Citation
Slipping into the void

Material Information

Title:
Slipping into the void the specialty access gap for medically uninsured community health center patients
Creator:
Werner, James Joseph
Publication Date:
Language:
English
Physical Description:
411 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Medically uninsured persons ( lcsh )
Medicine -- Specialties and specialists ( lcsh )
Health services accessibility ( lcsh )
Health services accessibility ( fast )
Medically uninsured persons ( fast )
Medicine -- Specialties and specialists ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 394-411).
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by James Joseph Werner.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
51821284 ( OCLC )
ocm51821284
Classification:
LD1190.L566 2002d .W47 ( lcc )

Full Text
SLIPPING INTO THE VOID:
THE SPECIALTY ACCESS GAP FOR MEDICALLY UNINSURED
COMMUNITY HEALTH CENTER PATIENTS
by
JAMES JOSEPH WERNER
B.S., Ohio State University, 1980
M.S., Ohio State University, 1983
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Health and Behavioral Sciences


2002 by James Joseph Werner
All rights reserved.


This thesis for the Doctor of Philosophy
degree by
James Joseph Werner
has been approved
by

Date


Wemer, James Joseph (Ph.D., Health and Behavioral Sciences)
Slipping into the Void: The Specialty Access Gap for Medically Uninsured
Community Health Center Patients
Thesis directed by Professor Kitty K. Corbett
ABSTRACT
The purpose of this study was to gain an understanding of specialty access issues for
medically uninsured patients at three community health center (CHC) clinics located
in a western state. Qualitative methods were used to gather information from
primary care clinicians, specialists, patients, and healthcare administrators. Forty-
eight in-depth interviews and four focus groups were conducted on the subject of
access to specialty healthcare for uninsured patients. Demographic and census data
were reviewed, and direct observations were made of patient referrals and access-
related activities.
Study findings indicate that as budgetary restrictions at a major referral center
intensified, access to specialty care became increasingly inconsistent for uninsured
CHC patients. Local hospitals and specialists came under increasing pressure to
care for those traditionally referred to the academic medical center. Private
specialists reported that a variety of adverse financial factors made them hesitant to
accept significant numbers of uninsured patients. These included overall reductions
in revenues due to low reimbursement rates from insurers, the anticipation of bad
debts, and the fear of malpractice litigation.
CHC clinicians employed a many innovative strategies designed to fill gaps in
specialty access. Sixteen different strategies were identified, including expanding
the scope of practice, rewarding local specialists for charity care by referring
significant numbers of insured patients to them, and building persuasive cases using
lab tests and procedures before requesting referrals.
Marked differences between two medical communities were observed with respect
to specialty access for uninsured patients. The quality of access was reported to be
associated with the level of support from the local hospital, the convictions of local
IV


physician leaders in the community toward charity care, and the ability of CHC
clinicians to gamer the respect and cooperation of the medical community.
Understanding access difficulties from the perspectives of those most involved can
provide insights into solutions. It is hoped that future investigations of access for
medically indigent CHC patients might benefit from the findings of this study.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
v


DEDICATION
I dedicate this thesis to my parents, Kenneth and Norinne Werner, for encouraging
me every step of the way.


ACKNOWLEDGEMENT
This research was supported in part by a grant from the American Academy of
Family Physicians Foundation.
I wish to thank Professor Kitty Corbett for skillfully guiding me through the
research process, for providing expert scientific advice, and for sharing my
enthusiasm for the topic; Professor Craig Janes, for his dedication to the Program in
Health and Behavioral Sciences and for generously giving of his time and energy to
further my education; Professor Cindy Bryant, who provided scientific expertise,
skillful editing, and steady encouragement; Professor Mark Dignan, for his patience
and commitment to seeing me through; Professor Tillman Farley, who provided
much assistance with this project; and the clinicians and patients who gave of their
time and energy to help further my understanding of access for the medically
uninsured.


CONTENTS
FIGURES..............................................................xx
TABLES...............................................................xxi
CHAPTER
1. INTRODUCTION.................................................1
Outline...................................................3
The Medically Uninsured...................................4
Uninsured Hispanic Persons................................5
Reduced Safety Net........................................6
Specialty Referrals.......................................7
Specific Aims.............................................9
Overview of Research Methods.............................11
Case Studies......................................11
In-Depth Interviews...............................12
Participant Observations..........................13
Focus Groups......................................13
Terminology..............................................14
A Personal Note..........................................15
Glossaries...............................................16
viii


2. BACKGROUND.................................................17
Overview of the Health Insurance Problem................18
Access to Healthcare....................................20
Background and Theory............................20
Measuring Access.................................24
Barriers to Access...............................27
Employer-based Health Insurance..................30
The Working Uninsured............................35
Numbers of Uninsured.............................40
Colorados Uninsured.............................43
Reasons for Increase in Numbers of Uninsured.....44
Increasing Healthcare Costs......................45
Efforts to Control Healthcare costs..............48
Profile of the Uninsured.........................49
Access for Rural Residents.......................51
Uninsured by Choice..............................53
Consequences of Being Uninsured.........................54
Health Services Utilization......................54
Illness Type and Utilization.....................56
Delaying Care....................................57
IX


Uninsurance and Health Status
58
Income and Health Status.............................59
Sub-standard Care....................................60
Relationships between Poverty, Access, Insurance
and Health Status.....................................62
The U.S. Hispanic Population................................66
Demographics.........................................66
Hispanic Persons and Health Insurance................69
Hispanic Health Status...............................70
Hispanic Healthcare Utilization......................71
Barriers to Accessing Healthcare for Hispanics.......72
Undocumented Immigration.............................75
Ethical Considerations in Healthcare Resource Allocation....82
Basic Principles.....................................82
Justice and the Allocation of Limited Resources......85
Health Resource Allocation and U.S. Policy...........88
Resources to Meet the Needs of the Uninsured................93
The Healthcare Safety Net..........................93
Medicaid.............................................95
Colorados Medicaid Program....................96
Community Health Centers.............................97
x


Community Health Centers in Colorado
101
j
I
1

Colorados Program for Indigent Patients.........102
Hospital Care of the Uninsured...................105
Academic Medical Centers..................106
Emergency Department Care.................109
Charity Care by Physicians.......................112
Health Insurance for the Uninsurable...........114
Primary Care...........................................116
The History of Primary Care......................117
The Efficiency of Primary Care...................117
Community-Oriented Primary Care..................119
Specialty Referrals....................................120
Referrals between Medical Providers..............120
Roles and Responsibilities in Consultations and
Referrals.......................................121
Costs Associated with Referrals..................123
Reasons for Consultations and Referrals..........124
Primary and Specialty Care Provider Differences..124
Summary................................................126
3. METHODS...................................................130
Methodology............................................131
xi


Qualitative Inquiry..................................134
Study Methods...............................................137
Initial Steps: Family Physician Focus Groups.........137
Focus Group Sample............................138
Focus Group Findings..........................139
Gaining Access to the Study Setting..................141
Overall Design.......................................143
Ethnographic Emphasis.........................143
Ethnographic Methods..........................146
Key Informants................................150
Study Context........................................150
Sampling Strategy....................................151
Description of the Sample............................152
The Community Health Center...................153
Designations of Urban and Rural...............155
The Clinics...................................156
The Case Studies.....................................162
Case Study Sampling...........................164
Case Study Protocol...........................170
Number of Case Study Interviews...............171
In-Depth Interviews..................................171
xn


Participant Observations........................174
Analytic Techniques.............................178
Transcription............................178
Overview of the Analysis.................179
Components of the Analysis......................181
Coding...................................181
Memoing..................................183
Pattern Coding...........................183
Theme Development........................185
Summary...............................................188
4. FINDINGS.................................................191
CONTEXT AND ENVIRONMENT..................................191
Historical Context....................................191
The Role of University Hospital.......................196
The Rationale for Restriction Access............198
Outpatient Gateway to Inpatient Care............201
Inaccessible Specialties........................202
Coping with Change..............................207
Other Community Health Centers Perspectives....210
Increasing Morbidity............................211


Increasing Demand...............................212
Local Access...........................................214
Summary................................................216
5. CHC CLINICIANS PERSPECTIVES ON PHYSICIANS
RESPONSIBILITY TO CARE FOR THE MEDICALLY
INDIGENT..................................................218
Private Entities Not Sharing the Load..................219
Physicians Incomes and the Obligation to Provide
Charity Care...........................................220
The Professions Public Image..........................222
Not Necessarily a Profession of Humanitarians..........223
Regard for Charitable Specialists......................225
Summary................................................227
6. FACTORS THAT DETER SPECIALISTS FROM SERVING
MEDICALLY UNINSURED CHC PATIENTS..........................228
Adverse Financial Impacts..............................229
Uninsured Patients Threaten Financial Viability.229
Low Reimbursements from Health Insurance Payers....234
Patients Defaulting on Debts....................237
Uninsured are University Hospitals Responsibility.238
Burden of Emergency Department Call.............239
Skepticism about Medical Indigence.....................242
Cultural and Language Barriers.........................243
xiv


Fear of Litigation....................................245
Disillusionment During Medical Training...............247
Aggressive For-Profit Business Strategies.............248
Tension between Professional Values and Profit
Motives.........................................250
Summary...............................................251
7. FACTORS THAT ENABLE SPECIALISTS TO PROVIDE
CARE TO UNINSURED PATIENTS..............................254
Cooperative Referral Strategy.........................255
Positive Role Models..................................257
Humanitarian Concerns.................................260
Spiritual Vocation....................................261
Exemplary System......................................262
Rewarding Relationships with Providers and Patients...263
Cooperation from Hospitals and Physicians.............265
Summary...............................................268
8. STRATEGIES TO OVERCOME ACCESS-RELATED
CHALLENGES..............................................271
Strategies for Gaining Access.........................271
Expanding the Scope of Practice.................271
Patient Safety and Malpractice Risks.....273
CHC Limits for Expanded Practice.........278
xv


Benefits for Clinicians
279
Conservative Treatments in Expanded
Practice....................................280
Telephone Consultations Facilitate Expanded
Scope of Practice...........................281
Deal-Making.........................................285
Over-Referral.......................................288
Brief Consultations.................................289
Brief Hospitalization...............................290
Balancing Referrals to Specialists..................292
Balancing Referrals to Institutions.................294
Clinician to Clinician Communication................295
Relationship Building...............................298
Case Building.......................................299
Specialty Clinics...................................301
Interpersonal Approaches............................303
Appealing to the Values of Specialists.......304
Humanizing the Patient.......................305
Consistently Showing Appreciation............306
Differences in the Utilization of Strategies...............307
Summary....................................................307
xvi


9. MEDICAL COMMUNITIES AND LOCAL ACCESS
TO SPECIALTY CARE..........................................312
The Communities..........................................312
Plains.............................................313
Foothills..........................................314
Grasslands.........................................315
Assumptions about Local Access...........................316
Differences in Local Access..............................317
CHC Clinicians Explanations.......................321
The Role of the Hospitals................................323
Plains.............................................324
Foothills..........................................325
Physician-Leaders in the Medical Communities.............329
Plains.............................................330
The Clinic Directors Role..................332
Foothills..........................................333
The Clinic Directors Role..................336
Summary..................................................340
10. PATIENTS INDIVIDUAL ACCESS BARRIERS
AND PERSONAL EXPERIENCES...................................343
Barriers and Solutions...................................343
Financial Hardship.................................343
xvii


CICP Unacceptable to Private Specialists...344
Advance Payment Barrier....................345
Payment Plan Solution......................347
Unsatisfactory Alternatives................348
Cultural and Language Barriers....................352
Addressing Cultural and Language Barriers.356
Transportation Barriers...........................357
Transportation Solutions...................359
Patients Personal Experiences...........................362
Loss of Control...................................362
Personal Impact...................................363
Summary..................................................367
11. CONCLUSIONS................................................369
Summary of Findings......................................369
Applied Healthcare Ethics................................374
Developments Since the Investigation.....................376
Suggestions for Mitigating Access Difficulties...........378
Clinician Training and the Transfer of Knowledge.........383
Implications for Patients Safety........................386
Epilogue.................................................387
xviii


APPENDICES.....................................................388
A. Acronym Glossary.....................................388
B. Pseudonym Glossary...................................390
C. Specific Aims of the Study with Corresponding Themes
and Pattern Codes....................................391
BIBLIOGRAPHY...................................................394


FIGURES
Figure
1.1 Predictive model for access to healthcare.................................23
1.2 Linear relationships between income, insurance, access,
and health status.........................................................63
1.3 Cyclical relationships between income, insurance, access
and health status.........................................................64
3.1 Study model: naturalistic inquiry and triangulation of methods...........146
3.2 Study contexts...........................................................151
3.2 Model of the inferential analysis with recursive grounding...............181
xx


TABLES
i
Table
1.1 Estimates of access difficulties based on different definitions
of access...................................................................27
3.1 Themes of qualitative inquiry................................................136
3.2 Summary of methods....................................................149
3.3 Demographic profiles for participating clinic locations..............162
3.4 Pattern codes.........................................................185
3.5 Pattern codes with corresponding themes..............................187
f
XXI


CHAPTER 1
INTRODUCTION
Primary care physicians working in Americas community health centers (CHCs)
frequently provide care to medically uninsured patients. They routinely determine
which health problems warrant specialized medical care and refer those patients to
appropriate specialists. While low-cost primary care services are available through
CHCs, specialty care is not. Consequently, medically uninsured patients may have
difficulty obtaining access to care from specialty providers or hospitals. Without the
means to pay for these services, medically uninsured community health center
patients may slip into the void between primary care and specialty care, and go
without needed healthcare services. The size of the specialty access gap in the
nations healthcare safety net varies between states and communities.
Restricted access is a potential problem for the more than 41 million medically
uninsured Americans, representing 14.6 percent of the U.S. population (U.S. Census
Bureau, 2002). The proportion of the population without health insurance has
consistently risen over the past three decades, and shows few signs of abatement.
Healthcare costs have steadily outpaced increases in real wages since the 1970s,
1


rendering health insurance unaffordable for millions of low-wage workers. In
addition, those lacking proof of citizenship may be barred from obtaining specialty
care at public hospitals. Estimates indicate that 5 million to 11 million undocumented
persons reside in the U.S. (Thurman, McCruden, Hamric, & Goodman, 2000;
Krikorian, 2001), and healthcare access for undocumented individuals is very limited.
Viable specialty care options for medically indigent patients are few. Emergency
department (ED) care is available to all. However, EDs specialize in providing care
for emergent conditions and are not equipped to provide many other forms of medical
care, nor do they provide preventive or longitudinal care. Historically, academic
medical centers have opened their specialty clinics for referrals of the medically
uninsured. For many academic hospitals, however, market pressures have reduced
revenues, and increasing numbers of uninsured patients have drained the hospitals
limited resources. As a result, many have been forced to cut back on the amount of
unreimbursed care that they can afford to dispense.
Providing adequate access to specialty care for the medically indigent is a vexing
problem that represents a significant hole in the nations healthcare safety net. The
intent of this research study has been to examine the problem of access for the
uninsured from the vantage points of primary care providers, patients, specialists, and
2


healthcare administrators. Their perspectives, in their own words, are the data that
allow in-depth understanding of these issues.
Outline
This chapter provides an introduction to the research study. It presents the Specific
Aims of the study and provides an overview of the research methods that were used.
It also provides an introduction to the following topics: the numbers of medically
uninsured individuals in the U.S., the impact of medical uninsurance on Hispanic
persons, the state of the nations healthcare safety net, and the ways in which referrals
are used to gain access to specialists. Finally, this chapter provides operational
definitions for terms that are used in the document, and offers an explanation of why
this topic was selected for investigation.
Subsequent chapters provide more in-depth presentations of background information,
methods, findings, and conclusions. Extensive background information is provided in
Chapter 2. Chapter 3 provides a detailed description of the studys methods and
analytic procedures, and presents the key patterns and themes that emerged. The
findings of the study are presented in Chapters 4 through 10. Conclusions drawn
from the findings are provided in Chapter 11.
3


The Medically Uninsured
Although the U.S. is one of the worlds wealthiest nations, over 41 million people
lack basic health insurance, a figure that represents 14.6 percent of the nations
population (U.S. Census Bureau, 2002). Virtually every study on the use of medical
services has reported that a lack of health insurance represents a major barrier to
medical care (Freeman, Aiken, Blendon, & Corey, 1990). Over the last two decades,
the portion of the population without health insurance has increased significantly
(Kuttner, 1999), affecting access to health services for millions. Presently, there is no
evidence to indicate that this trend will reverse at any time in the future.
Why so many U.S. citizens lack basic coverage is a challenging question with no
single, clear-cut answer (Friedman, 1991). Over the past few decades, declining
employer-sponsored coverage has been cited as a primary reason. From 1995 to
1999, however, the rate of employer-based coverage in the U.S. increased, ostensibly
due to a strong economy. However, these gains were more than offset by declines in
Medicaid and in private non-group coverage (Holahan & Kim, 2000). Thus, the trend
toward increasing numbers of uninsured continued during a period of rapid economic
expansion and decreasing unemployment. Since the economic downturn that began
in 2000, reports indicate that employer-based coverage has declined, resulting in
increasing numbers of uninsured (Austin, 2002a).
4


The majority of the uninsured are low- and moderate-wage workers and their
families. Approximately one-third of U.S. adults age 18-64 years are uninsured
(Davis, Rowland, Altman, Collins, & Morris, 1995; Ome, Fishman, Manka, &
Pagnozzi, 2000). Most uninsured adults are either working full or part time (Schoen
& Puleo, 1998). Their incomes are often too high to qualify for Medicaid but too low
to permit them to afford health insurance premiums.
Health resource allocation issues have often been considered within the context of
social justice. Theories relevant to the economic macroallocation of healthcare
include egalitarianism, utilitarianism, and libertarianism. These opposing social
perspectives have led to significant conflicts in setting forth a national healthcare
policy. The countrys predominant healthcare system is largely based on a libertarian
perspective, in which those with the financial means to afford health insurance
purchase it in the free market. Those who cannot afford health insurance are often
left to the vagaries of a system not designed to accommodate them.
Uninsured Hispanic Persons
Significant numbers of low-income Hispanic persons receive healthcare from
community health centers in Colorado, which was the setting for this research study.
Persons of Hispanic ethnicity are the most rapidly growing segment of the U.S.
population, now representing 12.5 percent of the U.S. population. Between 1990 and
5


2000, the Hispanic population increased by 77 percent in Colorado, to 754,000 people
(Aguilar, 2001). Hispanic persons represent 17.1 percent of the states population
(U.S. Census Bureau, 2001).
Rapid population growth in the Hispanic sector coupled with high rates of poverty
and uninsurance have resulted in an increasing need for health services for low-
income Hispanic persons. Nationally, 35.3 percent of Hispanic persons were without
health insurance in 1999, a substantially greater proportion than non-Hispanic Whites,
11.0 percent of whom were uninsured in 1999 (U.S. Census Bureau, 2000). Between
1977 and 1992,40 percent of the increase in uninsured persons in the U.S. was
accounted for by individuals of Hispanic origin, and Hispanics of Mexican origin
alone accounted for 27 percent of the increase (Berk, Albers, & Schur, 1996).
Persons of Hispanic ethnicity are least likely to have health insurance in Colorado
when compared to persons of other ethnicities. Approximately 26 percent of
Hispanic persons in the state lack health insurance, compared to 20.8 percent of non-
Hispanic African Americans and 13.3 percent of non-Hispanic Whites (Yondorf,
2001).
Reduced Safety Net
A loosely-knit healthcare safety net comprises community health centers (CHCs) and
public hospitals in the U.S. Community health centers typically focus on primary
6


care services, often relying on public hospitals as referral centers. Public hospital-
based referral centers are experiencing reduced revenues due to competition in the
healthcare market while facing ever-increasing numbers of uninsured patients. These
factors threaten access to specialty care for medically uninsured patients who receive
primary care at CHCs, which have the stated mission of serving low-income
individuals.
Within Colorado, such pressures have caused the leading referral hospital for
medically indigent patients to progressively restrict the number of non-paying
patients it can accept (Sherry, 2001). Many community health centers have
traditionally relied on this academic medical center to provide specialty care for
uninsured patients. The financial issues facing the hospital are not easily resolved.
They have resulted in restricted access for poor residents in need of specialty care,
and challenges for the primary care providers who seek to obtain specialty referrals
for medically indigent patients. Similar circumstances are reported to be prevalent in
many communities across the nation.
Specialty Referrals
Most patients gain access to specialists through referrals made by primary care
clinicians. Although patient referrals have been studied in depth in the United
Kingdom, Australia, and the Netherlands, relatively little is known about how
7


referrals are conducted in the U.S. healthcare system (Nutting, Franks, & Clancy,
1992; Roland, 1988; Shortell & Anderson, 1971). Studies have examined a wide
variety of topics including the frequency of referrals to various specialties, the
reasons for referral, the frequency of written communication between primary care
physicians and specialists, the validity of referrals based on medical records, and the
effect of different types of health insurance plans on referrals. They have yielded
important information about the referrals in the U.S. and furthered the understanding
of the healthcare system.
Past referral studies, however, have generally involved patients with some form of
health insurance that provided coverage for both primary and specialty care. Because
of this, specialty access has been irrelevant to most studies of referrals, and thus, has
remained unexamined. In contrast, this study involved patients for whom the lack of
health insurance was a barrier to gaining access to specialty care.
This research study has sought to gain insight into access issues by examining
community health center clinicians attempts to refer uninsured patients to specialists.
Cases of access were studied in depth, and primary care clinicians were interviewed
about their current and past experiences seeking specialty access. The perspectives of
specialists were sought about access for uninsured patients. Patients were
interviewed about specialty access issues and the experience of being medically
8


uninsured. Healthcare administrators were interviewed about policies that affect
access for the medically indigent. In addition, detailed information was captured
about the context in which referrals were made and access was sought.
Specific Aims
The specific aims of the research study are:
1. Identity key issues and describe challenges that face a community health
center (CHC) attempting to obtain specialty care for medically uninsured
patients.
a. Describe limitations in referring uninsured and undocumented patients.
i. Describe political and economic constraints affecting a major
referral hospitals ability to provide care to uninsured.
b. Characterize systemic issues within local health systems that affect
access to specialty services for medically uninsured patients.
2. Describe the perspectives and experiences of a CHCs clinicians and staff
regarding the referral of medically uninsured patients.
a. Identify challenges routinely faced by the CHC clinicians in
attempting to secure specialty care for uninsured patients.


b. Characterize strategies and tactics developed by the CHC clinicians to
secure specialty care for uninsured patients.
c. Describe how the CHC clinicians address patients needs for specialty
healthcare when referrals cannot be obtained.
3. Characterize the perspectives and attitudes that specialists have regarding the
referral of medically uninsured patients to their practices.
a. Describe specialists views on providing care to uninsured patients.
b. Describe specialists experiences in providing care to medically
uninsured patients.
c. Characterize ways in which specialists attitudes towards uninsured
differ by community.
4. Describe the perspectives and experiences of a CHCs medically uninsured
patients with respect to specialty access.
a. Describe barriers faced by uninsured patient attempting to obtain
specialty care.
b. Describe the effects of lack of health insurance and restricted specialty
access on patients daily lives.
10


Overview of Research Methods
Qualitative research methods were used to examine referrals of medically indigent
patients from primary care providers to specialists (see Chapter 3 for a detailed
presentation). The design incorporated case study, in-depth interview, and participant
observational methods. Three separate clinics within a community health center
system consented to participate in the study. A family physician practicing in each
clinic served as the primary key informant for that clinic. Clinicians were
interviewed about their general referral practices and patterns, their expectations of
specialists, and their past and present experiences in obtaining specialty care for their
patients. Other clinicians, staff members, administrators, and patients from each
clinic also participated in interviews.
Case Studies
Each Clinic Director participated in one longitudinal case study of a specialty referral
made by him or her. This process captured the perspectives of primary care
clinicians, specialists, and patients as the referrals occurred and enabled the
investigator to acquire a broad understanding of access issues. Primary care
clinicians were interviewed multiple times during the course of patients visits to
specialists. Patients were interviewed twice during the referral process; once prior to
their initial visit with the specialist and once following their final visit with the
specialist. Specialists were interviewed following patients initial visits.
11


The case studies yielded data from 21 in-depth interviews and numerous informal
interviews. They provided important information that directly addressed the research
questions and shaped additional data gathering activities. The longitudinal case
approach permitted a holistic view of the referral process from the patients initial
visits with referring primary care clinicians through the consultations with specialists,
and enabled the identification and understanding of all aspects of the process. The
resulting data informed the design of subsequent data collection activities.
In-Depth Interviews
In addition to the interviews conducted through the case studies, 27 in-depth
interviews were conducted with primary care and specialty physicians, patients, and
health administrators about issues relevant to healthcare access. In-depth interviews
permitted the investigation of issues encountered by each of the parties directly
involved in healthcare access. The semi-structured interview method was flexible,
enabling an inductive approach that permitted exploration of newly emerging
constructs and patterns. Including the case studies, a total of 48 in-depth interviews
were conducted.
12


Participant Observations
The investigator conducted participant observations at each clinic in order to gain an
understanding of the contexts in which access was sought, and to observe access-
related events as they naturally occurred. Many access cases were informally
followed using this method. Participant observations enabled a broad range of
perspectives to be captured, added new and different types of information, confirmed
findings, and aided in the interpretation of meanings.
A sample of each participating physicians patient visits were observed and
interpersonal interactions, patient flow, and clinic operations were studied. Medical
providers were interviewed about their experiences referring patients, and cases were
followed. Nurses and medical assistants were interviewed about the mechanics of the
referral process at each clinic, and clinic administrators were interviewed about
referral services and community referral programs.
Focus Groups
Four focus groups were conducted during the course of data gathering. Two initial
focus groups were held with a broad sample of primary care doctors from across the
U.S. The initial focus groups preceded other data gathering activities and served to
inform the design of the project. A third focus group, conducted after a substantial
amount of data was collected, involved family physicians from community health
13


centers across the country. Its purposes were to learn about access challenges in other
health systems and to gather information about strategies intended to surmount
specialty access barriers. A fourth focus group was conducted as a member checking
exercise with two key informant physicians for the purpose of verifying the research
findings.
This section has provided a brief overview of the methods. Chapter 3 contains more
detailed information about the study methods and the settings in which it was
conducted.
Terminology
Primary care physicians are members of medical specialties such as family medicine
and general internal medicine, and as such, are specialists. Please note, however, that
within this document, clinicians in the primary care specialties are referred to as
primary care clinicians, and physicians and other healthcare providers who accept
referrals of patients from primary care clinicians are referred to as specialists. This
convention was used by the providers, patients, and administrators participating in
this project, and it is used throughout this document to eliminate ambiguity.
14


A Personal Note
The initial intent of this study was to investigate the coordination of patient care
during referrals, with attention to communication between primary care clinicians and
specialists. However, physician-participants in the initial focus groups made me
aware of access issues at community health centers. As I began to collect preliminary
data about uninsured patients in the community health center clinics, it became
apparent that coordination of care was less relevant clinicians and patients than the
fundamental problem of gaining access to specialty care. This finding shifted the
focus of the study toward access to care for uninsured patients. Subsequent data
collection activities revealed information not found in previously published research
investigations. The topics of communication and coordination of care continued to be
included in the investigation; however, they were explored as within the study of
access to care for uninsured patients.
It has been argued that any attempt to investigate the causes of health disparities is
value-laden, and their investigation inevitably involves issues of social justice (Peter
& Evans, 2001). The subject of this investigation describes inequities in the access to
healthcare, which have been clearly shown to translate into health disparities. Despite
the compelling nature of this topic, I have sought to present this research in a voice
that sits nearer the dry dispassion of the scientist than the urgent plea of an advocate.
I have attempted to present the information in way that is fair, accepting that it is not
15


possible to remove every trace of personal and professional bias. My objective has
been to give voice to a microcosm of primary care clinicians, specialists, uninsured
patients, and health administrators who routinely grapple with these complex issues
Glossaries
Many acronyms are used in this document as abbreviations of the names of
institutions and programs. For the readers convenience, a glossary of these
acronyms is provided in Appendix A.
Pseudonyms have been used in an effort to protect the identities of institutions and
places. A glossary of these pseudonyms is provided in Appendix B.
16


CHAPTER 2
BACKGROUND
A thorough review of the relevant literature makes it possible for an investigator to
design and conduct a research study in an informed manner, and to interpret the
findings in light of what is already known. The topic of specialty access for
uninsured patients cuts across numerous bodies of knowledge. In order to understand
the key issues related to the topic, an extensive review of the literature was
conducted. This involved identifying, reading, abstracting, and summarizing key
segments of information. The broad areas addressed by this review included:
access to healthcare;
the health insurance problem in the U.S.;
health-related consequences of being uninsured;
healthcare ethics;
resources and systems designed to meet the needs of the uninsured;
the practice of referring patients to specialists;
health issues specific to people of Hispanic ethnicity and Latino heritage;
the nature of primary care.
17


This chapter provides background information from the existing literature
on these key topics and issues.
Overview of the Health Insurance Problem
In contrast to most developed nations, the United States has a sizeable percentage of
its population that is medically uninsured (Friedman, 1991; Kuttner 1999; Schroeder,
1996). With the exception of the U.S. and South Africa, the worlds developed
countries generally guarantee either government- or employer-sponsored health
insurance as a right of citizenship (Friedman, 1991; Schroeder, 1996). The medically
uninsured include the medically indigent, individuals whose clinical conditions make
them medically uninsurable, and the working uninsured (National Library of
Medicine, 2001).
Whether healthcare should be provided for all Americans has been the subject of
much controversy and debate and is one of the most important public policy issues
facing the United States (Hubbell, Waitzkin, Mishra, & Dombrink, 1989). The
subject of expanded or nationwide health insurance has been widely discussed as a
national policy issue numerous times over the past 50 years in 1948, 1965, 1973,
and most recently in 1993 (Schroeder, 1996). The passage of the Social Security Act
in 1935, the establishment of the Federal Employees Health Benefits Program in
1959, and the approval of the Medicaid and Medicare federal programs in 1965
18


marked substantial increases in public healthcare coverage for uninsured people
(Edge & Groves, 1998). In the 1960s, health insurance was initiated in collective
bargaining agreements between labor unions and management (Freeman et al., 1990).
During the same era, federal funding began for the establishment of community and
migrant health centers, which offered primary care health services to poor and
migrant workers. Since then, however, relatively little has been achieved with respect
to bolstering the healthcare system for poor and low-income people. The country
does not have a national healthcare safety net for the uninsured, but rather has a
patchwork of local safety nets that vary greatly in their ability to meet the needs of the
medically indigent (Baxter & Mechanic, 1997).
It appears that the political issue of determining how to deal with the medically
uninsured has been distorted by those on both sides of the argument. As with many
politically charged issues, the truth appears to lie between the extremes (Donelan et
al., 1996). From one perspective, healthcare is already available to the uninsured
through community health centers and emergency departments. Although these
entities provide low cost healthcare, this section will provide information about the
shortcomings of each. Individuals holding an opposing perspective have posited that
all who are medically uninsured are in a tragic situation. This section will also
provide information about healthy uninsured individuals and those who voluntarily
forgo health insurance.
19


Access to Healthcare
Background and Theory
Timely access to appropriate health services is considered important to the
maintenance of health. Access to care has been defined as the timely use of personal
health services to achieve the best possible health outcomes (Millman, 1993). The
concept of access to care has relevance to both primary and specialty care (Donaldson
et al., 1996). Various aspects of healthcare access are regularly tracked by
governmental agencies, private foundations, and private and public healthcare
systems using population-based survey methods. The results of these surveys are
used by policymakers, purchasers of healthcare, health plans, providers, regulators,
and consumers (Gold & Stevens, 1998). Each of these groups may have distinct
needs that may or may not be addressed by the administration of a specific access
measurement tool.
Aday and Anderson (1975) developed one of the first explanatory models of
healthcare access (see Figure 1.1). The model suggested that an individuals use of
healthcare may be predicted by the willingness to use services, the factors that either
support or impede their use of health services, and the need for care (Gold, 1998).
Because governmental policy often determines financing and manpower, health
policy was incorporated as the starting point of the model (Aday & Anderson, 1975).
20


The model suggests that two primary themes are predictive of healthcare access. The
first theme is potential access, which is simply defined as the presence of a source of
healthcare (Bashshur, Homan, & Smith, 1994), and the potential to use it. Potential
access is determined by structural indicators that are characteristics of the healthcare
system. Structural indicators include the availability of healthcare and the
organization of the delivery system. Availability is determined by the volume of
services available and the distribution of those services and providers. Organization
comprises organizational and structural health system characteristics that influence
ease of entry. Potential access also includes process indicators which are
characteristics of the patient population. These include both predisposing factors
(need for care, health beliefs, social structure) and enabling factors (availability and
organization of health services) (Gold, 1998). Predisposing and enabling factors may
be changeable (e.g., health beliefs, education, knowledge) or not readily changeable
(e.g., age, race, social structure).
The second theme in the model is realized access, which is based on utilization and
patient satisfaction pertaining to obtaining and receiving healthcare (Bashshur et al.,
1994). Realized access consists of an objective and a subjective outcome indicator.
The objective indicator is the utilization of health services, which comprises the
specific measurable aspects of utilization including the type of utilization, the site
utilized, the purpose of the visit, and waiting periods for visits. The subjective
21


outcome indicator is consumer satisfaction with care, which is a function of
convenience, availability, financing, provider characteristics, and the perceived
quality of the care received.
The components of the model are related to one another. Health policy affects the
delivery system by increasing or decreasing resources or by developing programs to
reduce the health risks of the target population (Aday, Anderson, & Fleming, 1980).
The system may also indirectly affect a populations utilization of services and
consumers satisfaction with care (Aday et al., 1980). A given populations
characteristics may influence use and satisfaction independently of delivery system
properties (Aday et al., 1980). Finally, utilization of services may influence the
consumers satisfaction with the overall delivery system.
22


Figure 1.1. Predictive Model for Access to Healthcare
Source: Aday, L.A., & Anderson, R. M. Equity of access to medical care: A conceptual and empirical
overview. Medical Care, 12. 1981, 19(12), Supplement, 6. 1981, Lippincott, Williams & Wilkins.
Reprinted by permission of Lippincott, Williams & Wilkins.
23


Although the model by Aday and Anderson (1981) has served as an important
starting point for study of access, some have criticized it for not effectively
addressing the influence of integrated healthcare financing and delivery systems,
which are common today (Gold, 1998). In response to a need for a more
comprehensive model, the Institute of Medicine has developed a complex model
which incorporates managed care components such as health plan selection variables,
delivery system characteristics, enrollment choices, use of services, determinants of
continuity, clinical outcomes, and variables which mediate these processes (Millman,
1993). This model approaches access as a determinant of the characteristics of health
plans rather than a choice of individual healthcare consumers, and makes the
assumption that health plan coverage exists for each individual. The models
incorporation of outcomes measurements has advanced the study of access by
measuring quality and effectiveness of care. However, because of its emphasis on
integrated delivery systems and the assumption of coverage, its utility lies in
analyzing managed care markets, and its relevance to this project is quite limited.
Measuring Access
It is difficult to determine the degree to which health services are accessible in the
United States. Population-based surveys are often employed to determine access to
healthcare, but problems arise when different conceptual perspectives about access
are incorporated into measurement tools, often resulting in widely disparate estimates
24


of access. One of the most commonly used measures of access is whether or not an
individual has health insurance coverage (Berk & Schur, 1998). It is widely used as a
proxy for assessing healthcare access. However, the simple presence or absence of
health insurance does not determine a lack of access but is considered an enabling
characteristic because it is highly correlated with the use of health services (Berk &
Schur, 1998). It does not provide a direct measure of the amount of care received;
however it is a discrete, dichotomous measure that is regarded to be relatively
objective, leaving little room for ambiguity or subjectivity (Berk & Schur, 1998).
Another measure of access is an individuals perception of whether or not they have
been able to receive all of the healthcare that they believed was necessary (Berk,
Schur, & Cantor, 1995). This concept is substantially less objective than the presence
or absence of health insurance but may be a more direct indicator of the ability to get
healthcare, which is central to access. Areas of ambiguity inherent in this measure
include the varying thresholds at which different individuals believe that healthcare is
necessary and the window of time for which the respondent may be asked to recall
information. Longer periods time are subject to increasing levels of recall bias. In
addition, the operational definition of healthcare may differ between instruments. In
some surveys it is limited strictly to medical care, and in others it includes dental
care, prescription drugs, eye care, and mental healthcare (Berk & Schur, 1998).
25


Two other measures of access that are less frequently used include the use of a
hospital emergency room as a usual source of care and the presence of health
problems but the absence of a doctors office visit (Berk & Schur, 1998). The first is
a fairly objective measure but is neither necessary nor sufficient to lacking access.
The second is significantly more subjective, as the definition of an un-addressed
health problem is subject to the interpretation of the respondent.
As discussed, the number of persons reported to have access difficulties differs
greatly with the definition of access that is used. A summary of differing estimates of
access based on alternative definitions is shown in Table 1.1. The original source of
this information was the 1994 Robert Wood Johnson Foundation National Access to
Care Survey, which incorporated five different measures of access. The estimates of
access varied greatly depending on the operational definition used.


Table 1.1. Estimates of access difficulties based on different definitions of access
Definition Number of persons (millions)
Uninsured 34.5
Inability to obtain medical care 14.7
Inability to obtain a specific service3 41.5
Use of emergency room as usual source of care 6.3
Fair or poor health status, no visits 4.5
Source: Berk, M. L. & Schur, C. L. Meaursing access to care: Improving information for
policymakers. Health Affairs. 1998, 17(1): 182. 1998, The People to People Health Foundation,
Inc. Reprinted by permission of The People to People Health Foundation, Inc.
Refers to inability to obtain at least one of the following: medical or surgical care, dental care,
prescription drugs, mental healthcare or counseling, and eyeglasses.
Although the measures discussed here have been frequently used, Gold (1998) has
suggested that the concept of access continues to evolve over time, because
healthcare systems change continually. It may be necessary to not only measure the
ability enter the health system, but also the ability to obtain care once entry has been
achieved (Gold, 1998), and to measure processes such as the ease with which a
system can be negotiated by patients of different backgrounds and demographic
characteristics.
Barriers to Access
Health-related barriers are beliefs an individual has concerning the tangible and
psychological costs associated with taking a specific health action (Melnyk, 1988).
Originally formulated as a component of the Health Belief Model, barriers have been
27


used to help explain why individuals may not take specific health actions. These
specific health actions include health service utilization behaviors, or attempts to
access healthcare services. Thus, obstacles to accessing healthcare services have
been referred to as barriers to access.
Barriers to access have often been dichotomized as either structural or individual.
According to Melnyk (1988), the following kinds of structural barriers have been
identified:
System: an observable characteristic of the health system, or the
subjective perspectives offered by individuals;
Time: travel time, waiting time, and lead times for appointments;
Distance: distance to the provider and lack of transportation;
Cost: service costs and costs for insurance coverage;
Availability: unavailable or inaccessible services and the efficiency
of appointment systems;
Organization of services: fragmented service systems and
inappropriate referrals;
Discrimination: racial or ethnic discrimination;
Provider-Consumer relationship: shortcomings in the relationship
between provider and patient.
28


Melnyk (1988) has also identified the following types of individual barriers:
Demographic factors: income, socio-economic status, age, and
education;
Attitudes: patient attitudes and psychological factors;
Knowledge: patients health-related knowledge and understanding of
health insurance programs;
Effort: the effort that must be put forth in order to obtain healthcare;
Cultural factors: language, ethnicity, cultural, and culturally-related
naivete;
Family characteristics: large family size, family responsibilities, and
influence of family members.
Like Melnyk (1988), Himmelstein and Woolhandler (1995) found that primary
utilization obstacles included financial cost. Cost was identified as the major barrier
for both the insured and uninsured (Himmelstein & Woolhandler, 1995), but because
the poor also have fewer non-financial resources than those in higher income groups,
non-monetary barriers to healthcare impact the poor most heavily as well.
Even among the insured, however, barriers to care exist. An analysis of the National
Medical Expenditure Survey (NMES) found that three-quarters of those who reported
being unable to obtain needed services already had healthcare coverage and nearly
half had private insurance (Himmelstein & Woolhandler, 1995). The primary barrier
29


reported was cost. Thus, the cost of obtaining health services may be a major
obstacle even for those who have health insurance coverage.
With respect to obtaining health insurance, three types of barriers have been
identified: employment restrictions, cost, and preexisting medical conditions
(Blendon et al., 1992). The poor are most directly affected by employment
restrictions because they tend to work in service-sector jobs that do not include
healthcare benefits. Further, they are often unable to afford the cost of private health
insurance due to low wages. Preexisting medical conditions affect individuals across
all income groups equally, although coverage is often available, but at a significantly
higher cost. Because of this, lower income individuals with preexisting conditions
may be barred from purchasing health insurance more frequently than their higher
income counterparts.
Restricted access is a potential problem for the more than 41 million medically
uninsured Americans, representing 14.6 percent of the U.S. population (U.S. Census
Bureau, 2002).
Employer-based Health Insurance
Employer-based health insurance has been the cornerstone of healthcare coverage in
America for many years. The country has traditionally relied on a voluntary system
30


that is based on the good will of employers to provide healthcare coverage to
workers. An unspoken agreement has existed between employers and employees to
the effect that if a person was employed, he or she would receive health insurance
benefits subsidized to some degree by the employer (Friedman, 1991).
Today, nearly two in three Americans rely on employer-based health insurance
coverage (Kuttner, 1999). Other estimates indicate that 80 percent of the nonpoor
(persons with incomes above 200 percent of the federal poverty level) obtain health
insurance through employment (Yegian, Pockell, Smith, & Murray, 2000). Of the
nonpoor that do not obtain health insurance through their employer, only 7 percent
purchase individual insurance, and the remainder go without coverage (Yegian et al.,
2000).
In order to attract a limited pool of skilled workers, employers offer participation in a
group health insurance plan as a benefit of employment. However, when there is a
relatively large pool of lower-skilled workers seeking low-skill jobs, employers may
be able to secure their services without offering fringe benefits such as health
insurance (Seccombe & Amey, 1995). Job characteristics may explain why some
employers provide low skill workers with health insurance benefits. The likelihood
that insurance is offered increases if the job is unionized, if the company is large in
scale, or if the employer is a multiple site company. Thus, low-skill workers working
31


for small, single site, non-unionized companies are more often without employer-
based health insurance than those working for larger, unionized firms (Seccombe &
Amey, 1995).
An important current trend is the erosion of employer-based health coverage
(Friedman, 1991). The decline in employer-sponsored coverage has been linked with
a rise in the overall uninsurance rate, as more than 80 percent of the growth in the
uninsured over the past 5 years was among those with incomes above 200 percent of
the federal poverty level (Holahan & Kim, 2000). Some employers have eliminated
coverage altogether, but most have cut back on the choice of offerings and shifted
costs to employees by placing a cap on the employers contribution and by choosing
plans with greater out-of-pocket expenses (Kuttner, 1999).
Why the foundation of employer-based health insurance has deteriorated over the past
few decades has been the subject of both conjecture and study. A common
explanation is that employment has expanded in the market sectors that do not
provide health benefits and declined in the sectors that do (Blendon, Young, &
DesRoches, 1999; Friedman, 1991; Long & Rogers, 1995). Several forces are said to
be responsible for this: the employment shift from industries such as unionized
manufacturing to service industries where unions are relatively rare (Friedman, 1991),
the trend away from full-time employment toward part-time employment, the use of
32


consultants and temporary workers rather than full-time employees, and greater self-
employment (Long & Rogers, 1995). Such shifts began in the 1980s, a time of rapid
increase in the proportion of the working uninsured population.
However, an analysis of the 1980 to 1987 Current Population Survey data conducted
by Long & Rogers (1995) refuted notions that the proportion of the uninsured
population increased because employment shifted from full-time to part-time, that
consultants and temporary workers replaced hired workers, and that more self-
employment occurred. Their findings indicate that self-employment declined during
that period. The shift in employment sectors from manufacturing, mining, and other
industries to the services sector occurred; however it was not large enough to explain
more than 15 percent of the increase in uninsured people (Long & Rogers, 1995).
Although these factors may become more relevant over a longer period, they do not
appear to account for more than a minor fraction of the reduction in coverage during
the 1980s.
Instead, the decline resulted from falling coverage rates across all U.S. industries.
Acs (1995) found that for the period from 1988 to 1991, falling family incomes
accounted for most of the decline in overall insurance coverage. A decline in
insurance coverage across all industries, firm sizes, employment statuses, income
levels, and demographic groups accounted for most of the decline in employer-
33


sponsored insurance coverage among workers (Acs, 1995). Interestingly, rising
unemployment and economic shifts from one industry segment to another accounted
for little of the overall decline. Acs (1995) concluded that fewer employers were
offering health coverage, workers were finding it difficult to pay for their share of the
premiums, and individuals without access to employer-sponsored plans were finding
it harder to afford non-group insurance.
Kroenick and Gilmer (1999) conducted an analysis of U.S. Current Population
Survey data from 1979 to 1995 for adults aged 19 to 64 who did not receive health
insurance as a dependent or through a public program. Findings indicated that real
income per worker remained virtually the same from 1979 to 1995, while the
proportion of individuals without health insurance coverage increased substantially.
Per capita healthcare spending increased much more rapidly than personal income
during this time period. The authors concluded that rising health expenditures made
insurance increasingly unaffordable for both employers and workers, and accounted
almost entirely for the overall decline in health insurance coverage over the 25-year
period (Kroenick & Gilmer, 1999).
From 1994 to 1999, the trend toward declining rates of employer-based coverage
reversed and began to increase (Holohan & Kim, 2000), most likely due to very
strong economic growth. However, those increases were not been sufficient to
34


outweigh simultaneous declines in Medicaid and the increasing cost of participating
in employer-sponsored plans, since the proportion of the population with health
insurance continued to decline through the last decade.
The Working Uninsured
The notion that a job is a ticket to a better life does not appear to hold true with
respect to health insurance (Seccombe & Amey, 1995). An examination of the
segment of the population below the poverty line shows that only one of four poor
working persons receives health insurance from an employer in comparison to more
than two of three non-poor workers (Seccombe & Amey, 1995). Low-wage workers,
concentrated in the service sector, are especially likely to work for employers that do
not offer group health insurance coverage (Medoff, Shapiro, Calabrese, & Harless,
2001). Others work for employers that offer coverage only to those meeting certain
employment qualifications. One in three uninsured working adults have reported that
their employer offered health insurance to at least some of their employees, but were
excluded because they either worked part-time or had not worked for the employer
long enough (Blendon et al., 1992).
In comparison to the unemployed, the working poor may actually be at a
disadvantage in acquiring healthcare coverage. In a telephone survey, Hubbell et al.
(1989) found that adults in the work force were approximately twice as likely as
35


adults not in the workforce to lack regular source of care and to lack a physician visit
within the past year. An analysis of the 1977 National Medical Expenditure Survey
(NMES) found that the working poor were two times more likely to be uninsured than
were poor persons without jobs (Berk and Wilensky, 1987; Seccombe & Amey,
1995). Thus, being employed in a low-wage job may in fact restrict access to
healthcare. While Medicaid is often available to the poorest citizens, many poor
working adults earn more than is allowed under states Medicaid regulations, but not
enough to afford private insurance premiums. Only five states offer premium
subsidies to adults with incomes above 200 percent of the poverty level, and there is
no serious discussion of offering any such programs at the national level (Yegian et
al., 2000).
Uninsured workers are employed by both large and small firms. A survey conducted
in Massachusetts indicates that 65 percent of the working uninsured are employed by
organizations with fewer than 100 employees (Blendon et al., 1992). However, many
uninsured work for large firms, as is the case with seasonal and agricultural workers
(Friedman, 1991). Three of four uninsured adults are working full (58%) or part time
(18%), often with incomes that bar them from Medicaid (Schoen & Puleo, 1998). In
an analysis of data from the National Medical Expenditure Survey, only 10.2 percent
of the working poor were found to receive public insurance, while nearly half were
completely uninsured (Seccombe & Amey, 1995).
36


The working poor have a variety of reasons for lacking health insurance, including
being seasonal workers, in poor health and therefore deemed uninsurable, in risky
jobs, unaware that they qualify for public or private coverage, or simply choosing not
to acquire coverage (Friedman, 1991). Thus, the majority of the uninsured are tied
directly or indirectly to a workplace that is not a source of health insurance
(Friedman, 1991).
The growth of the uninsured has been concentrated among the working poor (Schoen
& Puleo, 1998; Seccombe & Amey, 1995). In 1996,52 percent of full-time workers
with incomes below the poverty line had no insurance (Kuttner, 1999). A national
study indicates that 84% of the uninsured are either employed or are dependants of
employed persons (Seccombe & Amey, 1995). Kroenick & Gilmer (1991) found that
the decline in employer-sponsored coverage from 1979 to 1989 was confined to low-
income workers and that coverage levels did not decline for those earning $30,000
per year or more.
Cost is the primary reason that the working poor go without coverage (Donelan,
1996). In 1983,45.5 percent of employers paid employees healthcare premiums in
full, but by 1998 that figure had declined to 26.6 percent. Rising premiums and
greater out-of-pocket expenses have caused some eligible employees to forgo
37


coverage (Kuttner, 1999, Kroenick & Gilmer, 1999). As premiums continue to
escalate over time, the employee bears an increasing financial burden for coverage.
Low-income workers may be unable to afford the cost of premiums, while their
income bars them from qualifying for assistance programs. The uninsured are often
unable to pay for healthcare out-of-pocket, which reduces the likelihood that they will
access needed medical services (Freeman et al., 1990). In a study of patients one year
after termination from Medicaid, sixty-eight percent of the group reported a specific
episode in which they could not obtain needed care, and 78 percent of those cited cost
as the reason for not obtaining care (Lurie, Ward, Shapiro, & Brook, 1984; Lurie et
al., 1986).
Some employers have dropped coverage altogether due to spiraling costs (Kroenick
& Gilmer, 1999; Kuttner, 1999), and this has most affected low-wage workers. From
1979 to 1998, the proportion of blue-collar and service employees who received
health coverage through their jobs declined 17 percentage points, from 63 to 46
percent, while the rate declined 9 percentage points for white collar workers (Medoff
et al., 2001). In a number of industries, the proportion of management and
professional employees receiving health benefits through their jobs remained steady
or improved.
38


For organizations that have chosen to continue to provide coverage, an analysis
conducted by Medoff et al. (2001) indicates that benefits payments by private sector
employers increased from $0.74 per hour in 1986 to $1.28 per hour in 1999.
However, after adjusting for inflation, the average real purchasing value of
contributions fell by 17 percent. Instead of contributing additional funds to fill the
gap, many employers have either dropped coverage or required that employees to
make up the difference by increasing the share of workers contributions. The decline
in the proportion of health insurance payments by employers was most severe for the
bottom third of low-wage earners, who received approximately one-half of the health
benefit compensation per hour as workers in the top one-third income bracket
(Medoff et al., 2001).
Other working poor have health insurance but are underinsured in that they have
medical needs that are not covered by their health plan. The underinsured have health
insurance but cannot afford to use it. Individuals are also considered underinsured if
high health plan co-payments cause them to go without needed medical treatment. In
addition, some health plans may create underinsurance by having requirements so
stringent and difficult to meet that they result in the denial of medically necessary
treatments (Kuttner, 1999). The growing financial burden of participating in
employer-sponsored plans has been associated with an increase in uninsurance rates,
as more than 80 percent of the increase in the uninsured over the past 5 years has
39


occurred among those with incomes greater than 200 percent of the federal poverty
level (Holohan & Kim, 2000).
For individuals below 200 percent of poverty, cuts in Medicaid and reductions in
private non-group coverage have caused an increase in the proportion uninsured
(Holohan & Kim, 2000). Over the past 5 years, the loss of employer-sponsored
coverage and cuts in Medicaid have resulted in an increase in the proportion of the
population that is uninsured. Since this has continued in the face of record economic
growth and expansion and record tax surpluses, it begs the question as to what the
national uninsurance rate might grow to during a time of prolonged economic
recession.
Numbers of Uninsured
Theoretically, healthcare coverage is available to all U.S. citizens through one of four
routes: Medicare for the elderly and disabled, Medicaid for low-income women and
children and individuals with certain disabilities, employer-sponsored coverage, or
self-purchased coverage for those who cannot obtain any of the previous three
(Friedman, 1991). However, according to Census Bureau figures, 41.2 million
residents lacked healthcare coverage for all of 2001 (U.S. Census Bureau, 2002).
This figure represented 14.6 percent of the U.S. population in 1999, and is a slight
increase in the percentage of uninsured (0.4 percent or 1.4 million people) from 2000
40


measurements. With the exception of 1999 when there was a slight decrease in the
percentage of uninsured, the proportion of Americans lacking health insurance has
increased every year for the past decade (U.S. Census Bureau, 2002). Because the
tools used by different government and foundation-supported surveys to measure the
numbers of uninsured are often different, estimates of the number of uninsured
Americans vary, as shown in Table 1. Most agree that these variations are due to
methodological differences in data collection procedures (Freeman et al., 1990).
The Census Bureau reports that a much larger number of people lack insurance for
part of any given year (U.S. Census Bureau, 1998a). Based on a federally-funded
study conducted from 1993 to 1995, this figure has been calculated to be about 71.5
million persons (U.S. Census Bureau, 1998a), while other estimates indicate that 50
to 60 million persons lack health insurance at some time during a calendar year
(Schroeder, 1996). Over a period of 2.25 years, 28 percent of the population was
reported to be without health insurance for at least 1 month (Blendon et al., 1992).
Among non-citizens, 42.6 percent of those residing in the U.S. in 1999 lacked health
insurance (U.S. Census Bureau, 2000). Some researchers suggest that it is difficult to
determine the number of residents without health insurance at any one time (Franks,
Clancy, & Gold, 1993; Long & Rogers 1995; Seccombe & Amey, 1995), and this is
especially true of those who are residing in the U.S. without legal documentation.
41


A national survey conducted in 1997 found that one-third of adults lacked health
insurance at some time in the previous two years, and that 57 percent of these
individuals had been employed full time (Kuttner, 1999; Ome et al., 2000). National
estimates conducted in the mid-1990s indicated that one-third of American adults
under the age of 65 have been uninsured for some period within the past 24 months
(Davis et al., 1995). Fifty-percent of Hispanics lack coverage for at least one month,
as compared with 37 percent of Blacks, and 25 percent of non-Hispanic Whites
(Kuttner, 1999).
Being medically uninsured is not always a short-term problem. A 1993 survey found
that two-thirds of the uninsured had been without coverage for over a year, and
almost half had been uninsured for more than two years (Davis et al., 1995). Those
most likely to have periods without coverage are earning low incomes and unable to
afford health insurance premiums (Davis et al., 1995; Kuttner, 1999). The lack of
insurance was reported to be due more often to the inability to afford insurance
premiums than to the loss of a job or to an employers failure to offer coverage
(Schoen & Puleo, 1998; Kuttner, 1999).
Policy attention for the uninsured has recently focused on children. Children without
health insurance have been found to be less likely to see a physician than children
with Medicaid (Davis, 1991; Kasper, 1987). Approximately 800,000 children were
42


reported to be taken to the emergency room for all their care in 1997 (Davis, 1991;
U.S. Census Bureau, 1998b). To counteract these trends, $24 billion in federal
monies were appropriated by Congress to the Childrens Health Insurance Program in
1999 (Kuttner, 1999). In some states, this program has been grafted onto Medicaid in
order to make it possible to expand coverage to otherwise non-qualifying children for
a small monthly premium, which is often waived. Other states have chosen to set up
parallel programs for children in low-income working families (Kuttner, 1999). The
program is available to immigrants, both legal and undocumented (Kuttner, 1999).
Although enrollment has been slow in some states, the program has been credited
with the recent decline in the number of uninsured children (Auge, 2000).
With respect to uninsured adults, however, relatively little is being done to meet their
needs. Nationally, 52.6 percent of those lacking health insurance in 2001 were from
25-64 years of age (U.S. Census Bureau, 2002). Non-elderly adults are 40 percent
more likely than children to be uninsured, and less than half as likely to have public
health insurance coverage (Spillman, 2000).
Colorados Uninsured
In the state of Colorado, approximately 633,000 citizens (15.8 percent of the
population) were reported to be without health insurance coverage during the period
from 1997 to 1999, according to the current population survey (Yondorf, 2001). This
43


represents a 0.6 percent increase over 1995 to 1997 figures. Between 16.8 percent
and 22.1 percent of the states population were uninsured at any time during the year
1999 (Yondorf, 2001). The elderly were most likely to have health insurance in
Colorado from 1997-1999, with only 1.5 percent lacking coverage (Yondorf, 2001).
A significant proportion of the states children (under age 18) did not have health
insurance from 1997 1999, with 14.1 percent lacking coverage (Yondorf, 2001).
More than 76 percent of uninsured persons in Colorado were between 18 to 64 years
of age. In 1999 it was reported that 21.4 percent of Colorado adults age 18 and older
had been without insurance for the previous 5 years or more (Yondorf, 2001).
Young adults (age 18 to 24) in Colorado were the most likely to be uninsured. On
average, 31.5 percent (120,961 persons) of young adults lacked health insurance from
1997-1999 (Yondorf, 2001). However, the greatest number of uninsured individuals
in Colorado were between the ages of 25 and 54 years, of which 316,790 were
uninsured (16.7 percent) on average from 1997-1999 (Yondorf, 2001).
Reasons for Increase in Numbers of Uninsured
There does not appear to be a clear consensus about why the numbers of uninsured
Americans has risen over the past two decades (Ome et al., 2000), from 21 million
(12.2 percent of the population) in 1977 (Berk et al., 1996) to over 41 million in 2001
(14.6 percent of the population) (U.S. Census Bureau, 2002). Certainly, the problem
44


is complex, given the nations patchwork scheme of healthcare coverage. A very
rapid increase in the number of uninsured occurred between 1980 and 1987, when
their numbers increased by nearly 25 percent (Long & Rogers, 1995). Numerous
shifts in the American workplace were occurring at that time, and many researchers
have focused on workplace insurance issues, since two-thirds of Americans with
health insurance have employer-sponsored coverage (Holahan & Kim, 2000).
Declines in employer-sponsored coverage from the 1970s to the mid-1990s have been
attributed to a variety of factors, including the shift away from manufacturing jobs,
increased part-time employment, increased self-employment, declining labor unions,
cuts in real wages, and rising healthcare costs (Acs, 1995; Holohan & Kim, 2000;
Kronick & Gillmer, 1999; Kuttner 1999; Long & Rogers, 1995).
Increasing Healthcare Costs
In 2001, healthcare represented approximately 14.0 percent of the United States
Gross Domestic Product (GDP) (Heftier et al., 2002), and over the last half of the 20th
century, per capita medical costs have increased by over 1,000 percent (Edge &
Groves, 1998). Unless major cost reduction systems are put into place, healthcare
costs are expected to grow to $16 trillion by the year 2030, accounting for nearly one-
third of the GDP (Edge & Groves, 1998). It is widely believed that such an
expenditure of the nations financial resources will be unsustainable.
45


Despite substantial expenditures in healthcare, one-sixth of Americans under the age
of 65 lack basic healthcare coverage. A primary factor in the rising numbers of
uninsured Americans is escalation in healthcare costs over recent decades. Between
1965 and 1982, total national expenditures for hospital care increased 964 percent,
from $12 billion to $135 billion, while the gross national product increased only 444
percent (Gibson, Levit, Lazenby, & Waldo, 1984). A rapid increase also occurred
from 1980 to 1989, when gross healthcare expenditures in the U.S. increased by
128%; that is more than a doubling of the cost of healthcare in a single decade
(Friedman, 1991). Gold (1998) explained the events that brought about skyrocketing
healthcare costs:
Public policy has led to a growth in the proportion of the population
that is insured and in the continuity of that coverage in the post-World
War II era. But the historical trade-off to gain provider support for
enactment of programs like Medicare and Medicaid was the inclusion
of policies that in effect gave providers a blank check to induce them
to support the program. These public policy trade-offs, in turn, have
influenced the evolution of the private insurance market (Marmor,
1974; Starr, 1973). With this has come an explosion in healthcare
costs that is now the prime impetus for the growth of managed care
and more competitively based market models.
In 1983, Congress passed the Medicare prospective payment system under which
payment rates were set in advance for each medical diagnosis (Edge & Groves,
1998). Because payments for a given illness were fixed, healthcare facilities finally
had a strong incentive to contain costs. By that time, however, nearly two decades of
excessive spending had occurred since the initiation of the Medicare program, which
46


was subject to abuse by providers and hospitals in an effort to maximize
reimbursements.
Kroenick (1999) has demonstrated that relative to personal income, per capita
healthcare spending approximately doubled from 1979 to 1995, while real wages
remained essentially the same. During the same period, the proportion of uninsured
workers closely paralleled the steady increases in per capita healthcare spending
(Kroenick, 1999). Health spending increases were passed onto employers and
employees through increased insurance premiums. The inability of both employers
and employees to absorb health insurance premium increases far in excess of their
real incomes largely explains the steady rise in the proportion of uninsured (Kroenick,
1999).
In healthcare, unlike other industries, technology has not brought about a decrease in
personnel needs. Rather the opposite has happened (Edge & Groves, 1998). As new
technologies emerge, new technical specialists must provide the necessary services.
With ever-improving technologies and medicines, the healthcare infrastructure
continues to expand and consume resources. The pressure to rapidly develop of new
technology is a long-standing trend in American medicine, fueled by the publics
desire for new treatments, and the powerful medical-industrial complex that has
developed to fulfill those needs.
47


Efforts to Control Healthcare Costs
Health insurers have taken a number of measures to mitigate the impact of rising
costs, including dramatically increasing healthcare premiums. They have become
more selective about whom they cover, the comprehensiveness of that coverage, and
the types of pre-existing conditions they choose to exclude. It has been reported that
three percent of the uninsured cannot obtain health insurance due to their history of
poor health, illness, or age (Davis et al., 1995). For those who do not have the benefit
of group or public coverage, obtaining insurance is dependant upon having a
sufficiently high income and very good health status (Friedman, 1991). Medical
underwriting, experience rating, refusal to cover those deemed uninsurable,
cancellation of policies on short notice, and unaffordable premiums are frequent if not
universal barriers to those seeking individual coverage (Friedman, 1991). The
presence of any preexisting health conditions, even minor ones, are used as by
insurers as justification for denial of coverage. This effectively excludes those who
need coverage the most, the poor, the sick, and those unable to acquire workplace
insurance (Friedman, 1991).
Insurers have also focused on controlling costs by eliminating waste and inefficiency
in healthcare delivery systems, and by implementing payment structures that provide
a disincentive to providers who are inefficient in their use of resources. The primary
48


expression of this has been termed managed care, and over recent decades it has
been implemented across healthcare systems in most major American population
centers. Managed care organizations have radically changed the face of American
healthcare, as they have sought to establish primary care providers as gatekeepers of
resource utilization, and to remove the financial incentives that resulted from the
traditional practice of allowing providers to have exclusive direction over healthcare
utilization. These measures have been credited with slowing the rise of healthcare
costs by placing financial disincentives on excessive healthcare expenditures.
However, their effect on the quality of healthcare has been the subject of much
controversy and debate.
Profile of the Uninsured
In a U.S. Census Bureau report, Campbell (1999) reported that those most likely to
lack health insurance were:
18-24 years of age
Possessing lower levels of education
Poor
Hispanic
Part-time workers
Bom in another country
49


Other characteristics of the uninsured included the following (Campbell, 1999):
Approximately one-half of poor full-time workers lacked health
insurance.
Twenty-five percent of households with incomes of $25,000 or less
did not have health insurance.
Medicaid insured 14.0 million poor people, but about one-third of all
poor people (11.2 million) had no health insurance.
A much higher proportion of Hispanics (35.3 percent) were without
health insurance than were non-Hispanic Whites (11.9 percent).
Income is a major factor in estimating levels of uninsurance. A 1986 survey
sponsored by the Robert Wood Johnson Foundation found that those earning 100-150
percent of the poverty level were uninsured at about the same rate (28.6 percent) as
those at or below the poverty level (Freeman et al., 1990). Above 150 percent of
poverty, however, only 6 percent of persons had no health insurance. The group
earning 100-150 percent of poverty continue to be of special concern since they
represent the working poor who are unable to qualify for Medicaid but generally
unable to afford health insurance. A survey conducted in Massachusetts found that
the majority of the uninsured were living in households where an adult is employed
(84 percent) and where the income is above the poverty level (66 percent). Eligibility
changes in Medicaid can greatly affect the ability of persons on the fringes of poverty
50


to obtain health insurance (Freeman et al., 1990), allowing them to qualify for
assistance at some times but not others.
Young adults (aged 18 to 24 years) are the demographic group most likely to be
uninsured. McManus, Greaney, and Newacheck (1989) found that the young
uninsured are typically male, Hispanic or Black, poor or near-poor, unemployed, high
school dropouts, living with others, and residing in the South or West. Income and
educational attainment were the two most important predictors of uninsurance for 18
to 24 year-olds, and cost was cited as the primary reason for not purchasing health
insurance (McManus et al., 1989).
Access for Rural Residents
Although the communities in which the clinics participating in this project were
located do not meet the definitions of rural by the U.S. Census Bureau, or non-
metropolitan statistical area by the Federal Office of Management and Budget, they
are small towns with strong rural influences. Each is located in an agricultural area,
and serves the residents of small farming communities.
In comparison to urban residents, rural residents are more likely to work for small
firms that do not offer health insurance, and they have lower incomes, making it more
difficult to purchase private insurance (Frenzen, 1993). Rural areas also have higher
51


rates of unemployment than urban areas, and rural residents are more likely to work
low-wage jobs (Frenzen, 1993; McGranahan, 1988). As a result, rural residents are
more likely to be poor and uninsured (Frenzen, 1993; Davis, 1991). In addition,
states with large rural populations have been found to have more stringent standards
for Medicaid than urban states (Frenzen, 1993), making it difficult to obtain public
healthcare coverage.
Healthcare utilization for rural residents is generally lower than for urban residents,
regardless of insurance status (Frenzen, 1993; Mueller, Patil, & Boilesen, 1998).
Since there are fewer physician and hospital services available in rural areas,
residents may face substantial barriers to accessing healthcare services. Barriers
include long travel distances because rural populations may be too small or dispersed
to support health facilities (Frenzen, 1993; Miller, Holahan, & Welch, 1995; Mueller
et al., 1998), and little choice of healthcare providers that may dissuade some
residents from seeking medical services. Rural residents rely heavily on community
health centers for primary care. In an analysis of 3 national surveys that were
conducted in 1994, rural residents were found to be approximately 4 times more
likely to be seen at a community health center than non-rural residents (Forrest &
Whelan, 2000).
52


Rural minority populations have more difficulty obtaining medical services than rural
or urban whites. Combining the factors of rural residency, lack of health insurance,
and ethnicity considerably lowers the likelihood of having seen a physician during the
previous year (Mueller, Ortega, Parker, Patil, Askenazi, 1999). Rural Latinos were
found to be 30 percent less likely than urban whites to have seen a physician in the
previous 12 months, and uninsured rural Latinos were 70 percent less likely (Mueller
et al., 1999). Although residency and ethnicity are significant predictors of health
utilization, the presence or absence of health insurance has consistently been found to
be the primary driving factor (Mueller et al., 1999). However, as has been previously
discussed, undocumented residency status and Hispanic ethnicity are associated with
uninsurance.
Uninsured by Choice
Although approximately 60 percent of the uninsured have incomes below 200 percent
of the poverty line, a significant proportion have incomes above this level, and one-
quarter have family incomes of at least 300 percent above poverty (Yegian et al.,
2000). A telephone survey was conducted to determine the income point at which
individuals purchase health insurance for themselves, and why some choose to remain
medically uninsured (Yegian et al., 2000). Interviews were conducted with uninsured
households in California with incomes of at least 200 percent of poverty. Ten percent
worked for an employer that offered health benefits to them, but cost was the
53


predominant reason for declining coverage. Forty-three percent felt that health
insurance was not a good value for the money, and the majority believed that
coverage cost about twice as much as they were willing to pay. When generic health
plans and prices were described to the interviewees, 53 percent reported that they
would purchase one of the options. Thus, misperceptions about the cost of insurance
and lack of awareness about available health plan options may be significant factors
in the lack of health insurance for the non-poor (Yegian et al., 2000).
Consequences of Being Uninsured
Health Services Utilization
A lack of health insurance has been associated with reduced access to medical care
(Hubbell et al., 1989; Franks et al., 1993). The analysis by Mueller et al. (1998) of
National Health Insurance Survey (NHIS) data indicates that the uninsured were half
as likely as the insured to use a physicians services, and numerous other studies have
demonstrated that uninsured persons use fewer health services than insured persons
(Freeman et al., 1987; Anderson, Chen, Aday, & Cornelius, 1987; McManus,
Greaney, Newacheck, 1989; Patrick et al., 1992). After adjusting for health status,
persons without health insurance have been shown to visit a physician less often than
the insured (Hafner-Eaton 93; Franks 93; Freeman 90). In analysis of data from the
National Medical Care Expenditure Survey, Americans with health insurance
received 54 percent more ambulatory care and 90 percent more inpatient hospital care
54


than those without health insurance (Davis, 1991; Davis & Rowland, 1983). One
year after termination of Medi-Cal (Californias Medicaid program) for some 270,000
medically indigent adults, only half could identify a regular doctor, and only two in
five thought they could obtain care when they needed it (Donaldson et al., 1996;
Lurie et al., 1984; Lurie et al., 1986).
Among those with chronic or serious illnesses, the uninsured are less likely than the
insured to receive medical care (Freeman et al., 1990). In a study of medically
uninsured patients in Orange County, California, patients with serious medical
problems were found to be less likely to obtain medical care than those with less
serious health problems (Akin, Rucker, Hubbell, Cygan, & Waitzkin, 1989). This
was attributed to the fact that community health centers could not accommodate
patients in need of intensive treatment or hospitalization. Chronically ill uninsured
have been shown to be half as likely to have seen a physician over the previous 12
months than their insured counterparts, and acutely ill uninsured were almost 33
percent less likely to receive physician care (Hafner-Eaton, 1993). When controlling
for differences in family characteristics, uninsured families have been found to be less
likely to make outpatient visits and have hospitalizations than insured families,
despite having more chronic health conditions and reporting a greater need for health
services (Patrick et al., 1992).
55


Illness Type and Utilization
The uninsured have been shown to cluster by type of illness. Although all insured
persons have a much lower likelihood of receiving physician care than the insured,
the acutely ill receive care at higher rates than either the chronically ill or well
populations (Hafner-Eaton, 1993). Perhaps because the acutely ill are perceived to
have a more immediate health risk over the short-term, they are able to obtain care
despite their lack of health insurance. In addition, acutely ill patients may be more
insistent than the chronically ill in seeking healthcare due to the compelling
experience of their own symptoms and the more visible nature of their illness.
Physicians may feel more motivated to treat the acutely ill because a complete cure
may be achieved. Chronically ill patients may also be more likely to have multiple
barriers (such as functional limitations) and have reduced incentives to seek services,
because care will not lead to a complete cure (Hafner-Eaton, 1993). The chronically
ill uninsured may suffer serious flare-ups that might be avoidable through routine
primary care (Weissman, Gatsonis, & Epstein, 1992).
Uninsured persons have been found to be more likely to have potentially avoidable
hospitalizations in comparison to insured (Billings & Teicholz, 1990; Weissman et
al., 1992). The medically uninsured are more likely to be treated in a in a hospital
outpatient clinic or emergency room for receive routine care that could be provided
through a primary care office (Freeman et ah, 1990).
56


Delaying Care
The uninsured are more likely to delay or go without healthcare for serious medical
symptoms more often than those with health insurance (Franks et al., 1993;
Weissman, Stem, Fielding, & Epstein, 1991; Ome et al., 2000). Going without
needed healthcare can affect both health status and the cost of care that may
eventually be needed when neglected health problems worsen (Davis et al., 1995).
Uninsured patients have been found to be sicker at the time of hospital admission than
the insured (Billings & Teicholz, 1990; Weissman et al., 1992), which may indicate a
lack of timely or routine visits to a primary care physician during the early stages of
illness.
In a 1993 study, Davis et al. (1995) found that 13% of all Americans reported that
they did not receive healthcare which they believed was necessary at some time over
the previous 12 months, and 30 percent postponed seeking medical care for financial
reasons. Among uninsured adults, one-third indicated that they did not receive
needed care, compared to 9 percent of insured. Seventy-one percent of uninsured
delayed seeking healthcare because of financial reasons, compared to 23 percent of
insured (Davis et al., 1995).
57


In a study of delayed care, Weissman et al. (1991) found that in comparison to other
patients, the odds of delaying care because of cost were more than 12 times higher for
patients who were both poor and uninsured. The cost of care was an important, but
not the primary reason for delaying care for many poor and uninsured. The most
common reason that patients delayed care was that they believed that the problem
would go away or was not serious (Weissman et al., 1991). This may suggest that the
poor and uninsured may have relatively low ability to discriminate between serious
and self-limiting conditions due to their relative unfamiliarity with healthcare. It may
also reflect cultural differences in how symptoms are perceived and managed by the
individual.
Uninsurance and Health Status
Not having insurance has been shown to negatively affect overall health status
(Hafiier-Eaton 1993; Lurie et al., 1986). The lack of access to timely and appropriate
medical care for preventable or early-stage illnesses may lead to chronic or refractory
health problems (Davis & Rowland, 1983). A Massachusetts study found that on two
key measures of health status general health and disability the uninsured reported
significantly more illness than those with insurance (Blendon et al., 1992). Forgoing
needed care can ultimately result in increased risk of mortality and has been causally
related to a higher mortality rate, possibly because of decreased access and lower
quality of care (Franks et al., 1993).
58


Income and Health Status
Not having health insurance is highly correlated with low income, which itself has
been shown to be associated with poor health (Anderson et al., 1987; House, Kessler,
& Herzog, 1990; Marmot, Rose, Shipley, & Hamilton, 1978; Marmot et al., 1991;).
Two large-scale studies of British civil servants have demonstrated a steep inverse
relationship between social class and mortality from a wide range of diseases
(Marmot, et al., 1978; Marmot et al., 1991). Differences in health behaviors such as
smoking, obesity, physical activity, blood pressure, and plasma cholesterol levels
could only partially explain differences in mortality. Because the subjects in this
British study had state-provided healthcare, access to health insurance was not an
issue.
Studies conducted in the United States have found somewhat similar results, although
because the U.S. does not have a national healthcare system, it is not possible to hold
access to healthcare constant. In a U.S. study of educational differentials and
mortality, Feldman, Makuc, Kleinman, and Comoni-Huntley (1989) found that
between 1971 and 1984, less educated men were at substantially greater risk of
mortality than their more educated counterparts. The rate of heart disease mortality
was the primary differential between educational groups. Better access to healthcare
was suggested as a possible reason that the more educated men lived longer (Feldman
59


et al., 1989), although access variables were not measured. House et al. (1990) found
that lower socio-economic status was associated with higher levels of morbidity and
more functional limitations. Access to medical care was not measured, although it
was speculated that access may explain some of the differential, and it was
acknowledged that access differences have been exacerbated over the past decade as
lower income workers have become increasingly likely to be uninsured (House et al.,
1990).
The burden of morbidity has been found to be substantially greater for the low-
income Americans. Based on the 1984 National Health Interview Survey data,
Anderson et al. (1987) reported that people with low incomes were more likely than
those with high incomes to have high blood pressure, eye diseases, respiratory
conditions, arthritis, hearing impairment, diabetes, heart disease, stroke, to sustain
injuries at home, and to have longer hospital stays.
Sub-standard Care
The poor have also been shown to receive substandard medical care, often because
they lack health insurance (Burstin, Lipsitz, & Brennan, 1992; Franks et al., 1993). In
a study of over 30,000 hospital medical records, Burstin et al. (1992) found that the
uninsured have a two-fold greater risk of negligent care than those with insurance,
even when patient race, income, and gender were controlled for in the analysis. The
60


majority of negligent injury to uninsured patients occurred in emergency departments,
where a disproportionate number of uninsured receive primary care. The pace of
diagnosis and treatment in emergency departments is often accelerated, which may
explain the higher incident rates (Burstin et al., 1992, Leape et al., 1991).
Many uninsured patients require high-risk obstetrical care, and others have serious
diseases such as HIV, kidney disease, and heart disease, requiring complex
treatments, costly medications, and surgery (Thurman et al., 2000). However,
depending on insurance status, there are differences in the ways that admitted patients
are treated in hospitals. The hospitalized uninsured are less likely to get specialized
services, and are more likely to die while hospitalized. In a study of hospital abstract
data from a national sample of nearly 600,000 admitted patients, the uninsured had
1.2 to 3.2 times higher death rate than insured patients when controlling for health
status (Hadley, Steinberg, & Feder, 1991). The same study also showed that the
uninsured were also less likely to receive high-cost or high-discretion procedures than
insured patients. The uninsured are also more likely to have a shorter hospital stays
than insured parties (Davis & Rowland, 1983; Freeman et al., 1987; Hadley et al.,
1991).
Woolhandler and Himmelstein (1988) reported that the socioeconomically
disadvantaged are most likely to fail to receive recommended screening tests, and the
61


primary predictor of inadequate screening was lack of health insurance. Thus, those
who carry the greatest burden of illness in society are most likely miss important
health screening exams, because they are also most likely to lack health insurance.
Relationships between Poverty, Access, Insurance and Health Status
As is evidenced by the literature, a substantial number of studies describe the
relationships between poverty, access, health insurance, and health status. While
associations between the factors have been demonstrated, causality generally has not,
since the great majority of studies have used cross-sectional designs.
The literature has shown that the following relationships exist:
Poverty is linked to poor health;
Lack of health insurance is linked to access difficulties;
Lack of access is linked to poor health;
Lack of health insurance is linked to poor health;
Lack of health insurance is linked to poverty.
62


The implied causal chain is shown in Figure 1.2.
Figure 1.2. Linear relationships between income, insurance, access, and health status
Low Income (poverty)
1.
Lack of Health Insurance
1
Restricted Access to Health Services
Poor Health Status
The logic of the chain is as follows: because health insurance must be purchased, low
income could lead to a lack of health insurance. Because insurance is needed ensure
full access to healthcare, the absence of health insurance could restrict access to
health services. Finally, accessing healthcare when necessary is recognized as an
important component of maintaining health.
There are some challenges to the assumptions behind this model. McKeown (1979)
demonstrated that in contrast to public health and sanitation, medicine has had a very
limited effect in improving health over the past century. However, while the
contributions of public health are highly significant, it is generally agreed that
medical care and preventive services have substantial value in the improvement of
health status (Patrick, Stein, Porta, Porter, & Ricketts, 1988), albeit oftentimes at a
significant financial cost to society.
63


The direction of causality has also been brought into question. While poverty is
clearly associated to morbidity and premature mortality, there has been considerable
debate over whether poverty causes poor health or poor health causes poverty (Patrick
et al., 1988). The link between poverty and health appears to be bi-directional
(Patrick et al., 1988), as poor health can certainly lead to poverty, just as poverty can
lead to poor health. Thus, these concepts might be better addressed by a cyclical
model, in which the variables and causal direction shown above remain the same, but
rather than terminating the chain, poor health status leads back to poverty (see Figure
1.3). A patient could enter the cycle at any point, at which the cause and effect
relationships would begin.
Figure 1.3. Cyclical relationships between income, insurance, access, and
health status
Lack of Health Insurance
Low Income (poverty)
Restricted access to health services
Poor health status
64


Patrick et al. (1988) questioned the relationships between income, access, and health
status and surmised that larger system factors should enter into such a model. In a
large-scale study that involved community health centers across the country in which
significant resources were allocated to achieve higher levels of access to medical
care, providing access was not found to be effective in raising the poor health status
of low-income patients. Similar results have been found in studies of Medicare
patients. Patrick et al. (1988) submits that low health status may be the result of the
problems of a class of poor people that are the means by which they are able to cope
with a more privileged world. Patrick (1988) proposed that a theory of social
causation should be developed, which includes the beliefs, attitudes, values, and
behaviors of the poor, the medical establishment, and mainstream society. The author
explains that an understanding of the roles of economic and political power as
potential mechanisms for causing inequality are essential as well.
Patrick et al. (1988) also suggests that the implementation of such a sweeping,
system-wide undertaking is unlikely, and recommends that these problems can be
addressed more easily through the widespread adoption of community oriented
primary care (COPC). COPC is the provision of primary care that involves
understanding the unique needs of a local community, and subsequently developing
strategies that draw upon community resources in order to address those needs. It
provides a model for primary care practice that identifies a specific population in
65


need, and identifies and addresses the health problems in ways that are specific to that
population (Nutting, 1990). Through COPC, the beliefs, values, and health behaviors
within a community can be understood and responded to within the context in which
patients live their lives. Programs targeted at specific health problems can then be
designed to improve the health outcomes in that community. In addition, longitudinal
measurements can be conducted to assess outcomes rather than simply improving
process variables, which do not always result in improved health status. Moreover,
COPC can result in strengthened interpersonal connections within the community.
Communities that build a rich network of non-market relationships provide a sense of
trust, sharing, and cooperation that increases the efficiency of human relationships
(Anderson & Boumbulian, 1997). Such efforts can indirectly prevent disease by
improving and enriching the quality of the communitys social environment, and
increasing social support between individuals.
The U.S. Hispanic Population
Demographics
Persons of Hispanic origin are the countrys the fastest growing ethnic group
(Hubbell, Waitzkin, Mishra, Dombrink, & Chavez, 1991). As of Census 2000, the
current U.S. population numbered 281.4 million, and persons of Hispanic ethnicity
represented 12.5 percent, or 35.3 million persons (U.S. Census Bureau, 2000). An
additional 3.8 million Hispanic persons were enumerated in the Commonwealth of
66


Puerto Rico. The nations Hispanic population has increased by 57.9 percent since
1990, compared with an increase of 13.2 percent for the total U.S. population (U.S.
Census Bureau, 2000), bringing the Hispanic population nearly equal to that of
African-Americans. Birth rates for Hispanics are three to four times that of the non-
Hispanic white population (Munoz, 1988).
Between 1990 and 2000, the Hispanic population increased by 77 percent in
Colorado, to 754,000 people (Aguilar, 2001). Hispanic persons now represent 17.1
percent of the states population (U.S. Census Bureau, 2000).
Mexican Americans are the most populous Hispanic subgroup, and their numbers
have risen faster than other Hispanic subgroups. From 1970 to 1980, the Mexican
American population in the U.S. grew by 93 percent; from 1980 to 1990 it grew by 54
percent; and from 1990 to 2000 it increased by 52 percent to 66.1 percent of all
Hispanics. Census 2000 population figures for other Hispanic subgroups in the U.S.
were the following: Puerto Rican (9.0 percent), Central American (14.5 percent),
South American (3.8 percent), Cuban (4.0 percent), Dominican (2.2 percent),
Spaniard (0.3 percent), and the remainder were of other Hispanic origins (U.S.
Census Bureau 2000).
67


In the 2000 Census, 43.5 percent of Hispanic persons lived in the western U.S., where
they accounted for 24.3 percent of the population. Colorado is currently ranked sixth
among states in the proportion of Hispanic residents (17.1 percent), and ninth in the
number of Hispanic residents (735,601). The vast majority of Colorados Hispanic
residents are of Mexican origin.
According to the 2000 census, 35.7 percent of Hispanic persons were less than 18
years of age, compared with 23.5 percent of non-Hispanic Whites. Hispanic persons
of Mexican origin had the highest proportion less than age 18 (38.4 percent). The
proportion of those age 65 and older ranged from just 4.5 percent for Mexican-
Americans to 21.0 percent for Cuban-Americans (U.S. Census Bureau, 2000).
Hispanic persons in the U.S. (on average) are poor. In 1999, 22.8 percent of
Hispanics were living below the poverty level, compared with 7.7 percent of non-
Hispanic Whites (U.S. Census Bureau, 2000). Hispanic children under age 18 were
much more likely than non-Hispanic children to be living in poverty (30.3 percent
versus 9.4 percent). Unemployment is high among Hispanic people. In the 2000
Current Population Survey, 6.8 percent of Hispanic persons in the civilian labor force
over the age of 16 were unemployed compared with 3.4 percent of non-Hispanic
Whites. Among Mexican-Americans, 7.0 percent were unemployed (U.S. Census
Bureau, 2000). Among full-time, year-round workers in 1999, 76.7 percent of
68


Hispanic persons earned less than $35,000 annually, in comparison to 50.7 percent of
non-Hispanic Whites.
Educational attainment among U.S. Hispanics is low, with more than two in five not
graduating from high school (U.S. Census Bureau, 2000). More than one-quarter of
Hispanic persons had less than a ninth-grade education (27.3 percent) compared with
only 4.2 percent of non-Hispanic Whites. Mexican-Americans were least likely to
have graduated from high school (51.0 percent), and were also least likely to have not
attained a bachelors degree (6.9 percent), in contrast to 23 percent of Cuban-
Americans, and 28.1 percent of non-Hispanic Whites.
Hispanic Persons and Health Insurance
Rapid population growth in the Hispanic sector coupled with high rates of poverty
and uninsurance have resulted in an increasing need for health services for low
income Hispanic persons. Hispanics are less likely to have than any other ethnic
group to have health insurance nationally (Hubbell et al., 1991) and in Colorado.
Nationally, 35.3 percent of Hispanics were without health insurance in 1999, a
substantially greater proportion than for Whites, of which 11.9 percent were
uninsured in 1999 (U.S. Census Bureau, 2000). Between 1977 and 1992,40 percent
of the increase in uninsured persons in the U.S. was accounted for by individuals of
Hispanic origin, and Hispanics of Mexican origin alone accounted for 27 percent of
69


the increase (Berk, et al., 1996). Approximately 26 percent of Hispanics in Colorado
lacked health insurance from 1997 1999, compared to 20.8 percent of non-Hispanic
African Americans and 13.3 percent of non-Hispanic Whites (Yondorf, 2001).
A major reason that many Hispanic persons in the United States lack health insurance
coverage is the unavailability of employer-based coverage for working-age Hispanic
groups (Berk et al., 1996; Trevino, Trevino, Medina, Ramirez, & Ramirez, 1991).
Recently arrived immigrants often work in low-wage jobs with few fringe benefits.
Data from the 1992 National Health Insurance Survey (NHIS) indicate that more
recent Hispanic immigrants have, by far, the highest probability of being uninsured
(Berk et al., 1996). Among Hispanics who have been in the United States for less
than 5 years, 62% were uninsured. Even for Hispanics bom in the United States,
however, the percentage of uninsured (23.4 percent) is higher than the U.S. average
(Berk et al., 1996).
Hispanic Health Status
Despite the steady improvement in the health status of Americans generally, Hispanic
persons in the United States have continued to be at greater risk for health problems
(Astrachan & Scherl, 1991; Solis, Marks, Garcia, & Shelton, 1990). Hispanics suffer
from diseases of neglect and limited access to health services (Munoz, 1988; Hubbell
et al., 1991). A report by the Secretary of Healths Task Force on Black and Minority
70


Health concluded that Hispanic persons suffered excess morbidity and mortality from
many conditions (Munoz, 1988). Mexican-Americans have three to five times the
incidence of diabetes mellitus compared to whites, and were also found to have much
lower medical surveillance rates for certain common female malignancies, such as
physicians breast examinations for cancer (Munoz, 1988). Some health problems
have been shown to be less severe among new immigrants than in subsequent
generations of Mexican-Americans.
Hispanic Healthcare Utilization
Deficiencies have been reported for Hispanic persons on important measures of
access, such as having a regular source of healthcare or having at least one
ambulatory office visit during the previous year (Hubbell, et al., 1991). Access to
care has been strongly related to utilization, as in the overall population. In
particular, having a routine place for care and having health insurance coverage
produced the strongest effects (Estrada, Trevino, & Ray, 1990). In studies of access
conducted in 1986, Latinos were found to have significantly lower measures of access
than Anglos. A California study found that there was no regular source of care for 31
percent of Latinos and 17 percent of Anglos (Hubbell et al., 1989). A 1986 national
survey yielded similar results, reporting that 31 percent of Latinos and 19 percent of
Anglos lacked regular source of care (Freeman et al., 1987). Compared with the
urban white population, ethnic subgroups have been found to be less likely to use a
71


physician, and Latinos are one of the subgroups most negatively affected (Mueller et
al., 1998). In an analysis of the National Health Interview Survey data, urban Latinos
were 10 percent less likely than urban whites to have seen a physician in the previous
12 months, and rural Latinos were 30 percent less likely (Mueller et al., 1998).
Mexican-Americans were less likely than non-Hispanic Whites to have routine
medical check-ups, dental and eye examinations, prenatal care, and family planning
services (Solis et al., 1990). In 1994, the number of primary care visits made
nationwide were 20 percent lower for Hispanic persons compared with White non-
Hispanic persons (Forrest & Whelen, 2000). In the same analysis, Hispanics were
also found to receive care from community health centers more often than any other
ethnic or racial group.
Barriers to Accessing Healthcare for Hispanics
Persons without health insurance coverage often experience greater difficulty in
obtaining access to healthcare, and lack of healthcare access leads to unfavorable
healthcare outcomes (Trevino et al., 1996). Findings from the Hispanic Health and
Nutrition Survey (HHANES) indicated that approximately one-third of the adult
Mexican American population had reported barriers to obtaining healthcare during
their most recent medical encounter (Estrada et al., 1990). These barriers to receiving
healthcare were significant in that three-fourths of the time the barriers were
sufficient to prevent persons from obtaining the medical care they sought (Estrada et
72


al., 1990). Mexican Americans with low educational levels and low incomes have
reported more problems accessing healthcare than other Mexican Americans (Estrada
et al., 1990). Spanish-speaking, foreign-bom, and less acculturated Mexican
Americans were found to be the most likely to be prevented from obtaining the care
they sought. Predominant barriers to healthcare were low income, younger age, less
acculturation, lack of health insurance coverage, functional limitations, and poorer
perceived health status (Estrada et al., 1990).
Mexican Americans have been found to make less use of healthcare than White and
Black non-Hispanics, Puerto Ricans, Cuban Americans and other Hispanics. Barriers
reported by Mexican Americans include, but are not limited to, language and cultural
differences, lack of transportation, geographic inaccessibility, financial constraints
such as the cost of healthcare and limited health insurance coverage, and isolation
from the mainstream culture (Estrada et al., 1990). Findings from the HHANES
survey indicated that one in five adult Mexican Americans reported that the cost of
medical care was a barrier to their receiving care. The concern over the cost of
healthcare among Mexican Americans would appear to be warranted, given that they
have the highest rate of being uninsured for medical expenses in the country (Estrada
et al., 1990). A smaller proportion of the adult Mexican American population
reported problems accessing medical care due to the lack of healthcare services
during nonworking hours (e.g. losing pay from work, hours of operation were
73


inconvenient, care was not available when needed). It may be that access to medical
care is in some part dependent upon the availability of expanded hours of operation
among healthcare providers due to the greater propensity of Mexican Americans to be
employed in hourly-paid service occupations (Estrada et al., 1990).
With respect to barriers related to acculturation, language but not ethnic identification
was the important predictor of utilization for Mexican Americans (Solis et al., 1990).
Two general hypotheses have been offered to account for Hispanics underutilization
of preventive health services (Solis et al., 1990). One posits that the level of
acculturation influences Hispanics utilization patterns. Studies examining the effect
of acculturation suggest that aspects of language may be especially important to our
understanding of Hispanics use of medical screening programs (Quesada & Heller,
1977). A second hypothesis attributes use of preventive health services to access to
care. Access can be defined in terms of factors that influence the ease with which
medical care can be obtained (Anderson et al., 1987). Such factors include the
availability of health insurance, having a routine place of care and a regular provider,
and its proximity to residence (Solis et al., 1990). After controlling for age,
education, and income, Solis and colleagues (1990) found that utilization of
preventive services for Hispanics was more strongly predicted by access to care than
by acculturation. After controlling for age, sex, employment status, insurance status,
income, and health status, Hubbel (1991) found that ethnicity failed to be a strong
74


predictor of access. These findings stress the importance of socioeconomic factors as
barriers to access, which result in health service underutilization.
Undocumented Immigration
It has been estimated that at least 5 million undocumented immigrants reside in the
United States, with an annual increase of approximately 300,000 (Thurman et al.,
2000) . Other estimates have suggested that as many as 11 million or more illegal
immigrants live in the U.S. (Krikorian, 2001). Enforcement of immigration laws
against illegal immigrants, the majority of whom are from Mexico, is limited. The
great majority of the undocumented are low-skill workers willing to accept low-
paying jobs without fringe benefits. Employment sectors such as agriculture and
construction have become substantially dependant upon these workers (Martin,
2001) . When significant numbers of illegal workers have been identified by the INS,
it is reported that members of Congress have sought to influence the agency to reduce
pressure on the affected industries within their electoral districts (Krikorian, 2001).
Hospitals in the border states of California, Arizona, Texas, New Mexico and Florida
are most affected by immigration. In Florida, a teaching hospital recently incurred
$312 million in charges in admissions and visits by treating undocumented residents
over a 3-year period. Of that, just $70 million (22 percent) was reimbursed (Thurman
et al., 2000). In the recent past, the states of California, Florida, and Arizona have
75


each sued the federal government for allowing illegal aliens to enter and remain in the
country, and for not providing funding for adequate public services to support them
(Pallarito, 1994). These lawsuits were unsuccessful.
In the mid-1990s, there was a backlash against un-funded federal mandates requiring
states to provide services to undocumented immigrants. State officials in California,
Arizona, and Florida protested a provision of the 1986 Omnibus Reconciliation Act
(OBRA86) that requires hospitals to treat all patients seeking emergency medical
services, regardless of patients financial or immigration status (Berk et al., 1996). In
a study completed by Los Angeles County, the net cost to the County for providing
healthcare to immigrants was approximately $350 million per year, with
undocumented immigrants accounting for half of the costs (Evans, 1995).
Immigrants used more services than were paid for by their contributions in taxes and
other revenues to the County. Equally troubling was the discovery that the native-
born population also used more services than they generated in taxes and other
revenues, in excess of $200 million (Evans, 1995).
It has been suggested that a generous policy toward illegal immigrants serves to
increase the rate of illegal immigration (Thurman et al., 2000). In addition to jobs,
some illegal immigrants may come to the U.S. in search of healthcare services
unavailable in their native countries. According to Thurman et al. (2000), the word
76


on the streets in many Latin American cities is that you can get good, free healthcare
in the U.S.; you just have to get there. Sick individuals may feel that they have little
to lose and much to gain by entering the U.S. illegally in an attempt to obtain needed
health services. There have been documented cases of people entering the country
specifically to obtain medical care, and when it is completed, returning home without
paying (Pallarito, 1994). This type of medical immigration may be self-
perpetuating (Thurman et al., 2000), as others learn about compatriots who entered
the U.S. illegally and obtained high quality health services.
While anecdotal evidence is engaging, research has not supported the assertion that
undocumented workers enter the U.S. primarily for healthcare or social services.
Berk, Schur, Chavez, and Frankel (2000) conducted a survey of undocumented
residents in El Paso, Houston, Fresno, and Los Angeles. One-half of undocumented
Latinos cited that employment was their most important reason for immigrating. The
second most frequent reason cited was to unite with family and friends. An
insignificant fraction indicated that social services were their primary reason for
immigrating.
77


In addition, the work by Berk, Schur, Chavez, and Frankel (2000) indicated the
following:
The survey findings indicated that undocumented Latinos were less
likely than the overall population to use ambulatory health services,
less likely to be admitted to a hospital, but more likely to be admitted
to a hospital for childbirth (Berk et al., 2000). Their level of
participation in government programs was fairly significant.
Undocumented residents are technically ineligible for a number of
programs (non-emergency Medicaid, Aid for Dependant Children,
Social Security, Food Stamps), however many undocumented workers
indicated participation, apparently as a dependant of a legal resident or
citizen or because they have a child who is a legal citizen. Texas
stringent Medicaid program had very low levels of participation (less
than 3 percent), but Californias program had much higher levels, with
ten percent of undocumented participating in Medicaid in Los
Angeles, and 25.5 percent in Fresno. Thus, Californias generous
Medicaid plan incurs significant costs in providing services to
undocumented persons, although those costs represent just 4 percent of
Medicaid eligibles statewide. Rates of participation in Aid for
Families with Dependant Children (AFDC) were also significantly
higher in California than Texas. Los Angeles had the highest AFDC
participation, with 17.7 percent of undocumented Latinos
participating, while Houston had the lowest, with just 1.6 percent
participating. Participation in the Food Stamp program also varied
greatly, from a high of 48 percent in El Paso, to a low of 8.6 percent in
Houston. Participation in the Supplemental Nutrition Program for
Women, Infants, and Children (WIC) was high in all sites, with one-
quarter to one-half participating. Participation in public education was
high, with rates ranging from 40 percent in Houston to 67 percent in El
Paso. Approximately 90 percent of the children of undocumented
immigrants reported receiving free or reduced-price lunches through
their school. Across the sites, 42-58 percent of undocumented Latino
adults had at least one child who is a U.S. citizen. This has relevance
to public school attendance, AFDC, Food Stamps, and Medicaid.
Berk, Schur, Chavez, and Frankel (2000) concluded that excluding undocumented
workers from government-funded health services is unlikely to reduce illegal
78


immigration, and is likely to affect the well-being of their children (Berk et al., 2000).
However, while utilization of health services was low with the exception of
childbirth-related hospitalizations, participation in other government programs
appeared to be significant. This pertained even to programs for which undocumented
immigrants are technically ineligible.
As the authors have asserted, it does not appear be the quest for healthcare, but for
employment that draws undocumented workers. However, for undocumented
immigrants living on the edge of poverty, participation in government programs may
be essential to living in the U.S. Government services provide assistance to many
undocumented workers and their children, perhaps enabling them to achieve their
employment objectives. It is not known if the withdrawal of these programs would
reduce illegal immigration.
The lack of will to enforce existing immigration laws is indicative of the nations
ambivalence about undocumented workers. They serve as a ready source of
inexpensive, cooperative, non-unionized labor. Further, there is evidence that
immigration produces a net economic benefit for the U.S. as a whole, if immigrants
remain in the country for at least 15 years (Idea House, 2001; Smith & Edmonston,
1997). Initially, immigrants produce a deficit due to educational and perinatal
healthcare costs for school aged children, but after 15 to 20 years they may produce
79