that the incidence of chronic conditions is similar between the FFS
and MCO health care systems.
Hospital admission. The hospital admission data were analyzed to investigate
three factors: (a) the over all rate of hospitalization between FFS and MCOs; (b) the
rates of any hospital admission versus no admission between FFS and MCOs; and (c)
the rates of “high,” two or more, admissions between the two systems.
Overall, 17% of the senior population reported having been admitted to the
hospital at least once in the past year. There is no significant difference in the rate of
hospitalization between the FFS (18.8%) and the MCO (16.1%) populations in the
LACHS (x2 = 3.47,/? = .32) (Table 8). There is no significant difference in the rates
of no admissions and any admissions between the FFS and MCO plans (x2 = 1.10,
p = .30). This finding indicates that, contrary to earlier studies (Hill & Brown, 1990),
within the senior population, MCO beneficiaries are not being admitted to hospitals at
a significantly lower rate than within FFS beneficiaries (Table 8).
There is no significant difference in the rates of two or more admissions per
respondent between the respective types of health care coverage (x2 = 0.45,/?= .50).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
Table 8
Hospital Admissions fo r Los Angeles County Seniors by Health Insurance Type
No 3 or more
admissions 1 admit 2 admits admits
Insurance type
% (n) % (n) % (n) % (n)
FFS (n = 409) 81.2 (332) 13.5(55) 2.9 (12)
2.4(10)
MCO (n = 467) 83.9 (390) 9.7 (45) 3.9(18) 2.6(12)
% 2 = 3.47,/? < 0.325
This finding further supports the finding that patients in the FFS regime do not
have a higher rate of hospitalization than those in MCOs.
The findings regarding hospital utilization support Hypothesis I (c) that the
rates of hospital admissions are the same for the Senior FFS and MCO populations in
Los Angeles County.
Regular source o f care. In response to question 31 of the LACHS, whether or
not the respondent has one particular place or health care provider, more than 90% of
all seniors replied that they do have one regular source of care (RSC). However, a
significantly higher percentage of the MCO respondents (96%) than FFS respondents
(91%) stated that they have a regular source of care (x2 = 9.09,p = 0.003) (Table 9).
In determining if the respondents used a private doctor’s office or clinic, it
turns out that a significantly greater proportion of FFS seniors, obtain their health care
services at facilities other than their own private doctor’s offices or clinic. Of FFS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
Table 9
Regular Source o f Care: Whether or Not Los Angeles County Seniors Have a
Regular Source o f Health Care by Insurance Type
Yes, have a regular
source of care
No regular source of care
Insurance type
% (n)
% (n)
FFS (n = 369) 90.8 (335) 9.2 (34)
MCO (n = 465) 95.9 (446) 4.1 (19)
X2 =9.09,/? <0.003
seniors, 9.2% report either “no” or “other” regular source of health care compared to
4.1% of seniors enrolled in MCOs (x2 = 9.09, p = 0.003) (Table 10).
Table 10
Regular Source o f Care for Los Angeles County Seniors by Health Insurance Type
Insurance type
Doctor’s
office
% (n)
County or
community
clinic
% (n)
Hospital
outpatient
clinic
% (n)
Emergency
room
% (n)
Other
% (n)
FFS (n = 332) 88.9 1.8 6.0 0.9 2.4 (8)
(295)
(6)
(20)
(3)
MCO (n - 439) 89.1 2.5 4.8 2.5 1.1 (5)
(391)
(11)
(21)
(11)
X2 =5.45,/? <0.244
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
Hypothesis I (c), that the rate of people who report having a regular source of
care is the same for the Senior FFS and MCO populations in Los Angeles County is
not supported.
Post hoc analysis revealed that there is no significant difference in the rate that
FFS and HMO seniors use hospital outpatient, county or community clinics as their
RSC (x = 0.10,/? = .75). However, MCO seniors report using the emergency room
(ER) at more than twice the rate of FFS seniors RSC (2.5% to 0.9%). Although these
findings are suggestive of a trend (x2 = 2.96, p = .09) of greater use of ERs by MCO
members than by FFS respondents, the results should be interpreted with caution due
to the small number or respondents who report using the ER as their RSC (Table 10).
Quality o f Care Variables
Research question B: Is the quality of care comparable between the FFS and
MCO systems in Los Angeles County?
In this study, quality of care is assessed using the access and utilization
variables of Aday and Anderson (1974). The access variables include: access to and
waiting times for an appointment. The utilization variables include: (a) doctor visits in
the past year, (b) satisfaction, (c) afford medications-treatment, and (d) preventive
services. In order to investigate the relationship between the quality of care in the FFS
and MCO health care systems the following hypothesis and subhypotheses regarding
access and utilization were generated:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
Hypothesis II
The quality of care is the same for Senior FFS and MCO populations in Los
Angles County.
Hypothesis II (a)—access. It is hypothesized that the rate of people who report
having difficulty with access to health care services, is the same for the Senior FFS
and MCO populations in Los Angeles County.
Overall, almost 90% (87.5%) of the senior population finds it somewhat easy
or very easy to access their health care. Within the study population there was no
difference in the rates at which the MCO and FFS respondents report their access to
health care as very easy, somewhat easy, somewhat difficulty or very difficult (x2 =
1.30,/? = .73) (Table 11). There was no significant difference (x2 = 1.04,/? = .31)
between the rates of FFS (11.3%) or MCO (13.6%) respondents who reported that
access was somewhat difficult or very difficult. Hypothesis II (a) is supported.
Waiting time fo r appointment. Waiting time for appointments is comprised of two
components in this analysis. The length of time a patient must wait to obtain an
appointment is termed “wait time.” The length of time a patient must wait at an
appointment before being seen by the doctor is termed “office wait.” Hypothesis II
(b), regarding waiting time for appointments, is tested using the two sub-hypotheses
about “wait time” and “office wait.”
b. It is hypothesized that the waiting times for appointments are the same
for the Senior FFS and MCO population in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
Table 11
Patient Ease o f Access to Their Healthcare Provider fo r Los Angeles County Seniors
by Health Insurance Type
Insurance type
4
Very
easy
% (n)
3
Somewh
at
easy
% (n)
2
Somewh
at
difficult
% (n)
1
Very
difficult
% (n) Mean SD
FFS In = 399)
65.9 22.8 7.8(31) 3.5 (14) 3.51 0.79
(263) (91)
MCO (n = 456) 64.7 21.7 8.8 (40) 4.8 (22) 3.46 0.85
(295) (99)
X2 =1.30, p < 0.730, t = .87, p = 0.39
i. It is hypothesized that the “wait time” for an appointment are
the same for the Senior FFS and MCO population in Los Angeles County.
Within the senior population, almost 80% of the survey respondents report that
they can see their health care provider either with no appointment necessary or within
the same day-less than 2 days. Within the study population, there are no significant
differences in the waiting times for an appointment (x2 = 4.45, p = .49) (Table 12). In
addition, there is no significant difference between FFS and MCO respondents who
have to wait two days or more, and those who wait less than two days for an
appointment (x2 = 0.71, p= .40). Hypothesis II (bi) is supported.
ii. It is hypothesized that the “office wait” times are the same for
the Senior FFS and MCO population in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 0
Table 12
Waiting Time to Get an Appointment fo r Los Angeles County Seniors by Health
Insurance Type
Insurance type
No
appointment
necessary
% («)
Same day,
less than
two days
% (n)
3-6
days
% {n)
1-2
weeks
% {n)
3-4
weeks
% («)
4+
weeks
% (n)
FFS (n = 332) 18.7 67.1(230) 6.4 6.7 0.9 0.9
(64) (22) (23)
(3)
MCO (n = 439) 20.6 62.9 (263) 8.4 5.5 1.7 0.9 (4)
(86) (35) (23)
(7)
X2 =4.45,/? < 0.486
Overall, 95% of seniors are seen within 1 hour or less for their appointments. More
than 80% of the senior respondents reported that they were seen for their appointments
in 30 minutes or less. Another 12% of respondents reported that they were seen
between 30 and 60 minutes. Within the study population there is no significant
difference between the office wait times for the FFS vs. the MCO patients (x2 = 1.74,
p = .78). No significant differences in wait time for appointments were noted for either
short term (1 hour or less) or longer term (1 hour or more) waits between the FFS and
MCO respondents (Table 13). Hypothesis II (b. ii) is supported. Hypothesis II (b) that
FFS and MCO respondents experience similar waiting times is supported.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 1
Table 13
Waiting Time at Appointments fo r Los Angeles County Seniors by Health Insurance
Type
0-30 31-60 1 to 2 3 hours or more
Insurance type
minutes minutes hours
% (n)
% (n) % (n) % (n)
FFS (n = 347)
84.4 (293) 11.8 (41) 3.5 (12) 0.3(1)
MCO (n = 435) 82.1 (357) 12.9 (56) 4.6 (20) 0.5 (2)
X2 = 1.74,/? < 0.783
Utilization. The utilization factors of “quality of care” are analyzed using the
last visit, satisfaction, afford medication/treatments, and preventive services variables
of the LACHS.
Last Visit
c. It is hypothesized that the rate of seniors who have seen a doctor or
other health care provider in the past year is the same for FFS and MCO in Los
Angeles County.
Within the senior population, 94% of the respondents had seen a doctor or
other health care provider in the previous year. Within the study population there was
no significant difference (% 2 = 0.25,p = .61) between the FFS (93.3%) and MCO
(94.1%) populations with respect to having seen a health care provider in the previous
year (Table 14). Hypothesis 11(c) is supported.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 2
Table 14
Doctor Visit in the Past Year: Los Angeles County Seniors Who Reported
Seeing a Healthcare Provider in the Past Year
Insurance type At least one doctor visit No doctor visit
% (n) %(«)
FFS (n = 402) 93.3 (375) 6.7 (27)
MCO (n = 459) 94.1 (432) 5.9 (27)
X2 =025, p <0.615
Satisfaction
d. It is hypothesized that the level of satisfaction with the care that they
have received from their health care providers is the same for the FFS and MCO
populations in Los Angeles County.
Overall, MCO beneficiaries report a significantly higher average level of
satisfaction than do the FFS beneficiaries (t = -2.12,/? = .03). Hypothesis II (d) is not
supported.
In response to the question “Overall, how satisfied are you with the care you
are receiving. . 8 5 % of respondents age 65 years and over are either extremely or
very satisfied with their health care providers. W ithin the study population a higher
proportion of seniors in FFS (46.4%) than in MCOs (39.9%) responded that they are
extremely satisfied; whereas a higher proportion of MCO respondents (43.1%) than
FFS (41.9%) said they are very satisfied. In addition, 14.5% of MCO respondents and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 15
Patient Satisfaction with Their Healthcare Provider fo r Los Angeles County Seniors by Health Insurance Type
1 2 3 4 5
Extremely Very Somewhat Not very Not satisfied
satisfied satisfied satisfied satisfied at all
Insurance type % {n) % {n) % (n) % (n) % (n) Mean SD
FFS (n = 356)
46.4
41.9 9.6 1.7 0.6 1.68 0.76
(165)
(149) (34)
(6) (2)
MCO (n = 436) 39.9 43.1 14.5 2.1 0.5 1.80 0.79
(174) (188) (63)
(9) (2)
t - -2.12, p = .03
104
9.6% of FFS respondents said they are somewhat satisfied. Finally, 2.5% of MCO and
2.3% of FFS respondents are not very satisfied or not satisfied at all (Table 15).
Afford Medication/Treatment
e. It is hypothesized that the rate at which people report that they could
not afford certain medications or treatments is the same the same for the Senior FFS
and MCO population in Los Angeles County.
This hypothesis was further evaluated by testing five subhypotheses regarding
specific health care services.
i. It is hypothesized that the rate that people report that they could
not afford prescription medication is the same the same for the Senior FFS and
MCO populations in Los Angeles County.
ii. It is hypothesized that the rate that people report that they could
not afford mental health care or counseling is the same the same for the Senior
FFS and MCO populations in Los Angeles County.
iii. It is hypothesized that the rate that people report that they could
not afford dental care is the same the same for the Senior FFS and MCO
populations in Los Angeles County.
iv. It is hypothesized that the rate that people report that they could
not afford eyeglasses is the same the same for the Senior FFS and MCO
populations in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
v. It is hypothesized that the rate that people report that they could
not afford seeing a doctor is the same the same for the Senior FFS and MCO
populations in Los Angeles County.
Within the Los Angeles County population aged 65 years and older, more than
18% of seniors reported that they had to forego at least one of the listed services
because they could not afford it. Within the study population there was a significant
difference between the FFS and MCO respondents (x2 = 9.01,/? < 0.003). Of the FFS
respondents, 22.5% reported that they had to forego some needed services because
they could not afford them compared to only 14.6% of the MCO respondents (Table
16). Hypothesis II (e) is not supported.
For each of the specific services mentioned, a higher percentage of FFS than
MCO respondents said that they could not afford the service in the past year. The
differences were significantly higher for three of the services covered: (a) prescription
medicine (FFS 10.0%, MCO 6.1%) (x2 = 4.71,/? = .03); (b) mental health services
(FFS 4.8%, MCO 1.3%) (x2 = 8.77,/? = 1.0); and (c) eyeglasses (FFS 11.1%, MCO
5.4%) (x2 = 9.39,/? = 1.0). The fourth service, dental care, was suggestive of the trend
toward a higher rate of difficulty for the FFS respondents compared to the MCO
respondents (x2 = 2.70,/? = .10), but was not significant at the .05 level. In only one
of the services surveyed, seeing a doctor for a health problem, was there no solidly
significant difference (x2=0.20,/? = .65) between the responses for the FFS and MCO
populations (Table 15). Therefore, Hypotheses II (e. i, ii, iii, iv) are not supported.
Hypothesis II (e.v) is supported.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
Table 16
Financial Difficulty Accessing Health Services: Los Angeles County Seniors Who
Reported Needing Healthcare Service but Could Not Afford It
Insurance type
FFS
(n = 409)
% (n)
MCO
{n = 463)
% {n)
Significance
At least one service or procedure 22.5 (92) 14.6 (68)
***
Prescription medicine 10.0 (41) 6.1 (28)
**
Mental health care or counseling 4.8 (19) 1.3 (6)
***
Dental care 12.3 (50) 8.9 (41)
*
Eyeglasses 11.1 (45) 5.4 (25)
***
Doctor visit for a health problem 5.6 (23) 5.0 (23) NS
* P < .10, **p < .05, ***p <.01
Preventive services. This portion of the LACHS assessed seven different
preventive health services. Two procedures were common to all respondents: colo
rectal examination and HIV tests. Two of the questions were directed at male
respondents: (a) prostate examination and (b) testicular examination. Three questions
pertained to female respondents: (a) pap smear, (b) breast examination, and (c)
mammogram.
f. It is hypothesized that the use of preventive services is the same for the
Senior FFS and MCO population in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
i. It is hypothesized that the rate among all respondents for
receiving a colo-rectal exam is the same the same for the Senior FFS and MCO
population in Los Angeles County.
ii. It is hypothesized that the rate among all Senior respondents for
receiving a blood test for HIV/AIDS is the same the same for the Senior FFS
and MCO population in Los Angeles County.
iii. It is hypothesized that the rate among male senior respondents
for receiving a testicular exam is the same for the Senior FFS and MCO
population in Los Angeles County.
iv. It is hypothesized that the rate among male senior respondents
for receiving a prostate exam is the same for the Senior FFS and MCO
population in Los Angeles County.
v. It is hypothesized that the rate among female senior respondents
for receiving a pap smear is the same for the Senior FFS and MCO population
in Los Angeles County.
vi. It is hypothesized that the rate among female senior respondents
for receiving a physical breast exam is the same for the Senior FFS and MCO
population in Los Angeles County.
vii. It is hypothesized that the rate among female Senior
respondents for receiving a mammogram is the same for the Senior FFS and
MCO population in Los Angeles County.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
Of all the senior respondents, 46% had received a colo-rectal examination.
There was no significant difference for this procedure (x = 2.01, p = .16) between the
FFS (43.3%) and the MCO (48.2%) populations. For the HIV/AIDS blood test, 25%
of the seniors had received the test. A higher percentage of seniors in the FFS group
(27.8%) received the test than in the MCO group (22.3%). Although the difference
was not significant at the .05 level, it is suggestive of a trend of higher HIV testing in
the FFS sector (x2 = 3.17,p = .08) (Table 17). Hypothesis II (f. i) and Hypothesis II (f.
ii) are supported.
Table 17
Preventative Test or Examinations Administered in the Past Two Years: Los Angeles
County Seniors Who Reported Receiving Particular Examinations
Procedure (n)
FFS
% (n)
MCO
% (n)
Significance
Colo-rectal examination (393) 43.4 (174) 48.2 (219) NS
HIV tests (203) 27.8 (106) 22.4 (97)
*
Pap smear (359) 64.4(159) 65.4 (200) NS
Breast examination (444) 75.5 (189) 83.9 (256)
**
Mammogram (438) 72.3 (180) 84.3 (258)
***
Prostate examination (234) 72.4(113) 76.6(121) NS
Testicular examination (127) 43.2 (63) 44.1 (64) NS
* p < .\0 **/?<.05 ***/?< .01
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
Within the male portion of the senior population, 75% of the respondents had
received a prostate examination: (FFS 72.4%, MCO 76.6%) (x2 = 0.71, p = .39). Less
than 50% of male seniors had received a testicular examination (FFS 43.2%, MCO
44.1%) (x2 = 0.03, p = .87). The differences between the FFS and MCO populations
are not statistically significant (at .05) for these procedures (Table 17). Hypothesis II
(f. iii) and HII (f. iv), that the rate for receiving a testicular examination and the rate
for receiving a prostate examination, are the same for the FFS and MCO populations
are supported.
Among the female senior population, 65% had received a pap smear within the
previous two years. The rates for this procedure are within 1% of each other for the
FFS (64.4%) and MCO (65.4%) populations. Hence, there is no significant difference
(X2 = 0.06, p — .81) in the rates for females receiving pap smears between the two
populations (Table 17). Therefore, Hypothesis II (g. v) that the rate of receiving pap
smears was the same for female respondents in FFS and MCO health plans is the same
is supported.
Overall, approximately 80% of senior women had breast examinations and/or
mammograms. There is a no significant difference for the rate of breast examination.
However, there is a significant difference (at .05) between the FFS and MCO rates for
mammograms. For breast examinations, 76.9% of females in FFS compared to 82.8%
of females in MCO plans received the examination (x2 = 6.13,/? = .01). For
mammograms, 72.3% of female respondents in FFS compared to 84.3% in MCOs
received the procedure (% 2 =9.96, p = .0002) (Table 17).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 0
Hypothesis II (g. vi) and Hypothesis II (g. vii) that the rates for receiving
physical breast exams and the rate for receiving a mammogram were the same for FFS
and MCO respondents are partially supported.
Hypothesis II (g) that the use of preventive services was the same for the
senior FFS and MCO population in Los Angeles County is partially supported.
Summary
The findings regarding self-reported health status, chronic conditions, and
hospital admissions support Hypothesis I that the level of health risk factors is the
same for the senior FFS and MCO populations. The findings regarding access, waiting
time and last doctor visit directly support Hypothesis II that the quality of care is the
same for the senior FFS and MCO populations.
The findings on satisfaction did not support Hypothesis II. It should be noted
that the findings on affordability of medication/treatment, though not directly
supporting the hypothesis of equality between FFS and MCO indicate a higher
compliance rate of the MCO system relative to the FFS system (Table 18).
Post Hoc Multivariate Analysis
In order to assess the findings of this study relative to the themes developed by
Aday and Andersen (1974), post hoc multivariate analysis of the health risk and
quality of care variables was conducted. Odds ratios using logistic regression were
obtained. In this analysis, the health risk variables correspond to the predisposing (age,
gender, race-ethnicity), enabling (regular source of care, income and insurance
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
Table 18
Results o f Hypothesis Test o f No Significant Difference between FFS
and MCO Medicare Beneficiaries in Los Angles County
Supported
Health risk variables
Self-reported health status Yes
Presence of chronic condition Yes
Hospital admission in past year Yes
Regular source of care Yes
Quality of care variables
Access to health services Yes
Waiting time for an appointment Yes
Doctor visit in past year Yes
Satisfaction No
Could not afford medication or treatment
Prescription medications No (FFS>MCO)
Mental health care or counseling No (FFS>MCO)
Dental care No (FFS>MCO)
Eyeglasses Yes
Seeing a doctor Yes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 2
Table 18 (continued).
Utilized preventive services
Colo-rectal exam Yes
Blood test for HIV/AIDS Yes
Testicular exam Yes
Prostate exam Yes
Pap smear Yes
Physical breast exam Yes
Mammogram No (MCO > FFS)
supplement), and the need (health status, hospital admissions, and chronic conditions)
variables of Aday and Andersen (1974).
Within the predisposing variables differences between FFS and MCO were
found only in one category of the race-ethnicity variable. A significantly higher
proportion of Asian Medicare beneficiaries are enrolled in FFS than in managed care,
whereas there is no significant difference for the White, Hispanic or Black
beneficiaries (Table 19).
When considering the enabling variables there are several significant
differences between FFS and MCO. FFS enrollees are more than 40% less likely to
have a regular source of care than MCO enrollees {OR = .59). In addition, it was found
that average income is significantly more likely to be higher for the FFS beneficiaries
than MCO members (Table 19).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
Table 19
Odds Ratios for FFS vs. HMO Populations and Outcomes Using Variables
Derived from Aday and Andersen (1974)
Health risk variables
Predisposing Odds ratio Significance
Age 1.02
Gender 1.26
Race/ethnicity
White 0.90
Minority ® 1.20
Hispanic 0.99
Black 0.81
Asian 2.02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 19 (continued).
Enabling
Regular source of care 0.59
***
Income 1.12
***
Insurance supplement
Medicare only 0.31
**
Medicare +Supplement 2.15
***
Medi-Medi 1.35
*
Need
Health status 1.04
Hospital admissions 0.97
Chronic conditions 0.82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
Table 19 (continued).
Quality of care variables
Satisfaction 0.70
Access difficulty 0.99
Affording difficulty 1.65
Doctor visit in past year 1.05
Waiting time 0.95
Wait time 0.96
Preventive services
Pap smear (female) 1.33
Mammogram (female) 0.52 ***
Breast exam (female) 0.95
Testicular exam (male) 1.02
Prostate exam (male) 0.90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
there is no significant difference in the rate for each of the specific conditions
surveyed.
In addition, in order to evaluate the Medicare beneficiaries reporting multiple
chronic conditions, a post hoc analysis of the average number of chronic conditions
per respondent reporting having a chronic condition was calculated. However, no
significant difference (at .05) was found between the average numbers of chronic
conditions per respondent for FFS versus MCO (/ = .39) (Table 20). In order to test if
a difference in the rate of more serious chronic conditions exists between FFS and
MCO, the average number of conditions was calculated controlling for, omitting,
arthritis and hypertension. Again, no significant difference was found between FFS
and MCO in the rate of the selected chronic conditions (Table 20).
Significant differences between FFS and MCO were noted for two of the six
quality of care variables. FFS beneficiaries are significantly more likely (p < .01) to
report lower levels of satisfaction than MCO beneficiaries. In addition, FFS
beneficiaries are significantly more likely (p < .05) to report difficulty affording
various medications or treatments (Table 19).
There is no significant difference between FFS and MCO in the likelihood of
accessing routine preventive services for most of the procedures surveyed. However,
differences were noted for two of the procedures—mammograms and HIV testing.
Female respondents in FFS are half as likely to report having had a mammogram than
their MCO counterparts (OR = 0.52, p < .01). FFS respondents are significantly more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
Table 20
Comparison o f Total and Average Number o f Chronic Conditions for FFS vs. MCO
FFS
n — 405
HMO
n = 462
Arthritis 211 255
Diabetes 57 65
Heart disease 77 79
Cancer 24 24
Lung disease 17 32
Kidney disease 12 4
Hypertension 160 189
Total 558 648
Average all listed conditions1 1.378 1.403
Average omitting arthritis2 0.86 0.85
Average omitting arthritis and hypertension 3 0.46 0.44
No respondents in this sample reported having HIV.
1 t = 0.39; 2 t = 0.42; 31 = 0.43,;? < .05
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
likely to report having received HIV testing (OR = 1.44, P < .05) than MCO
respondents.
Summary
The post hoc findings for the logistic regression analysis are consistent with
and support the contingency table analysis. When considering the enabling factors
(regular source of care, income and insurance supplement), significant differences in
each of these variables were noted. FFS respondents are less likely to report having a
regular source of care than MCO respondents. FFS beneficiaries are more likely to
report higher levels of income. Regarding health insurance coverage, it is more likely
for FFS beneficiaries to have Medicare+Supplement and for Medicare only
beneficiaries to be enrolled in a MCO.
Regarding quality of care, FFS respondents are more likely to report lower
levels of satisfaction and more likely to report having difficulty affording medications
or treatment than MCO respondents. In addition, female FFS beneficiaries are less
likely than MCO beneficiaries to have had a mammogram as part of their preventive
services regimen.
The conclusions and implications regarding these findings are discussed in the
following chapter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
CONCLUSIONS, IMPLICATIONS AND RECOMMENDATIONS
Managed health care organizations have become increasingly important to the
provision of health care services to Medicare and other public health insurance
agencies. This study investigates whether risk selection prevails in a mature managed
health care market. It also addresses whether, within that market, managed care
organizations (MCO) are providing levels and quality of health care services
comparable to those provided in the more traditional fee-for-service (FFS) sector. The
present study utilizes a random sample, population based instrument— the 1997 Los
Angeles County Health Survey— to explore these issues.
The specific research questions addressed in this study are:
Research question A: Are the health risk factors similar between the senior
FFS and MCO populations in Los Angeles County? It is hypothesized that there is no
difference in the health risk factors between the FFS and MCO populations in Los
Angles County.
Research question B: Is the quality of care comparable between the FFS and
MCO systems in Los Angeles County? It is hypothesized that the quality of care is the
same for Senior FFS and MCO population in Los Angles County.
Health Risk Variables
Four variables are used to describe health risk factors: (a self reported health
status, (b) presence of chronic health conditions, (c) hospital admissions in the past
year, and (d) having a regular source of care (RSC).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 0
This study investigates whether the practice of favorable selection into HMOs
and other managed health care organization has continued, or if there is evidence that
the phenomenon of regression toward the mean has served to equalize the health risk
factors between the FFS and MCO paradigms. It is hypothesized that in a mature
managed health care market such as Los Angeles, the prevalence of favorable
selection would no longer be significant. This was posited because: (a) with the
increased presence of Medicare managed care programs, more people can retire into
the managed care programs; (b) chronically ill patients can maintain their relationships
with their physicians after they retire; and (c) the beneficiaries will continue to age
within the managed care program.
As shown by the results of this study, the overall level of favorable selection in
Medicare HMOs in Los Angeles County is insignificant.
Self-reported Health Status
This study’s finding that there is no significant difference in overall self-
reported health status is in contrast to earlier findings that report evidence of risk
selection in managed care (Brown, 1988; Hill & Brown, 1992). This study’s findings
on similar levels of health status between the FFS and MCO systems, indicate that the
phenomenon of favorable risk selection for MCOs may not persist in a mature
managed care market.
Chronic Conditions
The findings on the prevalence of chronic conditions do not suggest a situation
of adverse selection for FFS Medicare. Specifically, although the rate of chronic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
conditions was higher in the MCO regime than in the FFS system, the difference is not
significant. It may be instructive to see if there is a difference in the prevalence of
what might be considered more serious chronic conditions such as heart disease,
cancer, lung or kidney disease. Heart disease and cancer were more prevalent in the
FFS system, although not significantly so. Lung disease was more prevalent in the
MCO system. Although this was not significant at .05 it is suggestive (p < .10) of a
trend toward a higher rate of lung disease in the MCO system. The findings for kidney
disease were the only example of a significantly greater incidence of chronic
conditions within the FFS than MCO system. As noted earlier, the incidence of people
reporting this condition is small so that the implications of this finding are limited.
Medicare beneficiaries with multiple chronic conditions can pose an increased
health risk. Post hoc analysis was carried out to measure the rate of multiple chronic
conditions in the survey population. The analysis was carried out on all of the survey
chronic conditions, and controlling for the more serious conditions (heart disease,
cancer, lung disease and kidney disease). The post hoc analysis of multiple chronic
conditions indicated that there is no significant difference in the incidence of multiple
chronic conditions between FFS and MCO. Hence, the post hoc analysis supports the
hypothesis of no significant favorable selection for MCO over FFS in the Medicare
population.
Hospital Admissions
Although the rate of hospital admission is 2.7% higher for the FFS system than
for MCO, the difference is not significant. There is no significant difference noted in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 2
the various permutations of admissions (no admissions versus one, two or three
admissions). This finding differs from previous findings that MCO organizations
significantly reduce the rate of hospitalization compared to FFS (Luft, 1981; Langwell
& Hadley, 1989, 1990). Present findings support the position of Miller and Luft
(1994) that the rate of hospitalization within FFS is becoming more comparable to that
in the MCO system as hospitals and physicians respond to changes in the market and
Medicare reimbursement practices.
Unfortunately, LACHS data does not contain information on the number of
days of hospitalization. Such information might yield information on the risk or
prevalence of more serious health conditions, which require longer periods of
hospitalization.
Regular Source o f Care
Within the health risk variables, the only findings of significant differences
between FFS and MCO systems was in the rate for no RSC which was 5% higher in
the FFS system than in the MCO system (Table 9). Wholey and others. (1998) found
similar results that access to primary care providers tends to be higher in MCO than in
FFS.
Quality o f Care
The quality of care issue was addressed using access to health care and
utilization of health care services factors. Access was investigated using the ease o f
access and waiting times variables from the Los Angeles County Health Survey.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
Utilization was investigated using the doctor visit, satisfaction, afford medications/
treatment and preventive services variables.
Doctor Visit
Early studies of physician utilization tended to show reduced physician service
use in HMOs compared to FFS. Later studies show little differences between the two
groups (Miller & Luft, 1994). The present study demonstrates comparable physician
utilization between the FFS and MCO systems with approximately 94% of
respondents reporting seeing a physician in the previous year.
Satisfaction
Present findings are consistent with previous work (Alder, 1995), which found
that Medicare beneficiaries typically express high levels of satisfaction with their
health care. Within the study population, the MCO beneficiaries expressed a higher
overall level of satisfaction than the FFS beneficiaries.
Previously, Brown and his colleagues (1993) reported that although both HMO
and FFS beneficiaries rated their care as good or excellent, HMO enrollees were
significantly less likely than FFS patients to rate their care as excellent, or that they
were less satisfied with the quality of care they received (Clement et al., 1994). The
present findings suggest that the situation relative to the ratings of the MCOs has
improved somewhat. Although the finding that a larger proportion of FFS (46%) than
MCO (40%) seniors rate their satisfaction as extremely satisfied is suggestive of a
difference (x2 = 3.04, p = .08), the difference is not significant (at .05). At the other
end of the scale, there is no significant difference between the relatively small
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
numbers of FFS and MCO respondents who were not satisfied with their care. Overall,
MCO enrollees rated their satisfaction significantly higher than FFS beneficiaries
(Table 15).
These findings lend support to the contention that, if Medicare beneficiaries
remain in MCOs as the Medicare managed care market matures, their levels of
satisfaction become more comparable to if not exceed those reported for the FFS
system.
Afford Medication or Treatment
Within the study population, a significantly greater proportion of respondents
from the FFS sector had to forego some treatment or medication because they could
not afford it (Table 16). This finding is consistent with earlier studies that indicated
that patients in MCOs are more satisfied with the financial arrangements of their
health care (Miller & Luft, 1994; Wholey, 1998). These findings differ from those
reported by Clement and others (1994) who found, among other things, that MCO
patients were less satisfied with access to prescriptions.
It is important to note that patients in both FFS and MCOs, reported having to
forego their prescription medication because they could not afford them. Even within
the MCO system, where prescriptions are usually a covered benefit, 6% of the patients
could not afford their medications. Considering the importance of pharmacological
advances in treating, if not curing many of the illness and disabilities of the older
population, this finding indicates that a significant number of Medicare beneficiaries
are being deprived of potential health benefits due to financial constraints.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
The finding in this survey that 6% of the MCO Medicare beneficiaries have to
forego prescription medications is troubling, because in Southern California, most
MCOs ostensibly provide a prescription drug benefit. This barrier to obtaining
necessary medications might be the result of two factors within the MCO paradigm.
The first is the practice of requiring copayments for obtaining prescriptions. With co
payments ranging from $5.00 to $25.00, this can become a financial burden on
Medicare beneficiaries with limited incomes and multiple prescriptions to be financed,
ft has been noted that copayments may push some patients to try to save money by
neglecting to fill prescription for necessary medications (Soumerai & Ross-Degnam,
1990). In addition, the situation may have been exacerbated since the time of the 1997
LACHS. There has been a decline in prescription benefits along with other benefits
offered by Medicare+Choice plans. The rate of beneficiaries in Medicare+Choice who
are covered for prescription medications dropped from 84% in 1999 to 70% in 2001
(Mathematica, 2001).
The second potential barrier to obtaining medications is the implementation of
caps on the maximum annual amount for prescription medication that an enrollee is
allowed. Many MCOs have prescription benefit caps such as $1,000 or $2,000 per
year, as well as the copayment required for each prescription. Twenty-six percent of
Medicare+Choice enrollees have caps of $750 or less. Another 17% have caps of
$1,500 or less (Mathematica, 2001). So, Medicare beneficiaries with medications that
cost more than $100-$200 per month can find themselves with rather daunting health
care expenses.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
Preventive Services
MCOs have consistently been noted to be more effective at delivering
preventive tests, procedures and examinations (Miller & Luft, 1994; Potosky, 1998).
This study found less difference between FFS and MCOs in use of preventive services
than indicated by those earlier studies. However, a significant difference between
females within MCO over those in FFS persists at the rate of both breast examinations
and mammograms (Table 17). This finding is somewhat surprising considering that
Medicare has covered biennial mammograms since 1991.
Of particular note, is the dramatically low rate for both FFS and MCOs of
various cancer screenings (colo-rectal examination, pap smear, and testicular
examination) of Medicare beneficiaries (Table 17). Greenwald (1992) points out that
the most efficient way to reduce the serious consequences and even death due to
cancer is early detection. With up to 55% of male Medicare beneficiaries not receiving
certain routine preventive examinations, a major portion of the elderly population is at
risk for increased disability and mortality due to cancer.
Summary o f Findings
It is difficult to make direct comparison of theses findings to the results of
earlier studies, because of the differences in study methodologies employed. However,
the present study found much less evidence of continued favorable selection than
many of the health measures tested in earlier studies. For example, the Mathematica
study of Competition Demonstration HMOs (Brown, 1988) as well as others (Retchin
et al., 1992; Hill & Brown, 1992; Riley et al., 1996), have found statistically
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127
significant favorable selection on a range of measure such as self- perceived health,
functional disability, and bed days. This finding of limited favorable selection is
consistent with the findings of Dowd and others (1994), Haan (1997), and Mello
(1999).
Although there may be some difference in the health status of the Medicare
FFS and MCO beneficiaries, this difference does not seem to be manifested in the
actual prevalence of chronic conditions or increased rate of hospitalization.
This study found that there were few quality of care differences between the
FFS and MCO health care systems. The areas of most concern—prescription
medication and cancer screening—indicate problems common to both the FFS and
MCO systems— although they were more pronounced for the FFS system.
Policy Implications and Conclusions
The transition from traditional fee-for-service to managed care has
dramatically changed the structure of American health care in the past decade. The
model of prepaid care within integrated systems has become the prevailing system in
many parts of the country. Both state and federally sponsored health insurance
programs have followed the lead of commercial programs into managed care (Retchin
et al., 1995).
Government sponsored programs, through Medicare and Medicaid, deliver
care to more that 70 million people in the United States. With an expanding
beneficiary population and ever increasing expenses, government and policy agencies,
are looking for new programs aimed at containing costs. Because of its success in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
128
controlling the rise in health care costs in the private sector, managed health care held
out particular promise as a mechanism for expanding services and containing costs in
the public sector. However, from the earliest days of the HMO Demonstration Projects
(Brown, 1988) questions were raised about the managed care regime, including: (a)
Did managed health care organizations provide quality of care and access comparable
to the traditional FFS system?, (b) Were the costs savings reported for managed care
due to more efficient service or to the advantage of favorable selection?, and (c) Was
favorable risk selection a characteristic confined to the early stages of the program, or
is there evidence that it is a continuing phenomena?
The present study’s findings on satisfaction have important implications for
Medicare’s policy of encouraging the use of managed care organizations. The finding
that the beneficiaries with Medicare-Only or Medi-Medi coverage are as likely to be
satisfied with their care in MCO as FFS indicates that the Medicare risk program may
be proving beneficial to the most vulnerable of its beneficiaries.
This investigation demonstrated that managed care and FFS system are similar
in their impact on the populations served and the quality of care provided. Despite
these similarities, three issues have emerged which have important implications for
future health care policy initiatives.
First, the finding that 25% of women in the FFS system are not receiving
preventive screenings for breast cancer is a cause for concern. Since 65% of the
Medicare beneficiaries in Los Angeles County and more than 85% of the Medicare
beneficiaries nationally are still in FFS, a large number of eligible Medicare
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
129
beneficiaries are not being routinely screened for breast cancer. On the other hand, the
finding that managed care organizations have in Los Angeles County, achieved a
significant increase in the rate of breast cancer screening, demonstrates that a
conscientious program to increase screenings can achieve demonstrable results over a
broad range of providers.
The present study’s finding that the compliance rate for screening for testicular
cancer in males and colon cancer in both males and females is less than 50% is also
quite alarming. This finding indicates that a large portion of the elderly population is
at unnecessary risk for undetected and hence untreated cancer.
Because of the serious personal and public costs of delayed treatment for
cancer, these findings present important policy opportunities. Increased promotion of
the importance of cancer screening among primary care practitioners and
establishment of suitable standards for increased monitoring of compliance, could be
effective mechanisms for increasing screening rates. Appropriate reimbursement
scales are also necessary to ensure increased screening especially within the FFS
sector. Increased use of public service announcements on the importance of regular
cancer screening and increased promotion of the importance of cancer screening in the
health promotion and wellness campaigns conducted by health care organizations,
may improve public awareness of these important issues.
Second, prescription drugs are an important component of the health care
regimen of senior citizens. It has been noted that Medicare’s benefits package is based
on a model of acute care from the 1960s that is inadequate for the many elderly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
beneficiaries who have chronic illnesses (Iezzoni, Ayanian, Bates & Brustein, 1998;
Iglehart, 1999b). A critical shortcoming is Medicare’s failure to cover outpatient
prescription drugs (Federman, Adams, Ross-Degnan, Soumerai, & Ayanian, 2001;
Moon, 1999;). Medications are important factors in modem health care, especially for
the elderly or chronically ill. However, even those Medicare beneficiaries purchasing
Medigap coverage ,obtain only limited coverage for their prescription drugs. In
addition, many health plans are restricting rather than expanding their prescription
benefits (Moon, 1999; Henry J. Kaiser Foundation, 2001). This study supports those
findings that a significant part of the Medicare population is not able to afford
essential prescription drags. What is more, this situation even exists within the
managed care domain, which purports to cover prescription medications.
Although drags are an integral part of modem health care treatments, they
offer a means of risk selection. Plans that offer prescription drag benefits to their
beneficiaries may find that they attract sicker patients. On the other hand, plans that
restrict their drag benefits will discourage those who need them. This could result in
favorable risk selection for those health plans that restrict their drug benefits.
Therefore, to promote equitable risk selection as well as to provide adequately for the
elderly population, an expanded prescription drag plan should be part of the Medicare
benefits package.
Third, risk adjustment is a critical technical issue in health care reimburse
ment policy. It is important that adjustments to premium payments reflect differences
in health status, and thus the health care needs of beneficiaries (Wilensky &
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
Newhouse, 1999). Since the early Medicare demonstration projects indicated that
there was favorable selection and subsequent overpayment in the Medicare risk
program, HCFA and policy analysts have been searching for appropriate methods to
improve the inadequate payment formula (Newhouse et al., 1997). The study’s
finding that favorable selection may no longer be a significant factor between the FFS
and MCO populations, indicates that Medicare may now be obtaining the savings
intended when the Medicare risk program was implemented.
In addition, Medicare beneficiaries who select managed care as their option
may also be receiving considerable benefits. These include cost savings as well as
enhanced services from their MCOs. These benefits may be realized through improved
access to preventive care, vision and hearing care, and particularly, to prescription
drugs. However, the possibility that these benefits may be eroding, should be an issue
of concern for the current policy makers in the Centers for Medicare & Medicaid
Services (CMS).
In addition, this study has shown that there may be areas common to both FFS
and MCOs where improvement is possible. In the area of cancer screening and in the
provision of prescription drugs, both systems revealed weaknesses that need to be
addressed if we are to achieve true quality and accessibility for all Medicare
beneficiaries.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
Study Limitations and Future Research
Limitations
This study was based on a survey confined to Los Angeles County in
California. As such, its findings cannot necessarily be generalized to the American
population. The study is based on a population survey, so is subject to variability or
error due to self-reports and recall biases.
The LACHS was a telephone survey and so that portion of the population
without telephones was excluded from the study. Although this study is targeted at
seniors age 65 years and older, the frail elderly and those residing in long-term care
facilities, or those who could not participate in an extended telephone survey, were
excluded from the LACHS.
There are numerous differences in copayments and benefits offered by various
Medicare supplemental insurance policies and Medicare managed care programs
(Barents Group LLC, 1999) that are not controlled for in the LACHS. Therefore,
variations in access or utilization that may be a result of difference in benefits within
either system could not be addressed in this study.
Los Angeles County was selected as a “mature managed care market” because
of its extensive managed care penetration over many years. However, this is a cross-
sectional study based on the LACHS of 1997. The Medicare risk program is in a
dynamic phase and subject to dramatic expansions and unexpected contractions (CMS,
2001). Of particular importance, is the Medicare+Choice program implemented in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
133
1997, and the subsequent changes in the Medicare managed care program. The effects
of Medicare+Choice would not be reflected in the results of this study.
Future Research
This study investigated the relationships of two major populations of Medicare
beneficiaries, those in FFS compared to MCO. Future research is needed into the
various subpopulations of Medicare beneficiaries, such as those in poor health or those
with incomes less that twice the federal poverty level regarding issues of access,
utilization or satisfaction. Follow up studies which draw on larger samples of these
vulnerable populations could be valuable in further evaluations of health care service
issues within these groups.
Another area of research, which might be of particular relevance to other large
urban areas, is the issue of access, utilization and satisfaction for racial and ethnic
minorities and immigrant populations.
The issue of access, especially the use of the emergency room as a regular
source of care, by Medicare beneficiaries who are chronically ill, in poor health or in
low income or poverty, might present some insight into improving access or reducing
costs within the Medicare program.
The findings in this study could provide a basis for follow-up longitudinal
studies to evaluate the effects of the changes brought on by Medicare+Choice in Los
Angles County. In addition, follow-up studies could investigate if the program to
enroll more Medicaid beneficiaries in MCOs results in further reductions in the levels
of residual adverse selection in the FFS system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
134
The findings in this study indicate a need for an evaluation of the differential
affects of types of supplemental insurance on access and utilization of health care
services by Medicare beneficiaries. Research into the differential effects of type of
health insurance supplement, such as Medicare +Supplement or Medi-Medi, has on
access, utilization or satisfaction, could yield valuable insight into the performance of
the Medicare system. Some of the findings in the present study indicate that
beneficiaries who cannot afford comprehensive Medicare supplemental coverage, may
not be realizing adequate access to health care services, particularly preventive
services or medications/treatments.
The finding that 6% of the over 65 year olds surveyed are uninsured within an
age group that, nationally, has an insured rate of ostensibly, 99% also warrants further
study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BIBLIOGRAPHY
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
136
BIBLIOGRAPHY
Aday, L.A., & Andersen, R. M. (1974). A framework for the study of access to
medical care. Health Services Research, 9, 208-220.
Aday L. A., Sellers C., & Andersen, R. M. (1981). Potentials of local health surveys:
A state-of-the-art summary. American Journal o f Public Health, 71, 835-840.
Adler, G. S. (1995). Medicare beneficiaries rate their medical care: New data from the
M.C.B.S. Health Care Financing Review, 16, 175-187.
Andersen, R. M. (1974). A framework for the study of access to medical care. Health
Services Research, 9, 208-220.
Andersen, R. M., & Newman, J. F. (1973). Societal and individual determinants of
medical care utilization in the U.S. Milbank Memorial Fund Quarterly, 51, 95.
Anderson, J., Nelson, D., & Wilson, R. (1998). Telephone coverage and measurement
of health risk indicators: Data from the National Health Interview Survey.
American Journal o f Public Health, 88, 1392-1395.
Barents Group LLC. (1997, May). Changes in Medicare HMO payment policy: Effects
on HMO participation, benefits, and Medicare enrollment. Washington, DC:
The Henry J. Kaiser Family Foundation.
Beebe, J. (1988). Medicare reimbursement and regression to the mean. Health Care
Financing Review, 9(3), 9-22.
Blustein, J., & Weiss, L. J. (1998). Visits to specialists under Medicare: Socio
economic advantage and access to care. Journal o f Health Care fo r the Poor
and Underserved, 9, 153-168.
Brown, R. (1988). Biased selection in the Medicare competition demonstrations.
Princeton, NJ: Mathematica Policy Research Inc.
Brown, R., Clement, D., Hill, J., Retchin, S., & Bergeron, J. (1993). Do health
maintenance organizations work for Medicare? Health Care Financing
Review, 15, 7-23.
Burnette, D., & Mui, A. (1999). Physician utilization by Hispanic elderly persons:
National perspective. Medical Care, 37, 362-374.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
137
Carlisle, D. M., Siu, A. L., Keller, E. B. (1992). HMO vs. fee-for-service care of older
persons with acute myocardial infarction. American Journal o f Public Health,
82, 1626-1630.
Carrasquillo, O., Lantigua, R., & Shea, S. (2001). Preventive services among
Medicare beneficiaries with supplemental coverage versus HMO enrollees,
Medicaid recipients, and elders with no additional coverage. Medical Care, 39,
616-626.
Centers for Medicare & Medicaid Services (CMS). (2002). Managed care national
statistics. Retrieved March 12, 2002, from http://www.mcareol.com/factshts/
factnati.htm.
Clement, D.G., Retchin, S. M., Brown, R. S., & Stegall, M.H. (1994). Access and
outcomes of elderly patients enrolled in managed care. Journal o f the
American Medical Association, 271, 1487-1492.
Cohen R.A., Bloom B., Simpson G., & Parsons P. E. (1997). Access to health care.
Part 3: Older adults. Vital Health Statistics, 70(198), 1-32.
Congressional Budget Office. (1995). Managed care and Medicare, CBO
Memorandum, 26 April, 1995. In Lamphere, J., Neuman, P., Langwell, K.,
Sherman, D. The Surge In Medicare Managed Care: An update. Health Affairs
(May-June 1997) 16(3), 127-133.
Cook, A., & McCoy, J. (2001). Early experience under Medicare+Choice: trends in
MCO participation and enrollment (MPR Reference NO: 8565-012).
Washington, DC: Mathematica Policy Research, Inc.
Davis, M., & Burner, S. (1995). Three decades of Medicare: What the numbers tell us.
Health Affairs, 14, 231-243.
Dowd, B., & Feldman, R. (1985). Biased Selection in Twin Cities health plans. In R.
M. Scheffler & L. F. Rossiter (Eds.). Advances in Health Economics and
Health Services Research, 6 (pp. 253-271). Greenwich, NJ: JAI Press.
Dowd, B., Moscovice, I., Feldman, R., Finch. M., & Wisner, C. (1994). Health plan
choice in the twin cities Medicare market. Medical Care 32, 1019-1039.
Eden, J. (1998). Measuring access to care through population-based surveys: Where
are we now? Health Services Research, 33 (Pt. II), 611-624.
Eggers, P. (1980). Risk differential between Medicare beneficiaries enrolled and not
enrolled in an HMO. Health Care Financing Review, 3, 91-99.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138
Ellis R. P., Pope G. C., Iezzoni L.I., Ayanian, J. Z., Bates, D. W., & Burstein, H.
(1996). Diagnosis-based risk adjustment for Medicare capitation payment.
Health Care Financial Review, 17, 101-128.
Epstein, A. M., & Cumella, E. J. (1988). Capitation payment: Using predictors of
medical utilization to adjust rates. Health Care Financial Review, 10, 51-69.
Escarce, J. J., Epstein, K. R., Colby, D. C., & Schwartz, J. S. (1993). Racial
differences in the elderly: Use of medical procedures and diagnostic tests.
American Journal o f Public Health, 83, 948-954.
Farley, P. J., & Monheit, A. C. (1985). Selectivity in the demand for health insurance
and health care. In R. M. Scheffler & L. F. Rossiter (Eds.). Advances in Health
Economics and Health Services Research, 6, 23.1-248. JAI Press: Greenwich.
Federman, A. D., Adams, A. S., Ross-Degnan, D., Soumerai, S. B., & Ayanian, J. Z.
(2001). Supplemental insurance and use of effective cardiovascular drugs
among elderly Medicare beneficiaries with coronary heart disease. Journal o f
the American Medical Association, 286, 1732-1739.
Field Research Corporation (FRC). (1997). San Francisco, CA: Author.
Fowler, F.J. (1993). Survey research methods. Applied Social Research Methods
Series Volume 1. Newbury Park, CA: Sage Publications.
Fowler, F. J. (1995). Improving survey questions. Applied Social Research Methods
Series Volume 38. Newbury Park, CA: Sage Publications.
Frieman, M. P. (1998). The demand for health care among racial/ethnic sub
populations. Health Services Research, 3, 867-890.
Gluck, M. E. & Hanson, K. W. (2001). Medicare Chart Book (2n d Ed.). Washington,
DC: Henry J. Kaiser Family Foundation.
Gold, M., & Stevens, B. (1998). Synopsis and priorities for future efforts. Health
Services Research, 33 (Pt. II), 611-624.
Gomick, M. E., & Anstadt, G. W. (1999). The American health care system. The
New England Journal o f Medicine, 341, 917-921.
Grazier, K., Richardson, W. Martin, D. & Diehr, P. (1986). Factors affecting health
plan choice. Health Services Research, 6, 659-682.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
Greenwald, H. P. (1992). Who survives cancer? Berkeley, CA: University of
California Press.
Groves, R. M., Biemer, P. P., Lyberg, L. E., Massey, & Nichols, W. L. (Eds.). (1988).
Telephone Survey Methodology. New York: John Wiley & Sons, Inc.
Groves, R. M., & Couper, M. P. (1998). Non-response in Household Interview
Surveys. New York: John Wiley & Sons, Inc.
Haan, M., Selby J., Quesenberry, C., Schmittdiel, J., Fireman, B., & Rice, D. (1997).
The Impact of aging and chronic disease on use of hospital and outpatient
services in a large HMO: 1971-1991. Journal o f the American Geriatrics
Society 45, 661-61A.
Health Care Financing Administration (HCFA). (1996). 1996 Statistical Supplement,
Health Care Financing Administration. Health Care Financing Administration
(1998). 1998 Medicare Managed Care Summary Report. Baltimore, MD:
Health Care Financing Administration.
Health Care Financing Administration (HCFA). (1999). 1999 Statistical Supplement.
Baltimore, MD: Health Care Financing Administration.
Health Care Financing Administration (HCFA). (2000). Health U. S. 2000. Baltimore,
MD: Health Care Financing Administration.
Health Care Financing Review (HCFR). (1996). Statistical Supplement (pp. 124-133).
Baltimore, MD: U. S. Department of Health and Human Services.
Hellinger, F. J. (1987). Selection bias in health maintenance organizations: Analysis of
recent evidence. Health Care Financing Review, 9, 55-63.
Henry, J. Kaiser Family foundation, Health Research and Education Trust. (2001). The
Medicare Program: Medicare+Choice. [Fact Sheet #2052-03.]
Hill, J., & Brown, R. (1990). Biased Selection in the TEFRA HMO/CMP Program.
Princeton, NJ: Mathematica Policy Research, Inc.
Iezzoni, F. I., Ayanian, J. Z., Bates, D. W., & Burstin, H. R. (1998). Paying more
fairly for Medicare capitated care. The New England Journal o f Medicine, 339,
1933-1938.
Iglehart, J.K. (1996). The struggle to reform Medicare. New England Journal o f
Medicine, 334, 1071-1075.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
Iglehart, J. K. (1999a). The American Health Care System-Expenditures. The New
England Journal o f Medicine, 340, 70-76.
Iglehart, J. K. (1999b). The American Health Care System-Medicare. The New
England Journal o f Medicine, 340, 317-332.
Jackson-Beeck, M., & Kleinman, J. H. (1983). Evidence for self-selection among
health maintenance organization enrollees. Journal o f the American Medical
Association, 250, 2826-2829.
Kane, R.L. (Ed.). (1998). Managed care as a vehicle for delivering more effective
chronic care for older persons. Journal o f the American Geriatrics Society, 46,
1034-1039.
Kassirer, J. (Ed.). (1997). Quality of care. New England Journal o f Medicine, Special
Supplement. Massachusetts Medical Society.
Kramer, A.M., Fox, P. D., & Morgenstem, N. (1992). Geriatric care approaches in
health maintenance organizations. The Journal o f the American Geriatrics
Society 40, 1055-1067.
Kongstvedt, P. (Ed.). (1989). The managed health care handbook. Gaithersburg, MD:
Aspen Publications.
Kramer, A., Kowalsky, J., Lin, M., Grigsby, J., Hughes, R., & Steiner, J. (2000).
Outcome and utilization differences for older persons with stroke in HMO and
Fee-for-Service systems. The Journal o f the American Geriatrics Society, 4,
726-734.
Lambre, J. M., Defiese, G. H., Carey, T. S., Ricketts, T.C., & Biddle, A. K. (1996).
The effects of having a regular doctor on access to primary care. Medical
Care, 34, 138-151.
Lamphere, J. A., Neuman, P., Langwell, K., & Sherman, D. (1997). The surge in
Medicare managed care: An update. Health Affairs, 16, 127-133.
Langwell, K., & Hadley, J., (1989). Insights from the Medicare competition
demonstrations. Health Care Financing Review, 11, 65-79.
Langwell, K., & Hadley, J. (1990). Evaluation of the Medicare competition
demonstrations. Health Affairs, 2, 74-84.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
141
Leach, L. S., Yip, J. Y., Myrtle, R. C., & Wilber, K. H. (2001). Outcomes among
orthopedic patients in skilled nursing facilities: Does managed care make a
difference? Journal o f Nursing Administration, 31, 527-533.
Lee, A. J., Gehlbach, S., Hosmer, D., Reti, M., & Baker, C. S. (1997). Medicare
treatment differences for blacks and whites. Medical Care, 35, 1173-1189.
Lee Y., & Kasper, J. D. (1998). Assessment of medical care by elderly people: general
satisfaction and physician quality. Health Services Research, 32, 741-758.
Lichtenstein, R., Thomas, J., Adams-Watson, J., Lepokowski, J., & Simone, V.
(1991). Selection bias in TEFRA at-risk HMOs.” Medical Care, 29, 318-331.
Light, D.W. (1998). Keeping competition fair for health insurance: How the Irish beat
back risk-related policies. American Journal o f Public Health, 88, 745-748.
Lillard, L., Rogowski, J., & Kington, R. (1997). Long-term determinants of patterns of
health insurance coverage in the Medicare population. Gerontologist, 37, 314-
323.
Long, M. J., & Marshall, B. S. (1999). The relationship between self assessed health
status, mortality, service use, and cost in a managed care setting. Health Care
Management Review, 24, 20-27.
Los Angeles County Health Survey (LACHS). (1997). Field Research Corporation,
San Francisco, CA
Luft, H. S. (1978). How do Health Maintenance Organizations achieve their ‘savings’?
New England Journal o f Medicine, 298(24), 1505-1510.
Luft, H. S. (1980). Assessing the evidence on HMO performance. Milbank Memorial
Fund Quarterly, 58, 501-536.
Luft, H. S. (1981). Health maintenance organizations: Dimensions o f performance.
NY: John Wiley & Sons Inc.
Lurie, N., Moscovice, I. S., Christianson, J. B., & Popkin, M. K. (1992). Does
capitation affect the health of the chronically mentally ill? Results from a
randomized trial. Journal o f the American Medical Association, 267, 3300-
3304.
Manning, W. G., Leibowitz, A., Goldberg, G., Rogers, W., & Newhouse, J. P. (1984).
A controlled trial of the effect of a prepaid group practice on use of services.
New England Journal o f Medicine, 310, 1505-11.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
142
Manton, K., & Stallard, E. (1992). Analysis of underwriting factor for AAPCC.
Health Care Financing Review, 14, 117-132.
McCombs, J. S., Kasper, J. D., & Riley, G. F. (1990). Do HMOs reduce health care
costs? A multivariate analysis of two Medicare HMO demonstration projects.
Health Services Research, 25, 593-613.
Mello, M. (1999). Market penetration, biased selection, and utilization in Medicare
HMOs. (Doctoral Dissertation, University of North Carolina at Chapel Hill,
1999). Dissertation Abstracts International, 60-08B, 3834.
Menard, S. (1995). Applied logistic regression, Thousands Oaks, CA: Sage
Publications.
Miller, D. C. (1991). Handbook o f Research Design and Social Measurement (5th ed.),
p. 163. Newbury Park, CA: Sage Publications.
Miller, R. H. (1998). Health care organizational change: Implications for access to
care and its measurement. Health Services Research, 33, 653-680.
Miller, R. H., & Luft, H. S. (1994). Managed care plan performance since 1980: A
literature analysis. Journal o f the American Medical Association, 271{ 19),
1512-1519.
Miller, R. H., & Luft, H. S. (1997). Does managed care lead to better or worse quality
of care: A survey of recent studies shows mixed results on managed care plan
perfonnance. Health Affairs, 16, 7-25.
Moon, M. (1999). Will the care be there? Vulnerable beneficiaries and Medicare
reform. Health Affairs, 18, 107-117.
Morgan, R. O., Vimig, B. A., DeVito, C. A., & Persily, N. A. (1997). The Medicare-
HMO revolving door—the healthy go in and the sick go out. The New England
Journal o f Medicine, 337, 169-175.
Myers, D. (1992). Analysis with local census data: Portraits o f change. NY:
Academic Press.
Neuman, P., Maibach, E., Dusenbury, K., Kitchman, M., & Zupp, P. (1998).
Marketing HMOs to Medicare beneficiaries: Do Medicare HMOs target
healthy seniors? Health Affairs, 17, 132-139.
Newhouse, J. P. (1993). Free for all. Cambridge, MA: Harvard University Press.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
143
Newhouse, J. P., Buntin, M. B., & Chapman, J. D. (1997). Risk adjustment and
Medicare: Taking a closer look— The time for Medicare to experiment with
new payment methods to managed care plans is now. Health Affairs, 16, 26-
43.
Nicholls, W. L. (1988). Computer-assisted telephone interviewing: A general
introduction. In Groves, R. M., Biemer, P. P., Lyberg, L. E., & Massey,
Nichols, W.L. (Eds.). Telephone Survey Methodology (pp. 24-49). New York,
NY: John Wiley & Sons, Inc.
Oberlander, J. B. (1997). Managed care and Medicare reform. Journal o f Health
Politics Policy and Law, 22, 595-631.
Osmond, D. H., Vranizan, K., Schillinger, D., Stewart, A. L., & Bindman, A. B.
(1996). Measuring the need for medical care in an ethnically diverse
population. Health Services Research, 31, 551-571.
Pai, C. W., & Clement, D. G. (1999). Recent determinants of new entry of HMOs into
a Medicare risk contract: A diversification strategy. Inquiry 36(1), 78-89.
Potosky, A. L., Breen, N., Graubard, B., Parsons, E. P. (1998). The association
between health care coverage and use of cancer screening tests: Results from
the 1992 national health interview survey. Medical Care, 36, 257-270.
Potosky, A. L., Merrill, R., Riley, G., Taplin, S., Barlow, W., Fireman, B., & Lubitz, J.
(1999). Prostate cancer treatment and ten-year survival among group/staff
HMO and fee-for service Medicare patients. Health Services Research, 34,
525.
Preston, J. A., & Retchin, S. M. (1991). The management of geriatric hypertension in
health maintenance organizations. Journal o f the American Geriatric Society,
39, 683-90.
Price, J. R., & May, J. W. (1985). Selection and the competitive handling of health
plans in a multiple-choice, multiple insurer market. In R. M. Scheffler & L. F.
Rossiter (Eds.). Advances in Health Economics and Health Services Research
6, 249-263. Greenwich: JAI Press.
Retchin, S. M., & Brown, B. (1990). The quality of ambulatory care in Medicare
health maintenance organizations. American Journal o f Public Health, 80,
411-415.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
144
Retchin, S. M., Brown, R. S., Yeh, S. J., Chu, D., & Moreno, L. (1997). Outcomes of
stroke patients in Medicare fee-for-service and managed care. Journal o f the
American Medical Association, 278, 119-124.
Retchin, S. M., Clement, D. G., & Brown, R. S. (1994). Care of patients hospitalized
with strokes under the Medicare risk program. In H. S. Luft (Ed.), HMOs and
the elderly (pp. 167-194). Ann Arbor, MI: Health Administration Press.
Retchin, S. M., Clement, D. G., Rossiter, L. F., Brown, B., Brown, R., & Nelson, L.
(1992). How the elderly fare in HMOs: Outcomes from the Medicare
competition demonstrations. Health Services Research, 27, 651-669.
Retchin, S. M. & Hurley, E. (1995). The revision of government-sponsored health
care. American Journal o f Medicine, 98, 529-530.
Riley, G., Lubitz, J., & Rabey, E. (1991). Enrollee health status under Medicare risk
contracts: An analysis of mortality rates. Health Services Research, 26, 137-
164.
Riley, G., Potosky, A., Klabunde, C., Warren, J., & Ballard-Barbash, R. (1999). Stage
at diagnosis and treatment patterns among older women with breast cancer: An
HMO and Fee-for-Service comparison. Journal o f the American Medical
Association, 281, 720-726.
Riley, G., Rabey, E., & Kasper, J. (1989). Biased selection and regression toward the
mean in three Medicare HMO demonstrations: A survival analysis of enrollees
and disenrollees. Medical Care, 27(4), 337-351.
Riley, G., Tudor, C. M., Chiang, Y., & Ingber, M. (1996). Health status of Medicare
enrollees in HMO and fee-for-service in 1994. Health Care Financing Review,
17, 65-76.
Rogers, W., Wells, K., Meredith, L., Sturm R., & Bumam, M. (1993). Outcomes for
adult outpatients with depression under prepaid or fee-for-service financing.
Archives o f General Psychiatry, 50, 517-525.
Rossiter, F. F., Langwell, K., Tan, T. E., & Rivnyak, M. (1989). Patient satisfaction
among elderly enrollees and disenrollees in Medicare health maintenance
organizations: results from the National Medicare Competition Evaluation.
Journal o f the American Medical Association, 262, 57-63.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
145
Saag, K., Doebbeling, B., Rohrer, J. E., Kolluri, S., Peterson, R., Hermann, M., &
Wallace, R. (1998). Variation in tertiary prevention and health-service
utilization among the elderly: The role of urban-rural residence and
supplemental insurance. Medical Care, 36, 965-976.
Shortell, S., & Hull, K. (1996). The new organization of the health care delivery
system. In H. Sturat, & U. E. Reinhardt (Eds.), Strategic choices for a
changing health care system (pp. 101-148). Chicago, IL: Health Administration
Press.
Siddharthan K. (1991). Health insurance coverage of the immigrant elderly. Inquiry,
28, 403-412.
Soumerai, S., & Ross-Degnam, D. (1990). Experience of state drug benefit programs.
Health Affairs, 9, 36-54.
Thomberry, O.T., & Massey, J. T., (1988). Trends in United States telephone
coverage across time and subgroups. In R. M. Groves, P.P. Biemer, L. E.
Lyberg, J. T. Massey & W. L. Nichols (Eds.). Telephone Survey Methodology
(pp. 109-128). New York, NY: John Wiley & Sons, Inc.
Vimig, B. A., Morgan, R. O., De Vito, C. A., & Persily, M. A. (1998). Medicare
HMOs: Who joins and who leaves. American Journal o f Managed Care, 4,
511-518.
Wagner, E. R. (1989). Types of managed health care organizations. In P. Kongstvedt
(Ed.), The Managed Health Care Handbook (pp. 11-18). Gaithersburg, MD:
Aspen Publications.
Ware, J. E., Bayliss, M. S., Rogers, W. H., Kosinske, M., & Tarlov, A. R. (1996).
Differences in 4-year health outcomes for elderly and poor, chronically ill
patients treated in HMO and fee-for-service systems: Results from the medical
outcomes study. Journal o f the American Medical Association, 276, 1039-
1047.
Ware, J. E., Roberts, W. H., Davies, A. R., Goldberg, S. M., Brook, R. H., & Keeler,
S. D. (1986). Comparison of health outcomes at a Health Maintenance
Organization with those of Fee-for-Service care. Lancet, 1, 1017-1022.
Weiner, J., Dobson, A., Maxwell, S., Coleman, K., Starfield, B., & Anderson, G.
(1996). Risk-adjusted Medicare capitation rates using ambulatory and inpatient
diagnoses. Health Care Financing Review, 17(3), 77-99.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
146
Welch, W. P. (1985a). Regression toward the mean in medical care costs, Implications
for biased selection in health maintenance organizations. Medical Care, 23,
1234-1241.
Welch, W. P. (1985b). Medicare capitation payments to HMOs in light of regression
toward the mean in health care costs. Advances in Health Economics and
Health Services Research, 6, 75-96.
Welch, W. P. (1996). Growth in HMO share of the Medicare market, 1989-1994.
Health Affairs, 15, 201-214.
Wells, K., Hays, R., Bumam, M., Rogers, W., Greenfield, S., & Ware, J. (1989).
Detection of depressive disorder for patients receiving prepaid or fee-for-
service care: Results from the Medical Outcomes Study. Journal o f the
American Medical Association, 262, 3298-3302.
Wholey, D. R., Bums, L. R., & Lavizzo-Mourey, R. (1998). Managed care and the
delivery of primary care to the elderly and the chronically ill. Health Services
Research, 33(2), 322-353.
Wilber, K. H., & Yip, J. (2000). The quality o f post-acute health care in Los Angeles:
Does managed care make a difference? (Final Report to the John Randolph
Haynes and Dora Haynes Foundation). Los Angeles, CA: University of
Southern California.
Wilensky, G. R., & Newhouse, J. P. (1999). Medicare: What’s right? What’s wrong?
What’s next? Health Affairs, 18, 93-103.
Wilensky, G. R., & Rossiter, L. (1986). Patient self-selection in HMOs. Health
Affairs, 5, 66-80.
Williams, D. R., La Vizzomourey, R., & Warren, R. C. (1994). The concept of race
and health status in America. Public Health Reports, 101, 26-41.
Wolinsky, F. D. (1994). Health services utilization among older adults: Conceptual,
measurement, and modeling issues in secondary analysis. Gerontologist, 34,
470.
Yelin, E. H., Criswell, L. A., & Feigenbaum, P. G. (1996). Health care utilization and
outcomes among person with rheumatoid arthritis in fee-for-service and
prepaid group practice settings. Journal o f the American Medical Association,
276, 1048-1053.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
147
Zarabozo, C. (1989). Medicare risk contracting. In P. Kongstvedt (Ed.), The Managed
Health Care Handbook (pp. 249-262). Gaithersburg, MD: Aspen Publications.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
149
The Field Institute
550 Keamy Street
San Francisco, CA. 94108
1997 Los Angeles County Health Survey (LACHS)
Main Questionnaire
What kind of job do you think the Los Angeles County Department of Flealth Services
is doing in safeguarding the health of the people living here? Overall, do you think the
County Department of Flealth Services is doing an excellent, good fair or poor job?
Excellent.........................................................................1
Good...............................................................................2
Fair.................................................................................. 3
Poor.................................................................................4
DON’T KNOW .................................................DK
REFUSED................................................................REF
I am going to read various programs and services that the County Department of
Health Services provides. Based on what you have seen or heard, please tell me how
effective you feel these programs and services are in your community. (READ ITEMS
IN RANDOM ORDER) How effective do you believe these programs and services are
in your community very effective, somewhat effective, not too effective or not at all
effective.
Very Effective.................................................................1
Somewhat Effective........................................................2
Not Too Effective........................................................... 3
Not At All Effective.......................................................4
DON’T KNOW............................................................. DK
REFUSED.....................................................................DEF
(a) Ensuring that foods are free from contamination, such as through
restaurant and produce inspections. 1 2 3 4 DK REF
(b) Protecting the public from exposure to toxic chemicals and other
hazardous materials. 1 2 3 4 DK REF
(c) Ensuring safe drinking water. 1 2 3 4 DK REF
(d) Protecting the public from the spread of communicable diseases, such as
AIDS, hepatitis, and tuberculosis. 1 2 3 4 DK REF
(e) Collecting community health data, such as births, causes of death and
monitoring health trends. 1 2 3 4 DK REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
150
(f) Operating public health clinics and hospitals.
1 2 3 4 DK REF
(g) Providing services to seriously ill children
1 2 3 4 DK REF
(h) Providing emergency and trauma services.
1 2 3 4 DK REF
3. Overall, how easy or difficult is it for you to get medical care when you need
it? Would you slay it is very difficult, somewhat difficult, somewhat easy or
very easy?
VERY DIFFICULT............
SOMEWHAT DIFFICULT
SOMEWHAT EASY.........
VERY EASY......................
DON’T KNOW..................
REFUSE..............................
IF VERY OR SOMEWHAT DIFFICULT, ASK:
4. Have your problems in getting medical care when you need it gotten less
difficult, or stayed about the same over the past 12 months.
MORE DIFFICULT...................................................1
LESS DIFFICULT.................................................... 2
STAY THE SAME.................................................... 3
DON’T KNOW ......................................................DK
REFUSED.............................................................REF
.....1
.... 2
....3
.... 4
.DK
REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
151
5. When was you last visit to see a health care provider? By health care provider I
mean a doctor, nurse, or physician’s assistant. What month and year was this?
(a) ENTER M O N T H :_______________________
January 1 August 8
February 2 September 9
March 3 October 10
April 4 November 11
May 5 December 12
June 6 DON’T KNOW DK
July 7 REFUSED REF
(b) ENTER YEAR (LAST 2 DIGITS): 19___________________
DON’T KNOW DK
REFUSED REF
6. IF 1997 OR DON’T KNOW OR REFUSED TO 5b, ASK:
How may times have you seen a health care provider in the pas 3 months?
Do not count telephone calls or hospital stays. ___________ TIMES
DON’T KNOW..................................................... DK
REFUSED............................................................. REF
7. Thinking about the past year.. ..during the past 12 months was there ever a
time when you needed any of the following but didn’t get it because you could
not afford it? (READ IN RANDOM ORDER)
Yes No Don’t Know Refused
a. Prescription medicine 1 2 DK REF
b. Mental health care or counseling 1 2 DK REF
c. Dental care (including check-ups) 1 2 DK REF
d. Eyeglasses 1 2 DK REF
e. Seeing a doctor for a health problem 1 2 DK REF
RECORD GENDER OF RESPONDENT: MALE............................... 1
FEMALE.........................2
REFUSED.......................3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
152
9. What is your age?________________ REFUSED............. REF
IF REFUSED, ASK:
9b. We don’t need to know exactly, but generally speaking are you between
ages.
(READ CATEGORIES)?
18-24...............................................................................1
25-29.............................................................................. 2
30-39 3
40-49.............................................................................. 4
50-64.............................................................................. 5
65 or older...................................................................... 6
REFUSED............................................................... REF
10. How often do you exercise or participate in an active physical sport, such as
waling, running, jogging, swimming, bicycling or partake in any other type
of physical exercise that makes you sweat or makes your heart beat faster?
(READ LIST)
ALMOST EVERY DAY.............................................. 1
2-3 TIMES A W EEK................................................... 2
ONCE A WEEK............................................................3
2-3 TIMES A MONTH................................................. 4
ONCE A MONTH........................................................ 5
OCCASIONALLY, BUT LESS THAN
ONCE A MONTH........................................................ 6
NEVER.......................................................................... 7
DON’T KNOW.........................................................DK
REFUSED............................................................... REF
IF EVER EXERCISE OR PARTICIPATE IN AN ACTIVE PYSICAL
SPORT, ASK:
11. When you do exercise or participate in a physical sport, how long do you
usually keep at it on each occasion? (IF NECESSARY.) How many hours or
minutes?
HOURS
MINUTES
DON’T KNOW.........................................................DK
REFUSED............................................................... REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
153
12. How tall are you?______________FEET_________ INCHES
DON'T KNOW..........................................................DK
REFUSED................................................................REF
13. What is your weight?____________ LBS.
DON'T KNOW..........................................................DK
REFUSED................................................................REF
Do you consider yourself to be overweight, underweight, or about average for
your height?
OVERWEIGHT............................................................1
UNDERWEIGHT......................................................... 2
ABOUT AVERAGE.................................................... 3
DON'T KNOW..........................................................DK
REFUSED................................................................REF
15. What is your current employment status.. .that is, are you working full-time,
working part-time, unemployed, retired, a homemaker or keeping house,
disable or in school? (ANSWER CAN BE A MULTIPLE)
EMPLOYED FULL-TIME............................................. 1
EMPLOYED PART-TIME............................................2
UNEMPLOYED..............................................................3
RETIRED........................................................................ 4
HOMEMAKER/KEEPING HOUSE............................ 5
DISABLED...................................................................... 6
IN SCHOOL.................................................................... 7
DON'T KNOW.............................................................DK
REFUSED...................................................................REF
IF EMPLOYED FULL-TIME OR PART-TIME, ASK:
16. Is your business or organization mainly in manufacturing, retail trade,
wholesale trade, agriculture, health care, government, education, entertainment
or tourism, banking or financial services, telecommunications or utilities,
construction, household work, professional services (other than health care) or
something else?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
154
17. Are you covered by health insurance or some other kind of heath care plan,
including health insurance obtained through employment or purchased
directly, or any government or military programs such as Medicare, Medi-
Cal also know as Medicaid, ChampU.S., ChampV.A., and the Indian Health
Service?
YES......................................... 1
NO...........................................2
DON'T KNOW..................DK
REFUSED.........................REF
IF YES ASK:
18. Are you currently covered for health insurance under Medicare, the
government’s health insurance program for the elderly and disabled?
YES.......................................... 1
NO............................................2
DON'T KNOW...................DK
REFUSED..........................REF
19. Is your current health insurance coverage a fee-for-service plan that allows you
to go to almost any doctor or hospital or is it an HMO, PPO or another type of
plan that directs you to a list of doctors and hospital who are on the plan?
FEE-FOR-SERVICE 1
HMO/PPO//DIRECT PLAN/KAISER 2
DON'T KNOW DK
REFUSED REF
20. Are you currently covered for health insurance through your own or some
other family member’s employer, labor union, trade association or business?
YES 1
N O ...............................................................................2
DON'T KNOW DK
REFUSED............................................................ REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
155
21 Are you currently covered for health insurance through your own or some
other family member’s military insurance program like ChampU. S. or VA
coverage?
YES.......................................................... 1
N O ..................................................................................2
DON'T KNOW..........................................................DK
REFUSED................................................................REF
22. Are you currently covered for health insurance through a separate policy that
you or some their family member bought directly from an insurance provider?
YES................................................................................. 1
N O ..................................................................................2
DON'T KNOW..........................................................DK
REFUSED................................................................REF
23. Are you currently covered for health insurance under Medi-Cal, also known as
Medicaid, the government’s health insurance program for people on public
assistance or welfare?
YES.................................................................. ..1
N O ..................................................................................2
DON'T KNOW..........................................................DK
REFUSED................................................................REF
IF MEDI-CAL (Q23 = YES), ASK:
24. Can you go to any doctor who will accept Medi-Cal or is it an HMO, PPO or
other type of plan that directs you to a list of doctors and hospitals who are on
the plan?
ANY DOCTOR...........................................................1
HMO/PPO/DIRECTED PLAN/KAISER.................2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
25. Do you know the name of your health plan?
YES, KNOWS NAME............................................... 1
NO, DOESN’T KNOW NAME ........................ 2
REFUSED ...................................................... REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
156
26. During the past 12 months, have you had any periods when you had no health
insurance and you were not covered under any government health insurance
program, like Medicare or Medi-Cal?
YES................................................................................. 1
N O ..................................................................................2
DON'T KNOW..........................................................DK
REFUSED ...................................................... REF
IF NO, NOT INSURED FROM Q17, ASK:
27. When you get sick and need medical care, who pays for it—you, a family
member, relative or friend, do you not pay and try to work off the debt, do
you try to find free care, do you go without medical care, or what?
YOU/FAMILY/RELATIVE/FRIEND........................ 1
WORK OFF THE DEBT.............................................2
TRY TO FIND FREE CARE.......................................3
GO WITHOUT MEDICAL CARE............................. 4
OTHER (SPECIFY)..................................................... 5
DON’T KNOW.........................................................DK
REFUSE...................................................................REF
28. When was the last time you were covered by either government or private
health insurance or a health care plan—within the past 6 months, between 6
months and 1 year ago, between 1 and 3 years ago, 3 or more years ago, or
have you never had health insurance?
WITHIN PAST 6 MONTHS........................................1
6 MONTHS-1 YEAR AGO.........................................2
1-3 YEARS AGO..........................................................3
3 OR MORE YEARS AGO.........................................4
NEVER INSURED...................................................... 5
DON’T KNOW.................................... DK
REFUSED................................................................REF
THERE IS NO Q29
30. In the past 12 months, how may times, if any, were you admitted as an
overnight patient in a hospital or nursing home?
_____________ TIMES
DON’T KNOW .........................................................DK
REFUSED................................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
157
31. When you are sick or want advice about your health, is there one particular
place or health provider to whom you go most often?
YES................................................................................. 1
N O ..................................................................................2
DON’T KNOW .........................................................DK
REFUSED....................... REF
IF NO, ASK:
34. Is that because you have more than one place to go or is it because you have no
regular place to go?
MORE THAN ONE PLACE.................................... . 1
NO PLACE TO GO...................................................... 2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF MORE THAN ONE PLACE, ASK:
33. Is there a particular place that you go more often than any other place?
YES................................................................................. 1
N O ..................................................................................2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF NO PLACE TO GO, ASK:
34. I am going to read some reason why people do not always have a regular place
to go when they are sick. For each, please tell me whether this is or is not a
reason why you don’t have a regular place to go when you are sick. (READ
ITEMS IN RANDOM ORDER)
YES................................................................................. 1
N O ..................................................................................2
DON’T KNOW .........................................................DK
REFUSED................................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
158
35. I’m healthy and don’t need a doctor.
I don’t like or trust doctors.
I don’t know where to go for help.
My previous doctor is no longer available to me.
I don’t know any doctors who can speak to me in my own language.
It costs too much to go to a doctor regularly.
I don’t have a way of easily getting to the doctor.
IF YES TO Q. 31 OR YES TO Q.33, ASK:
36. Is this place or health care provider located in Los Angeles County or
someplace outside Los Angeles County?
INL.A. COUNTY.........................................................1
OUTSIDE L.A. COUNTY...........................................2
DON’T KNOW.................................. DK
REFUSED................................................................REF
37. Where is it located—in another county in California, another state other than
California, or outside the U.S.?
OTHER COUNTY........................................................1
OTHER STATE............................................................2
OTHER COUNTRY.................................................... 3
DON’T KNOW .........................................................DK
REFUSED................................................................REF
IF OTHER COUNTY, STATE OR COUNTRY, ASK:
37/38/39. Which other (county) (state) (country)?
COUNTY CODE_______
STATE CODE_________
COUNTRY CODE_____
DON’T KNOW DK
REFUSED REF {SKIP TO Q.54}
IF IN L.A. COUNTY, ASK:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
159
40. Which of the following best describes the place you usually go when you are
sick—a doctor’s office, a county or community clinic, a hospital outpatient
clinic, an emergency room, a healer other than a medical doctor, or someplace
else? IF NECESSARY READ: By healer, we mean a curandero (cue-en-dair-
o), an acupuncturist, an herbalist, or some other type of healer.
DOCTOR’S OFFICE/KAISER.................... 1
COUNTY OR COMMUNITY CLINIC 2
HOSPITAL OUTPATIENT CLINIC 3
EMERGENCY ROOM................................ 4
A HEALER OTHER THAN AN M.D 5
OTHER (SPECIFY)____________________
DON’T KNOW ......................................... DK
REFUSED ...............................................REF
IF EMERGENCY ROOM, ASK:
41. I am going to read some reason why some people use the emergency room for
their regular health care needs. Please tell me whether each is or is not a reason
why you use the emergency room for your regular health care needs. (READ
ITEMS IN RANDOM ORDER)
IS....1 IS NOT....2 DON’T KNOW....DK REFUSED REF
a. I don’t know where else to go.
b. I don’t have health insurance.
c. It’s convenient.
d. I’m familiar with it and have received care there before.
e. I’m covered by Medi-Cal and other doctors or clinics don’t accept it.
f. It takes too long to get an appointment at a regular doctor’s office or
clinic.
g-
It costs too much to go to a doctor’s office or clinic.
42. What is the name of the hospital you go to?
HARBOR-UCLA MEDICAL CENTER .... 1
LA COUNTY/USC MEDICAL CENTER .2
MARTIN LUTHER KING/DREW
MEDICAL CENTER............................... 3
OLIVE VIEW MEDICAL CENTER 4
OTHER...........................................................5
DON’T KNOW .......................................DK
REFUSED ...............................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
160
IF A COMMUNITY CLINIC OR HOSPITAL OUTPATIENT CLINIC, ASK:
43. What is the name of the (community clinic) (hospital outpatient clinic)?
CLINIC/HOSPITAL CODE_________
OTHER (SPECIFY)_________
DON’T KNOW......................................... DK
REFUSED................................................ REF
IF Q42 = OTHER, DK OR REFUSED REF, ASK:
44. Do you happen to know the nearest cross-streets or nearest intersections
where this (clinic) (hospital) is located?________________________________
DON’T KNOW......................................... DK
REFUSED................................................REF
45. How long have you been going to this (doctor’s office) (clinic) (hospital)
(place)? How many years or months? _________ years__________ months
DON’T KNOW......................................... DK
REFUSED................................................ REF
46. If your were sick and you needed to see a health care provider at this (doctor’s
office) (clinic) (hospital) (place), how long doe it usually take you to get and
appointment? (READ CATEGORIES)
NO APPROINTMENT IS NECESSARY
SAME DAY LESS THAN 2 DAYS ....2
....3
....4
....5
DK
3-6 DAYS...
1-2 WEEKS
3-4 WEEKS
DON’T KNOW
REFUSED....... REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47. About how much time does it usually take you to get from your home to this
(doctor’s office) (clinic) (hospital) (place)?(READ CATEGORIES)
LESS THAN 15 MINUTES...................................................1
15-30 MINUTES.................................................................... 2
31-60 MINUTES.................................................................... 3
1-2 HOURS.............................................................................4
MORE THAN 2 HOURS............................ 5
DON’T KNOW .................................................................. DK
REFUSED......................................................................... REF
48. How do you usually get to the (doctor’s office) (clinic) (hospital) (place)? Do
you usually drive yourself, does someone else drive you, do you take public
transit, walk, take a cab or what?
DRIVE YOUSELF..................................................................1
ANOTHER DRIVES..............................................................2
PUBLIC TRANSIT................................................................ 3
WALK..................................................................................... 4
c a b ............................................... :.........................................5
OTHER.................................................................................... 6
DON’T KNOW .................................................................. DK
REFUSE......................................................................... DREF
49. Once you get to this (doctor’s office) (clinic) (hospital) (place), how long do
you usually have to wait to see the health
0-30 MINUTES....................................................................1
31-60 MINUTES................................................................. 2
BETWEEN 1 AND 2 HOURS........................................... 3
3 TO 5 HOURS................................................................... 4
6 OR MORE HOURS.........................................................5
DON’T KNOW ............................................................... DK
REFUSED............................................................... REF
50. When you go to this (doctor’s office) (clinic) (hospital) (place), do you
normally see the same health care provider each time?
YES........................................................................................... 1
N O ............................................................................................2
DON’T KNOW.................................................................. DK
REFUSED......................................................................... REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
162
51. Think about you last visit to this (doctor’s office) (clinic) (hospital) (place),
did you have any trouble talking to the health care provider because he or she
did not speak to you in you own language?
YES 1
NO 2
DON’T KNOW............................................................DK
REFUSED....................................................................REF
52. Overall, how satisfied are you with the care you are receiving from this
(doctor’s office) (clinic) (hospital) (place)—extremely satisfied, very satisfied,
somewhat satisfied, not very satisfied, or not satisfied at all?
1
2
3
4
5
.DK
REF
53. Do you got to this same (doctor’s office) (clinic) (hospital) (place) when you
need routine preventive care such as a physical exam or check-up?
YES 1
NO 2
DON’T RECEIVE THIS TYPE OF CARE................ 3
DON’T KNOW............................................................DK
REFUSED....................................................................REF
54. Thinking now about the past two years... During the past two years, have you
had any on the following tests or exams administered to you by a doctor, nurse
or other health professional? (READ APPLICABLE ITEMS IN RANDOM
ORDER)
YES....1 N O ....2 DON’T KNOW....DK REFUSED....REF
a. A colo-rectal exam
b. A blood test for HIV infection or AIDS
c. (IF FEMALE) a pap smear, that is, a scraping for the cervix
d. (IF FEMALE) a physical breast exam by a doctor, nurse or other health
professional
e. (IF FEMALE AND AGE 40 OR OLDER) a breast exam called a
mammogram
EXTREMELY SATISFIED
VERY SATISFIED.............
SOMEWHAT SATISFIED.
NOT VERY SATISFIED....
NOT SATISFIED AT ALL.
DON’T KNOW...................
REFUSED...........................
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
163
g........................................(IF MALE) a testicular exam
IF FEMALE UNDER AGE 50, ASK:
55. In the past year did you have a baby?
YES................................................................................. 1
N O .................................................................................. 2
DON’T KNOW......................................................... DK
REFUSED................................................................REF
IF YES, ASK:
56. Did you receive any pre-natal care for your pregnancy?
YES................................................................................. 1
N O .................................................................................. 2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF YES, ASK:
57. At what month of your pregnancy did you receive your first pre-natal care
visit?
FIRST MONTH............................................................. 1
SECOND MONTH.......................................................2
THIRD MONTH........................................................... 3
FOURTH MONTH..................................................... .4
FIFTH MONTH............................................................ 5
SIXTH MONTH............................................................ 6
SEVENTH MONTH.....................................................7
EIGHT MONTH........................................................... 8
NINTH MONTH...........................................................9
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
164
58. In the past year, have you ever been treated or received any medical services
from the Los Angeles County Department of Health Services facilities, such as
a public hospital or health center?
YES...........................................................................................1
N O ............................................................................................2
DON’T KNOW ................................................................. DK
REFUSED ........................................................................ REF
59. Based on what you have seen or heard, please tell me whether you agree or
disagree with the following statements about the medical car provided at
hospitals, clinics and health centers operated by the county. (READ ITEMS IN
RANDOM ORDER) Based on what you’ve seen or heard, do you agree or
disagree?
AGREE 1 .... DISAGREE 2 NO OPINION 3
a. People have to wait too long to get care at County health facilities.
b. The quality of care at County health facilities is about the same as that
provided in most private health facilities.
c. The doctors and nurses at County health facilities are generally friendly
and helpful.
d. The people who work at the front desk at County health facilities don’t
seem to care much about patients.
60. In general, would you say your health is excellent, very good, good, fair or
poor?
EXCELLENT.......................................................................... 1
VERY GOOD..........................................................................2
GOOD........................................................ 3
FAIR........................................................................................ 4
POOR...................................................................................... 5
DON’T KNOW ................................................................. DK
REFUSED ........................................................................ REF
61. During the past 3 months, about how many days did illness or injury keep you
I bed more than half a day? This includes days while you were an overnight
patient in a hospital (IF NECESSARY) Just you best estimate.
DAYS
DON’T KNOW ................................................................. DK
REFUSED ........................................................................ REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
165
62. As far as you know, do you have any on the following medical conditions or
problems? (READ LIST IN RANDOM ORDER) Dou you have
AFTER EACH “YES” RESPONSE IN Q.62, ASK IMMEDIATELY:
(a-h) Are you currently being treated by a physician for
Q62 Q63
Yes No Don’t know Yes No Don’t know
Refused Refused
a. Arthritis 1 2
REF
DK 1 2 DK REF
b. Diabetes or sugar in the 1 2 DK 1 2 DK REF
blood REF
c. Heart disease 1 2
REF
DK 1 2 DK REF
d. Cancer 1 2
REF
DK 1 2 DK REF
e. Kidney disease 1 2
REF
DK 1 2 DK REF
f. HIV infection or AIDS 1 2
REF
DK 1 2 DK REF
h. High blood pressure or
hypertension 1 2
REF
DK 1 2 DK REF
64. How often you use seat belts when you drive or ride in a car or track? Do you
use them always, nearly always, sometimes, seldom or never?
ALWAYS.......................................................................1
NEARLY ALWAYS.................................................... 2
SOMETIMES................................................................3
SELDOM....................................................................... 4
NEVER...........................................................................5
DON’T DRIVE OR RIDE............................................6
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
166
65. Do you currently smoke cigarettes, cigars, a pipe or chew smokeless tobacco?
(IF YES: WHICH ONES(S)?) (ANSWER CAN BE A MULTIPLE)
NO, NON-SMOKER .............................................. 1
YES, CIGARETTES.................................................... 2
YES, CIGARS...............................................................3
YES, PIPE..................................................................... 4
YES, SMOKELESS TOBACCO................................ 5
DON’T KNOW.........................................................DK
REFUSED ...............................................................REF
IF SMOKES CIGARETTES, ASK:
66. How many cigarettes on average do you smoke per day? (IF NECESSARY)
Just you best estimate.
PER DAY
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
67. Did you ever smoke cigarettes, cigars, a pipe or chew smokeless tobacco at
least once a week? (IF YES) Which one(s) (ANSWER CAN BE A
MULTIPLE)
NO, NON-SMOKER.....................................................1
YES, CIGARETTES.................................................... 2
YES, CIGARS...............................................................3
YES, PIPE..................................................................... 4
YES, SMOKELESS TOBACCO.................................5
DON’T KNOW ........................................................DK
REFUSED................................................................REF
68. How often are you around people who smoke in your home—all of the time,
most of the time, only occasionally or never?
ALL OF THE TIME......................................................1
MOST OF THE TIM E................................................. 2
ONLY OCCASIONALLY........................................... 3
NEVER...........................................................................4
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
167
69. How often are you around people who smoke outside your home, such as at a
work place, school or church—all of the time, most of the time, only
occasionally or never?
ALL OF THE TIME......................................................1
MOST OF THE TIM E................................................. 2
ONLY OCCASIONALLY...........................................3
NEVER.......................................................................... 4
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
70. Do you have occasion to drink alcoholic beverages such as beer, wine or
liquor, or are you a total abstainer?
YES, DRINK ALCOHOL............................................ 1
NO, ABSTAINER........................................................ 2
DON’T KNOW ........................................................DK
REFUSED .......................................................... REF
IF YES, ASK:
71. How often did you drink any alcoholic beverages I the past 12 months on
average?
ALMOST EVERY DAY.............................................. 1
4-5 TIMES A W EEK................................................... 2
2-3 TIMES A W EEK................................................... 3
ONCE A WEEK............................................................4
2-3 TIMES A MONTH................................................ 5
ONCE A MONTH.........................................................6
OCCASIONALLY BUT LESS
OFTEN THAN ONCE A MONTH.....................7
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
168
72. If a drink is considered one can or bottle of beer, one glass of wine or cocktail
or shot of liquor, on the days that you drink alcoholic beverage, how many
drinks do you typically have per day, on average? (IF NECESSARY) Just you
best estimate.
___________________ drinks
DON’T KNOW.........................................................DK
REFUSED................................................................REF
73. So that the county can help prevent the spread of AIDS, we need to know more
about the sexual practices and drug use patterns of the general public in Los
Angeles County. Some of these questions are rather personal. If you prefer no
to answer a question, please tell me and I well simply go on to the next
question. We appreciate you cooperation in answering these questions.
74. Have you had sex with anyone in the past 2 months?
YES 1
NO 2
DON’T KNOW .........................................................DK
REFUSED................................................ REF
IF YES, ASK:
75. In the past 12 months, did you or your partner us a condom all the time, most
of the time, some of the time, rarely or never?
ALL OF THE TIME......................................................1
MOST OF THE TME................................................... 2
SOME OF THE TIM E................................................. 3
RARELY....................................................................... 4
NEVER.......................................................................... 5
DON’T KNOW ........................................................DK
REFUSED .............................................................. REF
76. In the past 12 months, how many different people have had as sexual partners?
____________partners
DON’T KNOW .........................................................DK
REFUSED................................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77. In the past 12 months, has your sexual activity been only with men, only with
women, or with both men and women?
MEN ONLY...................................................................1
WOMEN ONLY............................................................2
BOTH............................................................................. 3
DON’T KNOW............................................................DK
REFUSED ...............................................................REF
78. Which of the following best describes your sexual orientation—are you (a)
heterosexual or straight, (b) gay or lesbian, or (c) bisexual?
(a) HETEROSEXUAL/STRAIGHT/ “NORMAL” 1
(b) GAY/LESBIAN/HOMOSEXUAL...................2
(c) BISEXUAL.........................................................3
(d) TRANSEXUAL................................................. 4
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
79. Have you ever used illegal drugs or prescription drugs that were not prescribed
to you?
YES..................................................................................1
N O ...................................................................................2
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
IF YES ASK:
80. Was this within the past 12 months?
YES..................................................................................1
N O ...................................................................................2
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
170
81. How often have you taken illegal drugs or used prescription drugs that were
not prescribed to you in the past 12 months? (READ LIST)
ALMOST EVERY DAY.............................................. 1
4-5 TIMES A W EEK................................................... 2
2-3 TIMES A WEEK................................................... 3
ONCE A WEEK............................................................4
2-3 TIMES A MONTH................................................ 5
OCCASIONALLY BUT LESS THAN ONCE
A MONTH.............................................................6
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
82. In the past 12 months have you ever used needles to inject drugs such as heroin
or cocaine?
YES................................................................................. 1
N O ..................................................................................2
DON’T KNOW ........................................................DK
REFUSED .............................................................. REF
F YES ASK:
83. In the past 12 months have you ever shared needles with someone else or used
someone else’s needles?
YES................................................................................. 1
N O ..................................................................................2
DON’T KNOW ........................................................DK
REFUSED ...............................................................REF
Now, some questions about yourself for classification purpose.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
171
84. What is the highest level of school you have completed or the highest degree
you have received? (IF HIGH SCHOOL, ASK:) What was the highest grade
you completed?
8th grade or less...........................................................1
Grades 9-12................................................................2
High school grade.......................................................3
Some college/Trade school........................................4
College grad................................................................5
Post graduate degree.................................................. 6
DON’T KNOW......................................................DK
REFUSED............................................................ REF
85. What is your marital status? Are you (READ CATEGORIES)I
Married..............................................................................1
Not married but living together.......................................2
Widowed............................................................................3
Divorced........................................................................... 4
Separated.................................................................... 5
Never M arried............................................................6
DON’T KNOW ......................................................DK
REFUSED............................................................ REF
86. Are you a Latino or of Hispanic origin, such as Mexican-American, Latin
American, South American, or Puerto Rican?
Yes, Hispanic................................................................. 1
No, non-Hispanic...........................................................2
DON’T KNOW .........................................................DK
REFUSED................................................................REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
172
87. Which of the following best describes you Hispanic ancestry or ethic origin?
(READ CA TEGORIES)(ANSWER CAN BE A MULTIPLE)
Mexican.......................................................................... 1
Salvadoran......................................................................2
Guatemalan.................................................................... 3
Other Central American............................................... 4
South American.............................................................5
Puerto Rican.................................................................. 6
Cuban..............................................................................7
Other (specify)............................................................... 8
DON’T KNOW.........................................................DK
REFUSED................................................................REF
88. For classification purposes, we like to know what your racial background is.
Are you White, Black or African-American, Asian or Pacific Islander,
American Indian or an Alaskan native, or a member of another race?
(ANSWER CAN BE MULTIPLE)
W hite.............................................................................. 1
Black...............................................................................2
Asian or Pacific Islander.............................................. 3
American Indian/ Alaskan native................................ 4
Other (specify)............................ 5
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF ASIAN OR PACIFIC ISLANDER, ASK:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
173
89. Which of the following best describes your Asian ancestry or ethnic origin?
(READ CATEGORIES. ANSWER CAN BE A MULTIPLE.)
Chinese...........................................................................1
Korean............................................................................2
Filipino...........................................................................3
Japanese......................................................................... 4
Vietnamese....................................................................5
Asian Indian................................................................... 6
Cambodian..................................................................... 7
Hawaiian.........................................................................8
Guamanian..................................................................... 9
Samoan.........................................................................10
Laotian..........................................................................11
Other (specify)............................................................. 12
DON’T KNOW.........................................................DK
REFUSED................................................................REF
90. Do you do other members of your household have any of the following types
of equipment?
YES NO DON’T KNOW REFUSED
(a)
Personal computer 1 2 DK REF
(b)
A rifle or shotgun 1 2 DK REF
(c)
A answering machine 1 2 DK REF
(d)
A pistol or revolver 1 2 DK REF
91. Excluding fax or modem lines, does you household have any other different
telephone numbers that I could have dialed to reach you?
YES................................................................................. 1
N O .................................................................................. 2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF YES, ASK:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
174
92. Not including the number I dialed, how many other different residential
telephone line do you have there?_______ lines
DON’T KNOW.........................................................DK
REFUSED................................................................REF
93. Do you currently receive any cash payments, food stamps, or other forms of
public assistance from a federal, state or county agency other than Social
Security, unemployment or W orker's Compensation benefits?
YES................................................................................. 1
N O .................................................................................. 2
DON’T KNOW......................................................... DK
REFUSED................................................................REF
IF YES, ASK:
94. Which of the following do you receive?
Yes No Don’t Know Refused
(a) AFDC or Aid to Families with
Dependents 1 2 DK REF
(b) SSI or Supplementary Security Income benefits
1 2 DK
REF
(c) Food stamps 1 2 DK REF
(d) General relief or general assistance 1 2 DK REF
95. We don’t need to know exactly, but just roughly, could you tell me if you
annual household income before taxes is under $20,000, between $20,000 and
$50,000, or $50,000 or more?
Under $20,000................................................................1
$20,000-$49,000.............................................................2
$50,000 or M ore.............................................................3
DON’T KNOW............................................................ DK
REFUSED....................................................................REF
IF UNDER $20,000, ASK:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
175
95b. Is it under $ 10,000 or between $ 10,000 and $20,000?
Under $10,000................................................................1
$10,000-$ 19,000............................................................ 2
DON’T KNOW ......................................................... DK
REFUSED................................................................REF
IF $20,000-$49,000 ASK:
95c. Is it between $20,000 and $30,000, between $30,000 and $40,000, or between
$40,000 and $50,000?
$20,000-$29,000............................................................ 1
$30,000-$39,999............................................................2
$40,000-$49,999............................................................ 3
DON’T KNOW ......................................................... DK
REFUSED................................................................REF
IF $50,000 OR MORE ASK:
95d. Is it between $50,000 and $75,000, between $75,000 and $100,000 or
$100,000 or more?
$50,000-$74,000............................................................ 1
$75,000-$99,999............................................................2
$ 100,000 or more...........................................................3
DON’T KNOW.........................................................DK
REFUSED................................................................REF
96. Were you bom in California, in some other state in the U.S. or outside the
United States?
California........................................................................1
Other U. S. State............................................................2
Outside the U.S.............................................................. 3
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF OUTSIDE THE U.S., DON'T KNOW OR REFUSED, ASK:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
176
97. In which country were you bom?
Countries code:______
Other (specify)_________
REFUSED................................................................REF
98. Are you currently a U.S. citizen or not?
U.S. Citizen....................................................................1
Not a U.S. citizen...........................................................2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
99. How many years have you lived in the United States?
______________ years
DON’T KNOW.........................................................DK
REFUSED................................................................REF
100. And, how many years have you live in Los Angeles County?
_____________ years
DON’T KNOW.........................................................DK
REFUSED................................................................REF
101. Thinking back over the past 5 years, was there ever a time when you
considered yourself to be homeless; that is, you did not have your own place to
live or sleep?
YES................................................................................. 1
N O ..................................................................................2
DON’T KNOW.........................................................DK
REFUSED................................................................REF
IF YES; ASK:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
177
102. During the period when you considered yourself homeless, where did you
usually sleep? (READ CATEGORIES) (ANSWER CAN BE A MULTIPLE)
In a park..........................................................................1
On the street................................................................... 2
In an abandoned building............................................. 3
In a fiend or relative' s home.........................................4
In a homeless shelter.....................................................5
In a car or truck..............................................................6
In some other place........................................................7
DON’T KNOW.........................................................DK
REFUSED................................................................REF
103. What is your current zip code? (ALL ZIP CODES MUST BEGIN WITH
"9")?
Zip Code_____
DON’T KNOW
REFUSED.......
104. In what city or town do you live?
Cities code__________________
Other (specify)_______________
DON’T KNOW.........................................................DK
REFUSED................................................................REF
105. Including yourself, how many people currently live in your household?
106. Including yourself, how many are adults are 18 or older?
107. How many are teens between the age of 13 and 17?
108. How many are children age 3 to 12?
109. How many are infants age 2 or under?
IF NO TEENS OR CHILDREN IN HOUSEHOLD:
Confirm telephone number, including area code. These are all the questions I have.
Thank you very much for your cooperation.
.DK
REF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Care management for the uninsured: A force field analysis of the business case

PDF

An analysis of third -party regulatory audit systems for medical device safety

PDF

An organizational history of the National Aeronautics and Space Administration: A critical comparison of administrative decision making in two pivotal eras

PDF

A search for theory: Performance management to improve transportation safety

PDF

Evaluation of the Care Advocate Program: Bridging managed care and home community -based services

PDF

Adapting and applying a mission-focused strategic framework for emergency management

PDF

Interoperability among complex defense computer -based systems: The Navy case

PDF

Evaluation of professional services consultants in rural government

PDF

"Are we really in charge?": An analysis of the principal -agent relationship

PDF

Conversion of health care organizations from non -profit to for -profit status

PDF

A configuration study of multiagency partnerships as practiced in Taipei City government

PDF

Assessing United States information assurance *policy response to computer -based threats to national security

PDF

Information overload: Exploring management of electronic mail

PDF

Defending against bioterrorism: Lessons from the 2001 anthrax attacks

PDF

Banking on the commons: An institutional analysis of groundwater banking programs in California's Central Valley

PDF

Dual eligible beneficiaries and managed care: Ensuring consumer protection through state Medicaid contracts

PDF

Determining risk propensity of government program managers for high risk /high payoff projects

PDF

Getting 'how' and 'why' straight: A critical discourse analysis of the National Partnership for Reinventing Government's ideological discourse on information and communication technologies

PDF

Executive development and effectiveness: A study of mobility assignment experiences of career executives in federal research and development agencies

PDF

Bridled boldness: Reengineering the Federal Bureau of Prisons. Lessons learned

Asset Metadata

Creator Salisbury, John Arnold (author)

Core Title An analysis of health risk selection and quality of care under Medicare fee -for -service and Medicare managed care health care systems

School School of Policy, Planning and Development

Degree Doctor of Public Administration

Degree Program Public Administration

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag health sciences, health care management,health sciences, public health,OAI-PMH Harvest,Political Science, public administration

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Myrtle, Robert (committee chair), Myers, Dowell (committee member), Wilber, Kathleen (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-378415

Unique identifier UC11334911

Identifier 3103964.pdf (filename),usctheses-c16-378415 (legacy record id)

Legacy Identifier 3103964.pdf

Dmrecord 378415

Document Type Dissertation

Rights Salisbury, John Arnold

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA