A MODEL OF FACTORS cammmme To HOSPITAL
EMERGENCY DEPARTMENT USE. FOR NON-URGENT
HEALTH PROBLEMS _

Schazarsy Project "
fer the Degree ~cf M, S. N.
MICHEGAN STATE UNIVERSITY
ANNEE. DIRKSE “
1997. ‘

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

rEBU 1 (UN 1

.
U l! “’ l

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DaleDuepGS—p. 15

A MODEL OF FACTORS CONTRIBUTING To
HOSPITAL EMERGENCY DEPARTMENT USE
FOR NON-URGENT HEALTH PROBLEMS
By

Anne E. Dirkse

A SCHOLARLY PROJECT
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE IN NURSING

College of Nursing

1997

ABSTRACT
A MODEL OF FACTORS CONTRIBUTING TO
HOSPITAL EMERGENCY DEPARTMENT USE
FOR NON-URGENT HEALTH PROBLEMS
By

Anne E. Dirkse

Many factors contribute to the use of the Hospital Emergency Department
(HED) for health care problems which could better be addressed in a primary care
setting. Existing literature does not reveal a single model which identifies and links
these variables together. In this project, the ”Web of Causation” is the conceptual
model used to analyze the use of the HED for non-urgent health problems. This
model was selected because of its ability to visually depict the relationships between
the contributing factors which influence the use of the HED for non-urgent health
problems. The result of its application is the Dirkse Model. Implications for the
Advanced Practice Nurse (APN) are presented. Discussion focuses on areas of

practice, education, and research for the APN.

Copyright by
ANNE E. DIRKSE
1997

ACKNOWLEDGMENTS

I could have not chosen a better Chairperson than Dr. Joan Wood. Her
unceasing advice, support, and encouragement were instrumental in my successful
completion of this scholarly project. With her gift of direction and her assistance in
maintaining a focused schedule, I was able to complete the project within the desired
time frame. I also wish to extend my heartfelt thanks and appreciation to Dr. Linda
Beth Tiedje and Brigid Ann Warren for their willingness to provide assistance in their
areas of expertise.

With the committee’s approval to undertake this particular project, I was able
to fulfill my desire to focus on the issue of access to health care services. The
assistance of these three fine scholars and nurses who comprised my committee made
it all possible.

I would also like to express gratitude to my husband, David, and children,
Andrew, Emilie, and Elizabeth for their patience and support during the months in

which this project took precedence over home and family life.

iv

TABLE OF CONTENTS

LIST OF FIGURES ...................................... vi

CHAPTER 1

INTRODUCTION ........................................ 1
Statement of the Problem ............................... 2
Purpose of the Project ................................. 3

CHAPTER 2

CONCEPTUAL FRAMEWORK ............................... 5
Definitions of the Concepts .............................. 5
Review of Existing Models ............................. 10
The Proposed Model ................................. 17

CHAPTER 3

LITERATURE REVIEW ................................... 21
Variables in the DM at the Upper Level ..................... 22
Variables in the DM at the Middle Level ..................... 35
Variables in the DM at the Lower Level ..................... 44

CHAPTER 4

DISCUSSION ......................................... 46
Access to Primary Care ............................... 46
Developing Interventions .............................. 49
Established Interventions ............................... 50

CHAPTER 5

NURSING IMPLICATIONS ................................ 54
The APN as Practitioner ............................... 55
The APN as Educator ................................ 59
The APN as Researcher ............................... 61

CHAPTER 6

CONCLUSION ........................................ 63

LIST OF REFERENCES ................................... 65

LIST OF FIGURES

Figure 1 - The adapted path diagram of variables in

Andersen and Aday’s (1978) general model of health services ............ 12
Figure 2 - The health services system (Starfield, 1992) ............... 14
Figure 3 — Model of access to personal health care services (IOM, 1993) ..... 15
Figure 4 — The "Web of Causation" for adolescent pregnancy (Clark, 1996) . . . . 18

Figure 5 — The Dirkse Model for the Utilization of the HED
for Non-Urgent Health Problems .............................. 20

vi

CHAPTER 1

INTRODUCTION

Annual visits to a hospital emergency department in the past ten years have
increased by 22% (Young, Wagner, Kellermann, Ellis, & Bouley, 1996). While the
hospital emergency departments exist not only to treat clients with life-threatening
illnesses and injuries, they also serve clients with less serious conditions. Various
authors (Andren & Rosenqvist, 1987; Mayefsky, El-Shinaway, & Kelleher, 1991;
Hansagi, Carlsson, & Brismar, 1992; Young et al., 1996) report that 40% to 60% of
all Hospital Emergency Department (HED) clients who receive care could be treated
in a primary care setting. Concern about the use of the HED as a source of care for
non-urgent health problems is well-warranted. This concern is an indicator of the
current health care system problem of access to appropriate primary care services.
The HED has become a "dumping site" for many clients who do not have access to a
regular source of primary care. Since HED visits generate higher charges than
comparable visits to physicians in clinics and offices, the inappropriate use of the
HED has been cited as one of the contributing factors in the increasing cost of health
care. Furthermore, care in the HED is also typically fragmented and lacks continuity

(Institute of Medicine [IOM], 1993).

2

The reasons contributing to the inappropriate utilization of the HED are
multiple in nature and few sources agree on the direct causes. Many studies
retrospectively determine which visits are ”appropriate” by reviewing charts and
correlating the frequency of "inappropriate" use with client socioeconomic class and
the inability to access alternative sources of primary care (Haddy, Schmaler, &
Epting, 1987; Shesser, Kirsch, Smith, & Hirsch, 1991). In other studies clients are
interviewed at the time of presentation to the HED to determine the reasons for
utilization (Baker, Stevens, & Brooks, 1994; Young et al., 1996). Many of these
clients use the HED for its convenience, its twenty-four hour availability, and as a
regular source of care. These and other factors contribute to create a complexity of
factors which foster the inappropriate use of the HED for non-urgent health problems.
By identifying the reasons for using the HED for non-urgent health problems,
interventions can be developed which could improve quality of care and health care
outcomes. Examples are improved equity of access to primary care services, care
provided by appropriate sources, and care which is more cost~effective and of higher
quality.

Smment 9f the Problem

The problem of the utilization of the HED by clients who have non—urgent
health problems offers a complexity which has not been easily addressed by health
care planners and experts. A study of the reasons contributing to the misuse of the
HED can assist providers and policymakers in understanding the issues of access to

health care as a whole.

3

One of the reasons a person uses the HED for non-urgent health problems is
lack of access to appropriate primary care services. Thus, access to appropriate
health care is a related issue of interest. The Institute of Medicine and the American
Academy of Nursing have created committees to study the issues which contribute to
problems with the United States health care system. The Commission of Health Care
Policy of the American Academy of Nursing (1993) has summarized the issue of
access to health care as financial problems, delivery system problems, and health care
personnel problems (American Academy of Nursing [AAN], 1993). Both the IOM
and the World Health Organization, in conjunction with the Year 2000 Health
Objectives for the Nation (IOM, 1993), have developed indicators to monitor access
to health care. In addition, the AAN and the IOM committees also identified barriers
to health care. The efforts of these health organizations are helpful in understanding
some of the factors which contribute to the inappropriate use of the HED, for both
are concerned with problems of access to health care in the United States.

Ease of the Project

The purpose of this project is to develop a conceptual model of factors
identified in the literature which contribute to the use of the HED for non-urgent
health problems. Existing literature supports the presence of a problem and of
multiple factors contributing to the use of the HED for non-urgent health problems,
but there is no model which identifies the major contributing variables nor one that
attempts to link the variables together.

This project describes several general models in the literature which can be

used to examine the health services system and which may be appropriate for the

4

current issue of interest. Based upon this review, the "Web of Causation” model was
deemed the most appropriate conceptual model for the multiple factors contributing to
the use of the HED for non-urgent health problems. Thus, this model was utilized to
guide the development of the proposed model, the Dirkse Model for the Utilization of
the HED for Non-Urgent Health Problems, which reflects the contributing factors and
the relationships among the factors. It is proposed that the Dirkse Model can be used
as a basis for interventions which may prevent the use of the HED for non—urgent

health problems and may assist providers and policymakers.

CHAPTER 2

CONCEPTUAL FRAMEWORK

Definitions of the Congpts
Several concepts must be defined before a clear picture can be painted of the

intricate relationship of the factors in the problem. The following phrases are defined
for the purpose of this project: Hospital Emergency Department (HED), the client,
primary care provider, non-urgent health problems, and primary care.

Hospital Emergency Department (HED), The HED is defined as a public or
private hospital outpatient emergency room in which the primary use is for the
treatment of life-threatening or acute injuries or illnesses. The HED does not include
a medi-center for treatment of minor health problems. The providers of care in the
HED are typically specialized emergency medical technicians, physicians, and
registered nurses.

93% The term is synonymous with patient; it is an individual who, as an
outpatient, seeks care for a particular health concern.

Primgy care provider. A primary care provider is a family physician, a
pediatrician, an internist, a gynecologist, or an Advanced Practice Nurse (APN). One
study (Aiken, Lewis, Craig, Mendenhall, Blendon, & Rogers, 1979) revealed that one

of every five Americans receives continuing general health care from a specialist

6

physician. For the purposes of this study, a primary care provider is a physician or
an APN who is specialized in family practice, pediatrics, women’s health, or adult
health. The primary care provider’s goal is the attainment of primary health care,
defined later in this section.

Nen-urgent health problemg, There is a wide range of variance when defining
"non-urgent health problems. " The variance lies in the perception of severity of
illness between the consumer of health care and the provider of health care. A non-
urgent health problem is determined "non-urgent" if it meets the criteria reflected in
the guidelines of the American College of Emergency Physicians (Frey, Schmidt,
Derksen, & Skipper, 1994), an accepted standard of measurement for determination
of urgency of need. The non-urgent health problem is one that could be addressed in
a primary care setting. Examples include upper respiratory infections, skin
infections, rashes, or non-traumatic back pain. HEDs typically classify non-urgent
health problems as presenting complaints which have been present for over seventy-
two hours without any change in condition, and conditions in which delay of
treatment would result in no ill effects (Frey et al. , 1994). A complaint of chest pain
or a traumatic injury such as a compound fracture which needs immediate attention is
classified as urgent and thus appropriate to treat in the HED.

Primary gee, Starfield (1992) differentiates between "primary care" and
"primary health care” in the following: "primary care has as its goal conventional
primary medical care striving to achieve the goal of primary health care” (p. 6).
Collado (1992) and Starfield (1992) quote the Alma Ata definition of ”primary health

care" as:

7
Essential health care based on practical, scientifically sound, and socially

acceptable methods and technology made universally accessible to individuals

and families in the community through their full participation, and at a cost

that the community and country can afford to maintain at every stage of their
development in the spirit of self-reliance and self-determination. It forms an
integral part, both of the country’s health system, of which it is the central
function and main focus, and of the overall social and economic development
of the community. It is the first level of contact of individuals, the family and
community with the national health system bringing health care as close as

possible to where people live and work, and constitutes the first element of a

continuing health care process. (Collado, 1992, p. 407)

In this project the concept "primary care" is used as an approach in which the
outcome is "primary health care." ”Primary care” refers to the process of care given
by the primary care provider such as the APN. The APN’s primary focus combines
promotive and curative aspects of health care within the community (Collado, 1992),
and therefore is a "natural ally and promoter of primary health care goals” (Collado,
1992, p. 412).

To further define "primary care" Starfield’s (1992) conceptualization of
"primary care" makes a significant contribution. She describes primary care as the
delivery of first-contact medicine and the belief in the importance of longitudinal
responsibility for the patient which exists regardless of the presence or absence of
disease. According to Starfield (1992), primary care also includes the "integration of

physical, psychological, and social aspects of health to the limits of the health

8
personnel’s capabilities" (Starfield, 1992, p. 9). She continues with her primary care

definition by distinguishing it from secondary care (consultative) and tertiary care
(referral) by writing "primary care deals with more common and less well-defmed
problems, generally in community settings such as offices, health centers, schools, or
homes" (Starfield, 1992, p. 4). This definition includes four dimensions of primary
care, namely, first contact of care, longitudinal care, comprehensive care, and
coordination of care.

The first dimension of primary care is W. Inherent in the
organization of health services is the presence of a point of entry each time care is
needed. This can be called the first contact of care. The health care provider in such
a first contact is often referred to as a "gatekeeper." Intrinsic in the concept of a
gatekeeper is the idea that there should be one health care provider who serves as the
first contact of care for each health problem. There is also the belief that this
provider should be a primary care physician, a pediatrician, or a mid-level provider
such as an APN. The first contact of care also suggests that this contact is readily
available to each client when needed. Availability of a provider to offer health care
services when needed or to refer services to a proper alternative source of care
encompasses some of the concepts of primary care accessibility.

The second dimension of primary care is that of W. The
attainment of primary health care, as the goal of primary care, suggests that one
place, one individual, or a team of individuals serves as the source of care over a
period of time (Starfield, 1992). Individuals should recognize the source of care as

"theirs” and both the client and the provider should recognize the presence of a

9

relationship. This relationship encompasses the essence of longitudinal care, that over
time the client comes to know the practitioner and vice versa and that in this
relationship there are benefits. Many studies have shown that there clearly are
benefits in health care when clients have a regular source of care (Starfield, 1992).

The third dimension of primary care is that of eemprehensive eage and implies
a broad range of knowledge (Starfield, 1992). Although the purpose of primary care
is to provide continuous and comprehensive care which includes the basic services,
the provider may also be involved in secondary care or tertiary care by way of
referral (Millis, 1977). Comprehensive care must also involve problem recognition,
diagnosis, management, and reassessment (Starfield, 1992). Properly trained health
care providers must be able to utilize these aspects of health care, and if not able to
provide the treatment, must recognize the need for referral. Authorities (Millis,
1977; Starfield, 1992) agree that if done properly, comprehensive care promotes
effective and quality health care.

The fourth dimension is that of mrdinatien pf 35. Coordination of care is
essential for the attainment of primary health care; it is the "glue” which enables the
other three components to fulfill their potential. Coordination of care can be a
challenge for the provider. The client may frequently make visits to other health care
delivery sites or fail to follow through with needed referrals. Coordination of care
suggests that the provider is willing or recognizes the need for outside services and
that the client takes responsibility in the follow-through care which the provider

advises or encourages. The phrase "care management” has been frequently used to

10

describe this attribute in recent years, and may eventually become synonymous with
primary care provider coordination of care (Starfield, 1992).

These four dimensions of primary care—first contact of care, longitudinal
care, comprehensive care, and coordination of care— are needed before one can
establish that primary care has been realized by a client. While the four dimensions
are present in primary care practices, they are in reality significantly diminished in
the HED (AAN, 1993; Baker et al., 1994).

In summary, non-urgent health problems should be cared for in the primary
care setting by primary care providers. The goal of this project is to better
understand the factors which contribute to client use of the HED for non-urgent health
problems. With the development of a conceptual model which depicts the
contributing factors and their relationships, interventions which decrease client use of
the HED for non-urgent health problems can be developed by primary care providers,
including the APN.

Review of Existing Models

The literature review revealed one significant contributor and four existing
models of health care utilization which are relevant to the problem of non-urgent use
of the HED: the Andersen and Aday Model of health services (Andersen & Aday,
1978), the Starfield Model of the health services system (Starfield, 1992), the Institute
of Medicine Model of access to personal health care services (IOM, 1993), and the
"Web of Causation" model. However, the latter model was deemed most appropriate
as the model of choice, and therefore formed the general approach for development of

the proposed model for this project.

11
The Amerigp Agdemy of Nursing wprking mm The first significant

contributor to the proposed model is not a visual model but the American Academy of
Nursing’s (AAN) working paper, "Health Care Access Problems and Policy
Recommendations" (1993). The AAN (1993) focuses on four access problems:
vulnerable populations, financial access, health care delivery, and health care
personnel. The AAN (1993) offers recommendations for each specific problem area
in which APNs can improve equity in the access to primary care services. Some
examples of the recommendations include: expand the supply of primary care
providers in rural areas, identify and eliminate barriers to practice for APNs, increase
reimbursement for APNs, and increase scholarships for minorities to encourage
further education in nursing (AAN, 1993).

Ehaviofl Mgflel of health gape utilization, The earliest applicable visual
model was developed by Andersen and Aday (1978); they have done extensive work
related to access to health care services (see Figure 1). They propose a behavioral
model of health care utilization to assess equity of access to medical care. Variables
in this model are divided into predisposing variables, enabling variables, and illness
variables. The model provided in Figure 1 is a diagram of intercorrelated variables
which affect and predict health care utilization. This model has since been used by
Parboosingh and Larsen (1987) in assessing factors which influence the frequency and
appropriateness of utilization of the HED by the elderly.

The predisposing variables are described as those which are not easily

changed, or "immutable. " They include demographic and social structural variables

12

 

  
    
  

 
   
  

 

 

 

 

 

 

 

PREDISPOSING ENABLING NEED UTILIZATION
Age (P)
A
Family Income (E)
Number of M.D.'s (E)
r 1,000 population
Symptoms
V _ Number of (U)
Doctor “SH (5) a Doctor Visits
Insurance \x
Perceived (N)
Health
i
Particular Doctor (E)
Seen
v
Education of Head (P)

Figure 1. The adapted path diagram of variables in Andersen and Aday’s (1978)
general model of health services.

13

such as age, sex, living arrangements, marital status, education, and attitudes and
beliefs about illness and medical care (Andersen & Aday, 1978).

The enabling variables include situational or individual characteristics that
facilitate or impede utilization such as financial resources, availability and
accessibility of services, and established patterns of utilization (Andersen & Aday,
1978). They are viewed as ”mutable" and can be changed with interventions.

The third set of variables, "need of health care services,” are determined by an
individual’s reported symptoms and his/ her perceived health status. Andersen and
Aday (1978) emphasize that these variables are often mediated by the enabling factors
such as client income, insurance status, or provider per population ratio. They also
attempt to determine whether access to medical care is equitable or inequitable by
using predisposing, enabling, and illness variables. Although the behavioral model
proposed by Andersen and Aday (1978) is not recent in development, there are
important variables in the model which are related to this project’s issue. The
behavioral model, however, does not recognize mid-level providers as deliverers of
care in the health care system.

The Starfreld Mpdel, A second conceptual model is work proposed by
Starfield (1992); this visual model is a modification of Donabedian’s model of
Structure-Process-Outcome (Figure 2). Starfield developed the model to reflect the
structure of the health services system. This model illustrates the interrelationships of
the variables defined in Donabedian’s model. It attempts to measure the attainment of
primary health care through the four dimensions of primary care, that is, first contact

of care, longitudinal care, comprehensive care, and coordinated care. Starfield’s

14

 

 

 

 

t

Social
and
physical

CHVIIOIIIDCD

 

 

 

 

4
‘—

 

 

Personnel

t

rpmen

tron

aniza

Management and amenities

Range of services
Con

Org

Facilities and equ

trnuity

Accessi ility

inancing
ligible

Population 6

F

 

STRUCTURE

isron

of care

I
Prov

PERSONS

 

. . . LA”... .T Eve.

. . ... .huv . . ....u.... .... . .nw
.. .. . ....e. . . .... A...» .

WWI“? .rv. .35 a! .u .9.-

.. 3% Aw. x. Hw. . .

fisfilfiflxffi e
(w.

$1.35...
. ....

 

PROCESS "

I up! n
..n.. ..u n .. A ..u ..u ... ..n.
III I III In ky In ..I- - III-nulvln A. I l
..

.. v. ...... a...

.. .. . ... . .. .
SHARE»?
. . u. ..

my"... ...U

. . . . . . .Nva...~.w.

. .. .....n: . ....n... . .. ... ...Luvflt... AP.

. .. . . . «...... . . . .".. {<5 vhwrv. .u.
. . .nu ..

AP. .

ifii B3...

 

#-
..A. . twee...
..AEEw Amity... x ..

SEER...

An.

N." Hawk?

OUTCOME

 

2. The health services system (Starfield, 1992).

Fi ur

15

book, "Primary Care: Concept, Evaluation, and Policy," focuses on the discussion of
primary care services. As seen in Figure 2, multiple structural characteristics have a
two-way interaction with practitioners and clients. The outcome is reflected in
various aspects of health status. The Starfield Model (1992) is a general model of the
health services system, and not problem-specific.

Institute uf Medieine. The third model was proposed by the Institute of
Medicine (IOM) in 1993 as a project to develop a set of indicators to monitor access
to health care services useful to policymakers (Figure 3). This model incorporates

barriers to health care services (structural, financial, and personal) which affect the

 

 

 

 

 

 

 

 

 

 

 

 

 

BARRIERS
Structural
- Availability MEDIATORS
- How organized OUTCOMES
0 Transportation USE OF Appropriateness
. . SERVICES Health Status
Frnancral Efficacy Of 0 Mortality
- Insurance coverage Visits treatment . Morbidity
° Reimbursement levels -
. _’ _, . _p 0 Wellbern
- Public support Procedures Quality of . Funcfiong
providers
Personal
. . ui of Services
° Acceptabrlrty Patient Eq ty
° Cultural adherence
- Language
J- Attitudes

 

 

 

Figure 3. Model of access to personal health care services (IOM, 1993).

16

utilization of services. These barriers are affected by the mediators of
appropriateness, efficacy of treatment, quality of providers, and client adherence.
The combination of these variables results in outcomes of health status such as
morbidity, mortality, and equity of services. Several of the barriers defined in the
IOM model are problem specific in this project’s proposed model.

The "Web pf Causatiun" mgglel, The fourth conceptual model, the "Web of
Causation” or the ”web,” is rooted in multiple causation to explain the existence of
health and illness states, and provides guiding principles for epidemiological practice
(Spradley, 1990). The ”web" allows one to visualize all the variables affecting a
problem from early in the course of causation to the existing problem. With
visualization of these factors, one can begin to develop interventions that can prevent
development of the problem at the appropriate level of causation.

The ”web” was first coined in the literature in 1960 by MacMahon (Krieger,
1994) in an epidemiological textbook. Prior to 1960 the belief that a single agent was
responsible for a disease was beginning to lose credibility and was being replaced
with more complex models of host, agent, and environment as causative factors of
disease (Krieger, 1994). This belief originated with promoters of social medicine
who argued in favor of examining the social determinants of disease (Krieger, 1994).

The metaphor of the "web” can be viewed as "an elegantly linked network of
delicate strands, the multiple intersections representing specific risk factors or
outcomes, and the strands symbolizing diverse causal pathways" (Krieger, 1994, p.

890). It encourages epidemiologists to look for multiple causes and multiple effects,

17

to consider interaction, and to identify the many routes by which disease and social
issues can be analyzed and thus possibly prevented.

In this context the "web" is now being used more in nursing theory and
specifically as a conceptual model in the study of disease causation and presenting
social issues which contribute to health problems in community health nursing.
Existing literature utilizes the "web” in the study of the multiple causations leading to
myocardial infarction (in Spradley, 1990 and adapted from Friedman, 1988); in the
study of pressure ulcer prevalence, incidence and associated risk factors in the
community (Oot-Giromini, 1993); and in the study of contributing factors in pressure
ulcers in spinal cord injuries (Lehman, 1995). Figure 4 is an example of the ”web"
model as applied to the issue of adolescent pregnancy (Clark, 1996).

The Premfl Mmel

Although each of the previously described potential models is applicable for
this issue of interest, the "web” model is selected as the basis for understanding HED
utilization for non-urgent health problems. The ”web" is ideal for this issue; it
illustrates the multiple, interrelated causality of factors which can be adapted to a
particular problem. The variables in the proposed model, hereafter referred to as the
Dirkse Model (DM), reflect the elements of the "web,” and are based upon a review
of literature concerning the non-urgent use of the HED as well as barriers which
affect access to primary care services.

Since multiple variables contribute to and affect the use of the HED by people
who have non-urgent health care needs, this issue can be dissected by using the ”web"

model to understand personal, socioeconomic, social, and behavioral aspects of the

.\.<..

                             

          

6% 1.28 BEBE “88068 8a .8330 Lo 83.. 2:. in»; a _

                  

   
      

     

               

 
 

                                                   

                                                    

              

 

      

 

. A.
a. . .4 s
\I I \
.I.v \ ...u.

. ... . .. . \ WA \
....Swuwéwmskfiufiéwwfiéar

kuxéxwfiw... WEEKS? ...
.us . .5. ..AG. .v . . .s 1.5.. ruv .
m @me.n..m.mww>e 31...... e»... A... “we

-
.
non-

...? ...... . Heaven. .. .
Raw? £7?pr as...)
#3... 313.5. .v ...9. .2 5%.

 

                   
                         

  

                 

          

   
   
                          

                       

                                    

          

   

 

 

                   

    

                 

     

       

                      

     

                  

   

    
       

  

                      

  
         

   

             

                  

        

                 
  

              

                   

    

        

                 

          

            

 

 

"...... {4 ...éfixfin . A 93.}... ... ...Av. Lax . . . ...x .
s ..2 .12.... .i .. ...: A. . 3.. . .39.. .. P... «....s . . ....xw...\x .4". .9
a... .s. .. ...... . ... . .w..G.Mnn¢\..$..M ago“ .....A? «92535.». «NE... . . “...”...e. 51......" BEER”. ..mwaufimve ..
.... ......me $5.... BEES. .....x...» EAR... AVE... ......» Peg... .3... ... text}... Bet...
.... .u. a. 9....“ ......Rx .. a? Refine?“ Eve»? See B. ...... .. ans. Beggars... ....
. ...... «m Effie... as?" . ...“ «a. BREE. "...... EA... .....Rw ..m RENEEEAWNA
. .....z . . (3 A ....APAH A... syxmkaaw ......u... N.» 94.»? ...»... ...x. flofiw ..v.
t... .... Aw ..S... ...R Sate... .... A»... .. .. ......GE. . . u... 33.. ..e. P...
I” .n . .su ”\uvk}. x ... ifs...» Amara. r. $.33? 9%.? ....\.u“.r.wr ....wr .. ..u.v AS. .Aqu.
...... RAMA. A... A... .. .... .. «...... Bees... ......R...B........ . E... T...»
$.93 . 3...) > B}. v A... ... +5.1... ... . ......v. . .... ......
... .32 $3... ¢¢.Bv.e.e .w. ....n. BEEPAEE Se... AMP...» Eve ..
.. . .....Ex. .. $m...fi.~"..#.e.s . ...”..r ....3 getaway." . A"... . .....hfi..§.....«...¢

  

     
     
      

   

 

 

 

 

18

'D-arnrlnt' .

5.: vans-u.-
. .

1‘

 

54:22.

 

 

 

......B.
Vin}. .-
53.3.".

 
 

 

 

 
   

0-.
.\
-.- .-.

 
 

      

   

:-:.
.
Ava

P
‘3'

  
   

  

.-.
ha I

 

r-z- -
>~
t:
a

 

96— o. .

 

m. macaw—Ema >23 .0»

\' cl... 3

 

 

 

  

... 388» mm.” . . 0:858

9.ng 2 emcee
.v. A“

mic . .

 

 

 

 

    

..
9...? 3..

E

 

   

 

x9.
E8883

2 80:53

  

19

stated problem. The features of the "web" are problem—oriented, and therefore most
appropriate for the use of the HED for non-urgent health problems.

The variables in the DM are visualized beginning at the top of the model with
factors that affect the use of the HED before the problem exists. These variables
have an impact on later factors which occur at a time when a client considers seeking
care for a particular health problem. The factors which influence a client’s decision
to seek care are placed in the center of the DM. At the lower levels are the variables
which occur at the time a client uses the HED. At the lowest level of the DM is the
stated problem, that of utilization of the HED for non-urgent health problems (see

Figure 5).

20

Reuse.” 53am gaseoz .8 emu 2: .6 Saunas 2: .8 382 Big 2E. Adana

 

 

 

v .. ..
All-K . .. .-
.ucgw. when.

 

 

 

 

 

83.5:— .

homes—8am All' 50:0 . 895 9 83 :08
..o 8:32.— Boomg 3.. coo: . AllV
”Evans. 8832.... .808 .o Sam 9 83:02

4 4 4 P . s . 4
838a. .3 2.29355 02%

3033.. w 3:2: «3:5.» 5-525 . ceasing . 0552.
.0 SE: 9.6 . .
gongmwfimfim—o afiSbw—Wfi . oomEHhfiwo: . 8:30— . 80:0 .2, 9:25 «HEP—owe 803.8” 53 82.6.5 .
. 39.63 825m 9 308.. 8308? 50:0 . 30:0 .

3228 . .
Ho 55939.8(. .8 554.635 .3 :oaavoEEoou< ho b53339}. we banana}. ego—365. 23:0 noigom 82::

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5‘-

..

..“w.

. .

.9 . . . .

.....m .54....
+4..

'54 ‘1.”
4.5 523+,-
finJ-‘IZ'
E5421.
. "in."
+‘- -:-*'

GE»

 

1.
“c

      

View.)
{53"

S

       

          

        

 

..
.4.
.3 . . u. ..v
..5 I.- Aa . h. -
. .. ewe... Ewes
WV... N4"... .
I \<.I-.Vnu

  

    

..n.............n.xn.....u.....w...

a?
I
:9:

.‘E'

N
4'

   

. .1
...

...

    
 

      

. 6.53%...
. . ”44144....me ......

      

 
 

 

 

        

    

            

          
     

 
        

 
 

   

           

            

    

             
  

 

        

      

                  

        

......n..R........... .u....n...."..¢...n....+. ......4. .... . .. .. .
.53.."va. ..H 193.... 13.3.1333.“ . ...! .x AWN/MA<.¢H .u....uv.w.v~ .. ... .. I ”IN. $5.5...
BEBE... ...... Ea... BREE... BEE... new... Riva.”
....x. ... . . .. x.. . .
. 4. . ATP... 4.“ +A+.........~.wwv......... $3....wa ....
E.#.........%.........4.. ...»... ......4. ...... . .
...... . 3.....4. ... .....4. .
2....“ flaw...»

         

   

  

    

       

  

.3}...
a.

  

v a.

.... ... ..
....w. 4.. ....S..............¢ ...... a

.94er >3...+...uww$ .. .

. on..v/...\.I/.\)(\\? s AWL. fir.
. . 9%. ..
($0.... emu...

w...

        
    

        

      

         

    

    

   
                     

   

             
      

 

 

     

   
     

 

    

      

  

         

    

   

 

   

Haw“... ......»A... . ... .+. .v......\. a.

..me AMER 4.4.... ....fiwwwas... ......»th .. .... ..

...". . .. .31... . ...

...”... . . . .... . x. ... ....Be . ......2 . ..e ...

.. . .. 4...”... ....w...“.. ... .14. .44 .4. . .u

:9. ..r a} ..v. .1 (Alan... .ffiovfio. .YVJVN . . .
...... «mm. .9344”... .w.ew...u..e......................w. weeewsfi. .. .. ......amfi
$¢Mw¢flfl~wv$3flxé. ..waewkwfl...x .32... . 9.4%? ......” Jfifiauvawufifl
...... SEE}... .. .. E .. . . : .4...
......wufiwfi .A...R...........Mvfi. .«flwfi. . . ..
LEASE“? . . 4......

    

 

 

CHAPTER 3

LITERATURE REVIEW

The literature review enabled the investigator to identify the variables in the
DM. Figure 5 reflects these variables. The following description of factors
corresponds to the placement of the factors in the DM sequentially, top left to bottom
right. The initial factors, client demographics, client socioeconomic status, client
health status, and community attitudes are reflected at the top of the DM.

Factors at the next level include attitudes and beliefs, no insurance or
underinsured, presence of public aid, low formal education, the presence of health
problems, the presence of psycho-social problems, and the lack of a regular provider.
Continuing through the DM is illness perception by both the client and provider,
client satisfaction with services, client knowledge of access to available services, and
the five dimensions of access.

The last level in the DM prior to accessing care in the HED is the motivation
to seek care, the ease of access to services, perceptions regarding the need for
services, and the presence of a gatekeeper. These contribute to the decision of the
client to access care in the HED and the acceptance of the client in the HED resulting

in the use of the HED for non-urgent health problems.

21

22

Figure 5 also reflects horizontal and vertical relationships between factors.
The presentations of the factors in the literature review parallel their introduction in
the DM. Some of the literature cited is concerned with access to health care at an
appropriate source and is directly related to the client’s inability to access care at an
appropriate source.
Vmables in the DM at the Umr Eve]

gzlient demographics, For the purpose of this project, demographics includes
culture, race, gender, geographic location, and age. Many sources (Purdie,
Honigman, & Rosen, 1981; Thomas & Penchansky, 1984; Haddy, Schmaler, &
Epting, 1987; Blendon, Aiken, Freeman, & Corey, 1989; Frey, Schmidt, Derksen, &
Skipper, 1994; Berk, Schur, & Cantor, 1995; Young et al., 1996) agree there is a
relationship between client demographics and access to health care, but that the
relationship is indirect rather than direct. Thomas and Penchansky (1984), in their
classic study, view client demographics as having an indirect effect on a client’s
health motivations and perceptions. For example, the demographic of culture may
affect an individual’s evaluation of the importance of a symptom or knowledge of a
disease. They conclude from their study that sociodemographic characteristics relate
to behavior through clients’ beliefs and perceptions. Client perceptions are built into
the DM at the middle level as influenced by demographics, socioeconomic status, and
health status.

Although culture is included in the discussion of the demographics as affecting
the utilization of the HED, it tends to be interrelated with race, beliefs, or the identity

of a group of individuals rather than the stereotypic ”demographics” of an individual.

23

Merriam and Webster (1995) define culture as "the customary beliefs, social forms,
and material traits of a racial, religious, or social group” (p. 282). This definition is
much broader than race alone and is much more difficult to measure as an influence
towards access to health care. The culture of an individual affects one’s perceptions
related to health beliefs thus influencing the decision to use the HED as a source of
health care. As an example of culture, a Hispanic child with a cough and fever may
first be treated at home with an alcohol sponge bath as a folk remedy. If no
improvement is seen, the parents may seek care in the HED without trying other
means to treat a cough and fever which are conventional to Western medicine.

Race as a demographic variable in the use of the HED for non-urgent health
care, as well as access to health care, has been the topic of many studies. Frey et al.
(1994) reported in their study that 71 (41%) of 174 Hispanics compared to 70 (25%)
of 280 Caucasians used the HED inappropriately as determined by the criteria set by
the American College of Emergency Physicians (ACEP) (Buesching et al., 1985).
Race would also affect the criteria established by the ACEP, which consists
predominantly of Caucasians. The race characteristics of client use of the HED for
non-urgent health problems in the study of 1190 subjects by Baker et al. (1994)
showed that 226 (22 %) were Caucasian, 508 (45%) were Hispanic, and 285 (26%)
were African American. Pane, Farner, and Salness (1991), in their study of 1,000
patients who sought non-urgent care in a California HED, found that over half of the
study sample was Hispanic.

Blendon et al. (1989) found that African Americans lacked access to physician

care more than Caucasians, that is, African Americans reported poorer health status

24

and fewer physician visits than Caucasians. Although race is cited as an example of
inequality of access to health care, it is also noted that a larger percentage of
minorities than Caucasians are living below the poverty level. Nationwide, about
30% of Hispanic families are living below the poverty level compared with 15% of
the total United States population (Pane et al., 1991). This makes it difficult to
separate out race and socioeconomic status as factors in access to primary health care.
Blendon et al. (1985) suggested that race and culture together are directly related to
the acceptability of the providers who treat them and may indirectly affect access to
primary care services.

The demographics of gender and utilization of the HED do not appear
significant. One study (Purdie et al., 1981) found that 13 of 16 chronic users of the
HED were men, and that lack of appropriate services for their particular health
problems was the primary reason for HED use. Several studies reported similar
utilization of the HED by men and women (Baker et al., 1994; Young et al., 1996).
However, Thomas and Penchansky (1984) reported that gender influences various
aspects of accessibility to primary care services. Women reported satisfaction with
affordability of services in terms of payment arrangements, health insurance coverage,
and cost of services as influential in their decision to access a source of health care
(Thomas & Penchansky, 1984). As a result, women tend to utilize the HED for non-
urgent care when the care is perceived as more affordable than alternate sources of
care.

Geographic location, as part of client demographics, presents as a conflicting

predictor of HED utilization for non-urgent health problems. Young et al. (1996)

25
reported that of 6187 clients, 483 (8%) used the HED for geographical reasons, and

183 (3%) used the HED because of transportation problems. Thomas and Penchansky
( 1984), however, found there are many people who travel from outside an area to
seek care for services because the services are more available in terms of the
affordability and acceptability than a source of care in the immediate vicinity.

Several studies suggested that a higher percentage of people between the ages
of 19 and 43 use the HED more than other people (Shesser et al., 1991; Young et al.,
1996). Grumbach, Keane, and Bindman (1993) reported the mean age of clients with
non-urgent problems as 36 years, while Young et al. (1996) reported a mean age of
30 for the study clients. Purdie et al. (1981) reported the mean age of the chronic
HED client in their study as 48 years. Frey et al. (1994) reported higher rates of
inappropriate visits of children under the age of five.

W Thomas and Penchansky (1984) proposed that the
attitudes and beliefs of both the client and the provider affect the utilization of health
care. They referred to health care access as "the degree of fit between the client and
the health system" (p. 554). This degree of fit can be narrowed to include the match
between a client seeking care in the HED and the ability of the HED to absorb the
client into the system as a function of similar beliefs and acceptance. Although the fit
may be good, it is not always good longitudinally in terms of cost-effectiveness or
optimal outcomes. Thus, attitudes and beliefs of the client precede the level before
which a client decides to seek care, and are reflected as such in the DM.

The attitudes and beliefs of a provider of care in the HED may also affect the

treatment of the individual and influence client outcomes and client satisfaction with

26

care received. These attitudes and beliefs of the provider in the HED may be
influenced by the provider’s educational preparation. As an example, a provider of
care in the HED may treat a client experiencing a mild asthmatic attack with a
nebulizer treatment and then discharge. A primary care provider may treat this same
client with a nebulizer treatment, maintenance and episodic inhalers and provide
asthma management education.

In summary, most sources agree that client demographics are indirectly related
to other variables by affecting access to health care services. Thus, the demographics
of a client are depicted early in the DM and reflect both horizontal and vertical
relationships between the other model variables.

Client goioeconomig status. The key components of socioeconomic status
($138) in this project include income, education, and employment. The SES is
separated from demographics because of its stronger influence on the use of the HED
for non-urgent care. Many studies supported a connection between SES and
difficulties with access to an apprOpriate source of health care. The AAN (1993)
cited individuals living in poverty as having the greatest risk for health care access
problems. Similarly, the IOM (1993) recognized that individuals who are poor, those
who lack transportation, those without adequate health insurance, and those having a
lower level of education are less likely to obtain prOper health care. Pane et al.
(1991) concluded that among HED walk-in clients in their study, low-income
individuals were significantly more likely to use the HED as a routine source of
health care and to delay seeking care than higher income individuals. The delay in

seeking treatment by low income individuals could also contribute to the results of a

27
study by Epstein et al. (1988), in which clients of lower SES had longer hospital stays

and higher charges than clients of higher SES.

All of the individuals in the study of the non-urgent, chronic HED client by
Purdie et al. (1981) were either on public assistance, had no regular source of
employment, or were dependent on a relative for their support. Use of the HED for
care by the clients in this study was more likely related to the fact that the HED
cannot refuse care to anyone who presents for treatment if another appropriate source
of care is unavailable (Buesching et al., 1985). Kasper ( 1987) found that children
from high income families had a 20% greater likelihood of seeing a family physician
for care. Likewise, children from low income families had a much higher occurrence
of using the HED as their usual source of ambulatory care (Kasper, 1987). Berk et
al. (1995) found that of the 3450 persons in their study who were unable to obtain
care, 24.4% reported an income of less than $20,000 compared with 16.6% who
reported an income of $20,000-$50,000 and 7.9% who reported an income above
$50,000. The data from the study done by Pane et al. ( 1991) suggested that,
compared with higher-income and fully insured clients, poor people and the medically
indigent were having ”significantly more difficulty in accessing the health care system
and are routinely using the HED for basic health care needs" (Pane et al., 1991, p.
733). Pane et al. also suggested the overuse of the HED results from primary care
providers’ refusal to provide care. Pane et al. (1991) reported that of the 1000 study
population, 680 reported an income of less than $10,000, and 206 reported an income
of between $10,000 and $20,000. These clients were reportedly walk-in clients, with

income being a statistically significant predictor of routine HED non-urgent use.

28
No insurance/underinsurfl, The lack or shortage of insurance is related to the

ability of one being able to pay for services and is critical in one’s access to health
care services. There are numerous studies which explore the relationship between an
individual’s source of payment for HED services for non—urgent health care and
access to health care services. The AAN (1993) and the IOM (1993) are both
consistent in their studies that individuals who are uninsured and underinsured are at
risk for unequal access to health care services and are more likely to rely on the HED
for their source of health care (AAN, 1993). It is estimated that in 1989, 37 million
people in the United States were uninsured (AAN, 1993), with 75% either employed
or dependents of employees whose employer did not offer health benefits. Young et
al. (1996) found that of the 6187 people using the HED, 24% were uninsured while
76% either had public or private insurance. Pane et al. (1991) found that 40% of the
people using the HED for non-urgent care lacked insurance and that public aid and no
insurance together totaled 70% of the 940 people. Berk et al. (1995) found that of
the 3450 of those unable to obtain any health care, 33.7% were uninsured compared
to 13.6% who were insured.

W In order to provide basic health care for persons who
could not afford to pay either for health care or health insurance, the Medicaid
program was initiated by federal legislation in 1965 (Starfield, 1992). There are
many studies and statistics on the use of the HED by those who have Medicaid as
their source of insurance. The number of visits by Medicaid recipients to the HED
increased by 34% between 1985 and 1990 (Medicaid Access Study Group, 1994).

Many Medicaid recipients sought treatment in the HED for relatively minor health

29
problems. The Medicaid Access Study Group (1994) found strong evidence that

limited access to ambulatory care outside the HED was the reason the HED was used
for basic care. The study suggested factors influencing access to health care for
Medicaid recipients which included (a) refusal to grant walk-in appointments to those
with Medicaid, (b) no appointment possible within two days, (0) a walk-in
appointment after five o’clock in the afternoon not possible, and (d) request of a co-
payment.

St. Peter, Newacheck, and Halfon (1992) suggested that Medicaid does
improve access to care for poor children, but does not ensure them access to the same
location or to continuity of care as is available to non-Medicaid children. Kasper
(1987) supported the belief that Medicaid increases access for poor children, but that
access for care suggested reliance on the HED as a regular source of care. Buesching
et al. ( 1985) found that Medicaid payors used the HED appropriately at a lower
proportion than other payors and had a significantly higher proportion of
inappropriate use. Young et al. (1996) found that 46% of 6187 ambulatory clients
using the HED were on government support.

Although there is strong evidence that Medicaid assists the poor with basic
health care benefits, the evidence does not demonstrate access to a regular source of
care. One of the attempts by the government to remedy this problem is to assign a
regular health care provider to each Medicaid enrollee. This intervention will be
discussed later in the paper.

Employment status is related to several variables in terms of affordability of

health care because of the presence of insurance. Those with insurance have reported

30

the presence of a usual source of care in the form of a primary care provider.
Buesching et al. (1985) found that a higher percentage of inappropriate HED visits
were made by unemployed individuals (13.1%) than employed (6.8%). Grumbach et
al. (1993) found that 68% of 700 study clients waiting for care in the HED were
unemployed.

W A lower level of education is also associated with an
increase in the use of the HED, and is connected to lower socioeconomic status.
Baker et al. (1994) reported in their study of 1190 subjects that 22.9% of the users of
the HED as their regular source of care had no formal education, and 17.4% reported
an education level of high school or less. Rask, Williams, Parker, and McNagny
(1994), however, reported that a low level of education was not a strong predictor of
a regular source of care but was reflected in the individual’s delay to seek health care.
Thomas and Penchansky ( 1984) suggested that a lower level of education influences
the utilization behavior by fostering dissatisfaction with available services.

A lower level of education can affect health seeking behavior in relation to
client knowledge of disease process, prevention, and treatment. To quote the ACEP’s
introductory statement in the guidelines to appropriate HED use:

We feel that a patient has made an appropriate visit to an emergency

department when: An unforeseen condition of a pathophysiological or

psychological nature develops in which a prudent lay person, possessing an
average knowledge of health and medicine, would judge to require urgent and
unscheduled medical attention most likely available, after consideration of

possible alternatives, in a HED. (Buesching et al. 1985, p. 672)

31
This definition of appropriate use of the HED encompasses both knowledge and

cognition by an individual; the positive correlation between a lower level of education
and non-urgent use of the HED is easy to understand.

Client health stams, Similar to the demographics and their relationship to the
use of the HED for non-urgent health problems is the variable of client health status.
According to Adler, Boyce, Chesney, Folkman, and Syme (1993), socioeconomic
status is a strong and consistent predictor of morbidity and premature mortality,
therefore the DM depicts health status as connected horizontally to client
socioeconomic status. Empirical research has focused on the impact of poverty and
its correlates, such as poor housing and inadequate nutrition, while policy debate has
focused more attention on insurance coverage as a remedy to SES-related inequalities
in health (Adler et al., 1993).

Pregng of health problems, Both the AAN (1993) and the IOM (1993)
reported that populations with chronic illnesses, disabilities, and/or those who are
likely to have worse than average health status, experience barriers to access
appropriate health care services. Baker et al. (1994) reported that of the 589 clients
who reported the use of the HED as their regular source of care, 178 (30%) reported
poor to fair health. Berk et al. (1995) reported that 1031 (29.9%) of 3450 of their
study population who reported inability to obtain some aspect of health care also
reported poor to fair health, whereas 507 (14.7%) of the study population reported
good to excellent health. These studies suggested that people who are in relatively

poor health are more vulnerable to barriers in accessing proper health care.

32

Presence pf psyeho-flial problems, Andren and Rosenqvist (1985), in their
study of heavy users of the HED, found that the repeater group of individuals using

the HED had more psycho-social problems. In their follow-up study Andren and
Rosenqvist (1987) found that the group of clients who continued to be heavy users of
the HED reflected an increase in four variables: (a) the proportion of those
unemployed, (b) the number of unmarried, (c) the number with a high level of
loneliness, and (d) the number who tended to live alone. Andren and Rosenqvist
(1987) concluded that those who had less access to an informal social network used
the HED to a greater extent than those with better access to a social network.

The lack of social support and presence of psycho-social problems as a
function of the use of the HED for non-urgent health problems is suggested in the
study by Purdie et al. (1981). In their study, 15 of the 16 chronic users of the HED
were single, fourteen of the users experienced alcoholism, and nine were suffering
from a chronic psychiatric disorder. Additionally, the results of a study done by
Calnan (1984) suggested that social circumstances may be an important influence on
client demand for outpatient care in the HED.

Qemmunity attitudes, The willingness of a community to take responsibility
for insuring adequate health care for all its members plays a significant role in access
to primary care services. A community is not only the immediate population in close
proximity to the HED, but society in general. The IOM (1993) cites society’s ethical
obligation to ensure access as follows: ”Equitable access to health care requires that
all citizens be able to secure an adequate level of care without excessive burden" (p.

32). Community attitudes towards the health care of its members reflect ethical

33

beliefs and proactive steps to ensure care for all. The action a community takes
towards ensuring access to health services includes services that are affordable,
services that can accommodate a wide range of population needs, and services that are
readily available, acceptable, and accessible to its entire population.

Community attitudes affect several factors in the DM. One of the factors
affected by community attitudes is that of the supply of primary care providers. In
this project investigator’s own community with a population of 60,000, the
community hospital has a full time physician recruiter. This recruiter, under the
direction of the CEO and board of directors, who are members of the community,
makes decisions on the number and type of providers. These are also the people who
affect the recruitment and practice of APNs in the community. This particular
community is known for its conservative attitudes towards midlevel providers such as
the APN and affect the supply of the community’s primary care providers.

Community attitudes toward health care are reflected in an article in the
Wee] (Lozon, 1997) in which it was reported that a large corporation with
5,000 employees established an on-site medical center to provide health care. By the
provision of extensive health care services on-site, the company can offer more
economical, convenient, and accessible care for employees and their families. The
providers include a physician, a physician assistant, and an APN in family practice.
This is a move in a positive direction to eliminate use of the HED for non-urgent
health problems.

Community attitudes also affect the distribution of public aid to individuals by

way of state and local laws and guidelines, for states vary in their coverage of

34

Medicaid for individuals. In 1994, the states of New York and Tennessee each
reported that 13 % of their respective populations received Medicaid, whereas in the
states of Florida and Texas, only 8% of their respective populations received
Medicaid. This study also reported variances in the number of urban private-charity
clinics; Milwaukee reported a total of seventeen, while Boston reported one (Medicaid
Access Study Group, 1994). These statistics reflect differences in community
attitudes towards health care.

Laek pf regular previger, endersppply, pr maldistg’bng'pn, Again, one
influence the community has towards the use of the HED is the number of primary
care providers in a community. Buesching et al. (1985) found in their study that the
total number of inappropriate HED visits had fallen because of a substantial increase
in the proportion of primary care physicians to the population. Hilditch (1981) found
that after the establishment of a community health center in an underserved area and a
fivefold increase in the family physician per population ratio, the use of the HED
decreased. The fact that a person has a usual source of care that provides a point of
entry to the health care system is strongly related to the likelihood of the person
seeing a physician (Kasper, 1987). The study by Haddy et al. (1987) supported the
premise that clients who have a regular personal physician tend to present to the HED
with true emergency conditions more often than clients who do not have a personal
physician. While the American Academy of Nursing (1993) cities the undersupply of
primary care providers affects the access of selected p0pulations to appropriate health
care (AAN, 1993), there is nothing in the literature on the impact an APN has on the

non-urgent use of the HED.

35

Not only is there an undersupply of primary care providers, there is also a
maldistribution which contributes to an undersupply for vulnerable populations or
groups (AAN, 1993). Various studies have shown some relationship between not
having a regular source of care and individuals using the HED as their usual source
of care (Baker, Stevens, and Brook, 1994; Grumbach et al., 1993). Grumbach et al.
(1993) found that of 1190 clients in their study, 309 (28%) had no usual source of
care, while 178 (16%) identified the HED as their usual source of care. Pane, Farner
and Salness (1991) found that 213 of 1000 clients (24.8%) identified the HED as their
regular source of care. Shesser et a1. (1991) in their study sample of 325 stable walk-
in clients found that 72 (22.1%) had no previously identified source of personal health
care. The lack of a regular provider is a critical factor and contributes to the stated
problem; it will be discussed more fully later in the implication section.

To summarize the factors in the upper and middle levels of the DM, four main
important areas identified include: client demographics, client socioeconomic status,
client health status, and community attitudes. From these four variables arise
attitudes and beliefs, no insurance or underinsured, presence of public aid, low formal
education, presence of health problems, presence of psycho-social problems, and lack
of a regular provider.

Vag'ables in the DM at the Middle fivel

The following paragraphs describe the variables in the Dirkse Model which
affect the use of the HED when a client decides there is a need to access the health
care system for a particular problem. The decision is affected by many of the

variables depicted in the middle level of the model. These variables include: illness

36

perception by both the client and provider; client satisfaction with services; client
knowledge of access to services; and the five dimensions of access to services, that is,
availability, accessibility, accommodation, affordability, and acceptability. These
variables are affected both vertically by the previous factors, and horizontally, and are
illustrated as such in the DM.

Illness mrgptien, Illness perception by both the client and provider are
related to culture as a client demographic, and to formal education, a variable of
client socioeconomic status. Illness perception is also expressed in the perception of
urgency of need for care. Client perception of the need for prompt medical attention
is a factor in the decision-making process to seek care. This need is influenced by
culture, level of education, knowledge of appropriate use of services, and the
dimensions of access to care. Several studies suggest discrepancies between client
perception of urgency of need and provider perception of urgency of need (Baker et
al., 1991; Frey et al., 1994). Frey et al. (1994) found that in only 26.6% of 417
visits in their study was there agreement between the provider and the patient as to
appropriateness of the HED visit. This would suggest that in the remainder 73.4% ,
there was a perceptual discrepancy between client and provider.

Illness perception is interrelated to attitudes and beliefs of both client and
provider; the authors (Frey et al. , 1994) also suggest illness perception is impacted by
client education. Approximately 60.3% of the 417 study clients thought that
treatment was required earlier than the HED attending physician; this suggests that
client perception of severity of need determined the utilization of the HED resulting in

an inappropriate use (Frey et al., 1994). Frey et al. (1994) also reported that 65% of

37

417 visits demonstrated agreement between physicians and criteria by the authors that
the visit was appropriate, while the remainder 37% disagreed on appropriateness of

the HED visit.

Client satisfaction with services Client satisfaction with services is also a

 

function of demographics and is another factor in the use of the HED by clients with
non-urgent health problems. Several studies cite that clients prefer to use the HED
because of its convenience. Shesser et al. (1991) found that 77 (23.7%) of the 325
sample group preferred this setting because it was quicker and always available.
Young et al. (1996) found that while 121 (6%) of the 2043 clients without a regular
provider identified not having a family physician as the reason a client chose to come
to the HED, 1922 (46%) cited other reasons, such as the doctor’s office was closed,
they were able to get quick medical attention, and they too sick to wait for an
appointment.

Client satisfaction is used as an important indicator of measurement of the
quality of medical care. In a study by Hansagi, Carlsson, and Brismar (1992) of 758
clients who used the HED, patient satisfaction was found to be high with regards to
treatment (87%) and service/care (92%). The 380 clients who were classified as non-
urgent scored satisfaction slightly lower than the clients classified as urgent; the
investigators (Hansagi et a1. , 1992) found many of the non-urgent study clients had
been unable to be seen by their own primary care provider possibly affecting the level
of satisfaction with the HED.

Aside from being a consequence of utilization, client satisfaction is also

believed to be a major determinant of care provider choice in the future (Hansagi, et

38
al., 1992). Young et al. (1996) noted that non-financial barriers to primary care

services were related to satisfaction, that is, the clinic was not open, an immediate
appointment time was unavailable, or the clinic did not take walk-ins. Young et a1.
(1996) also cited personal preference for the HED as the reason an individual chooses
to use the HED for care of a non-urgent nature.

Client knowlgge of afiss t9 servigs. Client knowledge of access to services
for non—urgent health problems can be influenced by client demographics and
socioeconomic status (Thomas & Penchansky, 1984). Many clients are unaware of
the ramifications of seeking care for health problems in the HED that could better be
cared for in a primary care setting. Client knowledge of the availability of services
may be limited by the lack of education by providers or by guidelines and standards
(Starfield, 1992). Some clients may be unaware of the cost of care from the HED
versus the cost of care in a primary care setting (Young et al., 1996). The higher
cost of care in the HED also includes hidden costs such as duplication of services
including laboratory tests or radiographs (Gadomski, Perkis, Horton, Cross, &
Stanton, 1995). Discussion of client knowledge of access to services will be
expanded in the discussion and implication sections.

Five dimensipns pf mss te health w, The next factors which influence the
use of HED services are the five dimensions of access to health care proposed by
Thomas and Penchansky (1984) and recognized as influential in access to primary
care by Starfield (1992). These five dimensions are incorporated into the DM as
variables resulting from the influence of demographics, socioeconomic status, health

status, and community attitudes. The dimensions’ placement in the model reflect the

39

HED access point; it is here that the five variables affect the individual’s decision to
access care in the HED. The five dimensions, availability, accessibility,
accommodation, affordability, and acceptability, determine how a person seeks and
accesses care.

Availability, the first dimension, is described as the adequacy of the supply of
physicians and other providers, health care facilities, specialized programs and
services (Thomas & Penchansky, 1984). Factors which influence availability include
the ability to pay for the services, the provider to population ratio, and the presence
of a regular source of care. When there is an insufficient number of primary care
providers per population available, clients will access services of care which are
available, such as the services of the HED. Although Thomas and Penchansky (1984)
suggest that the doctor-to—population ratio should represent a clear measure of
availability, this factor was not significant in predicting utilization. A more recent
study by Baker et al. (1994) found that clients who identified the HED as their
regular source of care had 25% fewer physician visits than those with a source of care
other than the HED. The study suggested that the clients’ possession of a regular
source of care is a strong predictor of physician utilization in the primary care setting
or the HED.

Accessibility, the second dimension, is defined by both Thomas and
Penchansky (1984) and Starfield (1992) as the relationship between the location of
supply and the location of clients, taking into account clients’ transportation

resources. The accessibility of the HED in terms of structure alone was a factor for

40
11% of 6167 study clients who used the HED for non-urgent health problems (Young

et al. , 1996).

Accommodation, the third dimension of access, involves the manner in which
the supply resources are organized to accept clients and include appointment systems,
hours of operation, walk-in capabilities, and telephone services. Starfield (1992)
includes in the definition the extent to which the population perceives these aspects of
access to a care source as convenient. A client’s willingness to accommodate to these
factors predicts the utilization rate of the services. In the study by Young et al.
(1996), 50% of the 6187 study clients cited nonfinancial barriers including both
accessibility to the structure and appointment accommodation as the main reason for
the use of the HED for non-urgent health problems.

Affordability, the fourth dimension, has already been discussed as affecting the
use of the HED. While the cost of services for a primary care health concern is
notably much higher in the HED, the relationship between providers’ insurance or
deposit requirements and clients ability to pay is influential in the decision to use the
HED. Also affecting the client perception of affordability is the possible credit
arrangements with a provider. By law, a hospital cannot refuse to treat a person if
payment is not possible, whereas a private physician or a clinic can refuse to treat a
person with an inability to pay.

Acceptability, the fifth dimension of access, is defined as the relationship
between clients’ attitudes about personnel and practice characteristics of existing
providers (Thomas & Penchansky, 1984). This definition includes age, sex, location

and type of facility as well as religious affiliation of provider or facility. Provider

41

attitudes about acceptable personal characteristics of clients, including ethnicity and
client payment source, also influence the acceptability of a provider source. Figure 5
demonstrates that all four of the initial factors in the DM affect the five dimensions of
access.

Mdtivaddn m gk ease. The most important predictor of frequency of
utilization of health care services is client perception of the symptoms of an illness
(Parboosingh & Larsen, 1987). Client motivation to seek care is also influenced by
the combination of resources available at the time of need, such as financial
resources, availability and accessibility of services, and the sense of urgency of need
(Parboosingh & Larsen, 1987). A client may be motivated to seek care from his/her
primary care provider but the provider may not be available. Baker et al. (1994)
suggested a client with a regular source of care may be motivated to seek care at the
HED because of dissatisfaction with one’s own provider. One study (Young et al.,
1991) found the main motivation to seek care from the HED for non-urgent health
problems was the client’s desire for rapid treatment. The motivation to seek care
from the HED is a result of the previous variables and thus included in the DM.

We The ease of access to appropriate services is
affected by the acceptance of the client by the HED staff and the HED triage service.
The availability of alternative sources of primary care services at the time of need is
also a variable and has been discussed previously. The presence of triage personnel
able to make an appropriate referral to alternative sources of care or access to the
HED for care is also significant at this point in the DM. According to federal

Consolidated Omnibus Budget Reconciliation Act (COBRA) laws, the HED cannot

42

refuse treatment to a client who presents for care unless there is an alternate source of
care and a medical screening examination reveals a stable medical problem (Young et
al. , 1996).

Pereeptidns regnging [he 113d fer grvigs, While perceptions regarding the
need for services has been previously discussed, it is important to emphasize that
frequently there is a discrepancy between the perception of urgency held by the client
and the perception of urgency held by the provider (Frey et al. , 1994). Subsequently
the client may perceive more of an urgency of need than the HED provider of care
which results in client dissatisfaction with services. Further support for the client’s
perception is provided by Shesser et al. (1991) who found that most clients in their
study group expected to receive medical attention within twenty-four hours of their
decision to seek care.

Pregnce pf d ggjekmr, Starfield (1992) prefaces her chapter with the
phrase "Gatekeepers make sense” (p. 25). Most people do not know enough about
the technical details of medical care, as suggested in the previous sections on
perception of severity of illness and urgency need, to make informed judgments about
the appropriate time and source of care. Advice and guidance from a primary care
provider should be expected to facilitate the selection of the best source of care
(Starfield, 1992). The presence of a gatekeeper is inherent in health maintenance
organizations and has more recently been initiated by Medicaid in many states. The
purpose of a gatekeeper is to provide more rational, timely, and appropriate use of

resources. Therefore, there should be scientific rationale for the belief that primary

43

care providers can efficiently and effectively judge who should be referred to the
HED and who can be seen by a primary care provider (Starfield, 1992).

The concept of gatekeeper involves several dimensions. At the least a
gatekeeper functions to authorize a visit to an alternate source of care in a managed
care system. The decision to authorize a visit to an alternative care source may
depend upon whether or not the organization is paid for each enrolled member. Pre-
authorization for a visit as a means of gatekeeping is merely a phone call to an on-call
person to get approval to be seen at the HED.

At a more in-depth level, gatekeeping involves a client’s contact with his/her
primary care provider. When a visit or call to the primary care provider as
gatekeeper is required before the HED visit can be made, HED visits are reduced. A
study of 3000 patients enrolled in four Medicaid demonstration programs that required
a gatekeeper, demonstrated large reductions in the proportion of persons with at least
one HED visit. The reductions ranged from 27% to 37% for children and from 30%
to 45% for adults (Hurley, Freund, & Taylor, 1989).

Gatekeeping at the time of utilization of the HED is also reflected by the
presence of a triage system. The triage personnel may range from a registration clerk
in the smaller HED who registers everyone presenting for care at the HED to an
experienced registered nurse who assesses each client before treatment and determines
the level of acuity of illness. More recently, HED’s are attempting to triage non-
urgent clients to alternate sources of care if those sources are available.

Another type of gatekeeping which is growing in popularity is that of twenty-

four hour ask-a-nurse services. Ask-a-nurse services are offered by community

44

hospitals and managed care systems in order to more appropriately treat not only non-
urgent health care problems, but also urgent care problems. Ask-a-nurse services will
be discussed later in the implication section.

Vag'dples in the DM at the Qwer level

flisipn te assess ege in HED, At the bottom of the DM are the variables
which are the end result of interaction of the previous factors. The decision to access
care in the HED may be a cognitive process or an impulsive decision determined by
the perception of illness on the part of the client and client knowledge of services
available (Parboosingh & Larsen 1987). As discussed earlier, another factor which
affects the decision to access care in the HED is the availability of services in terms
of hours of service and ability to see walk-in clients (Hansagi et al., 1992). A client
who decides to seek care may realize that the HED will not require any payment at
the time of care, and thus the decision lies in affordability. Many times the decision
to seek care in the HED is not a cognitive process at all, but the result of the
perception of an urgency for care need.

Clients who are comfortable with the utilization of the HED for care may view
the providers of service in the HED as being more receptive, thus, the attitudes of
providers affect the use of the HED (Thomas & Penchansky, 1984). The client may
also have a high amount of satisfaction with care in the HED if care is rapid and
comprehensive in terms of diagnosis and treatment (Hansagi et al., 1992).

In summary, Figure 5, the Dirkse Model for the Utilization of the HED for
Non-Urgent Health Problems is a visual model of interrelated factors of causation. In

order to reap the benefits of the visual model, it is critical to discuss and formulate

45

interventions which may alter the variables and promote an improved outcome, that of
access to care for non-urgent health problems in a primary care setting. In the
following section, discussion will focus on interventions which are already impacting
the problem of inappropriate use of the HED. In the final section, interventions are
proposed in which the APN as an individual and APNs as a group of providers can

influence access to health care for clients.

CHAPTER 4

DISCUSSION

The development of the Dirkse Model for Utilization of the HED for Non-
Urgent Health Problems is significant for several reasons. In this section there are
three areas which have implications for health care professionals and deserve further
discussion. The first area of interest is that of access to primary care in general. The
second area is that of developing interventions to alleviate or obviate the problem
under scrutiny. The third area is that of interventions and areas of interest in which
there are already changes which attempt to decrease the use of the HED for non-
urgent health problems.

Aeeess te Primdg gage

Access to primary care is a complex issue with extensive literature written on
the topic, multiple suggestions, and varying interventions. While the DM is problem
specific, it can be generalized to access problems in alternate health care systems such
as a family practice clinic or a community health center. The visualization of the
contributing factors to the stated problem assists the health care professional interested
in the topic to understand the strengths and/or weaknesses in his/her own primary

care setting. The DM can be used in the evaluation of a health care system.

46

47

The DM can also be used to evaluate one’s own community HED for non-
urgent health care. In this evaluation, the strengths and weaknesses of access to
primary care in the immediate community would be of interest. For example, if a
community has a higher percentage of a vulnerable population known to have access
problems, identification of barriers to appropriate primary care services could occur.

Alternatively, the DM can facilitate the evaluation of the utilization patterns of
a community health center or a family practice clinic. (Questions of interest could
include: How quickly can appointments be made in the current system for an acute
problem? Does the setting offer walk-in or same-day appointments for an episodic
illness? What are the payment arrangements when the client does not have the ability
to pay for a visit? Is there a knowledge deficit on the part of the client concerning
hours of service or how to access services? Is the client dissatisfied with any aspect
of the service setting? With the DM, many of the factors identified in affecting
access to health care can be investigated and alternative interventions which increase
equity in access to primary care for all populations can be sought.

There are several assessment tools in the literature which can be used in
combination with the DM to further research in access to health care, and specifically
the population using the HED for care. Pane et al. (1991) used a pretested health
access survey in their study of indigent HED walk-in clients. In this study the
assessment tool was comprised of a written survey at the time of presentation for care
at the HED. Bilingual investigators were present to enroll patients, answer questions,

and collect surveys (Pane et al., 1991). The survey questions included client

48

demographics as well as questions regarding the reasons the clientd sought care at the
HED for non—urgent health problems.

Young et al. (1996) used a survey entitled "Twenty-four Hours in the Life of
an Emergency Department" for exploration of factors which prompted the study
population to seek care from the HED. The survey in this study was distributed
nationwide to site directors who in turn trained volunteers to administer the
questionnaire to prospective clients presenting for treatment at the HED. The
population used in the study by Young et al. (1996) was a nationwide cross—sectional
sample of 6441 ambulatory clients. Questions in the survey included the reasons why
the study clients used the HED for care and included such things as financial and
nonfinancial barriers to care, personal preference, and clinical reasons. The study
also compared clients with and without a regular provider who used the HED for non-
urgent care.

The questions in both of these studies (Pane et al., 1991; Young et al., 1996)
focused on various factors such as demographics, presence of a usual source of care,
and insurance status as well as the reasons why the HED was used by clients who
were considered stable walk-in patients. The variables in these surveys are reflected
in the DM; they are important factors for inclusion in future research. While many
of the variables in these two surveys are included in the DM, one could also use the
DM to develop an assessment tool for the purpose of further study in the problem of

non-urgent use of the HED.

49
Developing Infirventions

The second critical feature of the DM is its value in facilitating understanding
of the levels of prevention. In community health there are primary, secondary, and
tertiary levels of prevention of a health care problem; these are appropriate to discuss
in relationship to the DM and can be viewed as providing a framework for
interventions. Thus, a brief definition of the three levels of prevention are provided.

Spradley (1990) describes primary prevention of health problems as preceding
disease or dysfunction and that which keeps the problem from occurring. According
to Spradley (1990), "primary prevention involves anticipatory planning and action on
the part of community health professionals who must project themselves into the
future, envision potential needs and problems, and then design programs to counteract
them so that they never occur” (p. 16). The working paper from the AAN (1993)
and a committee established by the IOM Access Monitoring Project (1993) are
examples of primary prevention which promote equal access to health care services.

Spradley (1990) describes secondary prevention as the need to detect and treat
existing health problems at the earliest possible stage. Secondary prevention attempts
to discover a health problem at a point when intervention may lead to its control or
eradication. In the case of the use of the HED, secondary prevention is aimed at
redirecting access for non-urgent health care problem from the HED to appropriate
primary care services in which health outcomes can be maximized. The establishment
of a gatekeeper, which will be discussed later in greater depth, is a means of

secondary prevention.

50

The tertiary level of prevention "attempts to reduce the extent and severity of a
health problem to its lowest possible level (Spradley, 1990, p. 16). At this level in
the DM a client has already utilized the HED for a non-urgent health problem and the
focus lies in preventing repeated occurrence. Client education of more appropriate
sources of health care after the utilization of the HED and assistance in obtaining a
primary care provider are examples of tertiary prevention. In the discussion of
nursing implications, interventions will include primary, secondary, and tertiary levels
of prevention. Examples will be specific for the APN in a primary care practice.
Establishg Interventiens

The following paragraphs describe established interventions present in the
health care system to prevent non-urgent use of the HED by populations with
Medicaid, and to promote access to appropriate primary care services for this same
population. From the previous literature review on the increased use of the HED by
the population receiving Medicaid, recent changes in the system were noted. In an
attempt to discourage use of the HED by those with Medicaid, the Maryland Access
to Care Medicaid program (Gadomski etal., 1995) emphasized primary care and
appropriate health care utilization by incorporating the following elements of managed
care: client assignment to a primary care provider, gatekeeping, mandatory
enrollment, and fee for service. These elements are now implemented in other states.
Although the intent of the Maryland Medicaid program was to provide all enrollees
with a primary care provider who would also act as a first contact of care for all non-
urgent care needs, gatekeeping, while a safe practice, had no impact on subsequent

HED utilization by Medicaid participants (Gadomski et al., 1995).

51

The assignment of a primary care provider in managed care has resulted in
inconsistent outcomes. Frequently the primary care provider will grant permission to
access the HED for care because of unavailable appointment times. Enrollees may
also find it easy to change to another provider if they are not satisfied with the
assigned provider. To avoid these pitfalls, providers are reimbursed by a flat fee for
each enrollee and penalized for having the enrollee cared for at an alternative source
such as the HED (i.e., capitation of costs). The capitation issue in health care is a
topic of concern for it encourages a lower quality of care by increasing the total
number of clients in a primary care practice which may result in shorter visit time for
each client.

The establishment of managed care through health maintenance organizations
(HMOs) is an intervention to encourage non-urgent care by a single primary care
provider and limits the urgent care to be provided by HED personnel for which they
are educationally prepared. Managed care is a major health care reform in the United
States that both directly and indirectly affects the use of the HED. The intent of
managed care is to provide longitudinal care to individuals which allows them to
identify a source of care as ”theirs" (Starfield, 1992). The outcomes of managed care
ideally are higher quality of health care at a lower cost (Starfield, 1992). Managed
care is infiltrating not only private insurance but also, as mentioned earlier, public
aid. Managed care with the presence of a gatekeeper is part of the DM at a primary
level of prevention by way of community attitude, at the secondary level by way of

the presence of a gatekeeper to redirect care for non-urgent problem to a more

52

appropriate source, and at the tertiary level by way of non-reimbursement of HED
charges when used inappropriately.

Legislative changes in the past decade have focused on attempting to create a
narrower disparity in access to primary care services by changes other than managed
care. Some of these changes are discussed by the AAN in its working paper on
health care access problems and policy recommendations. The AAN (1993) suggests
that legislative changes are nwded to reorganize the care delivered at the federal,
state, and local levels through integrated planning and regulation. Health care
policymakers are influential in legislation that affects health care for at-risk
populations. While the establishment of community health centers increases the
primary care provider to population ratio, sliding fee scales for qualified persons, free
care for the indigent, and public aid for qualifying individuals are all ways a
community can increase access to health care.

Selby, Fireman, and Swain (1996) studied the effect of a co-payment on HED
use in a group-model HMO and found that the group with a co-payment had fewer
number of visits to the HED for non-urgent care. legislation regarding qualification
of individuals for public aid, the initiation of co-payment to share the responsibility of
payment, the funding of community health centers for the un-insured or underinsured,
and the assignment of a primary care provider, are all responsibilities of health care
policymakers and the community at large. These acts are the subject of the AAN
working paper (1993) for proposed changes needed to increase equity of access to

health services for all populations.

53

In summary, community and legislative involvement is presently in the midst
of implementing changes which attempt to decrease the use of the HED for non-
urgent health care problems and redirect care to primary care services. The DM
presented in this project can serve as an instrument to identify an array of
interventions at community and provider levels. The current and future contributions
of the APN in promoting access to primary care services and decreasing the non-

urgent utilization of the HED will be discussed.

CHAPTER 5

NURSING IMPLICATIONS

The nursing implications for the use of the DM in a primary care setting are
multifaceted. The DM can assist the APN in practice, research, and education. The
model can also provide a visual aid to illustrate all the factors which affect the access
to care for non-urgent health problems. This awareness will hopefully encourage
further study and research in the area of access to health care and in the APN’s own
practice setting.

In the earlier discussion of the Andersen and Aday (1978) model of health care
utilization, the concepts of mutable and immutable variables were introduced. When
developing interventions for the APN, it should be emphasized that there are variables
in the DM which cannot be changed. These variables include the client demographics
such as age, gender, culture, and race. Client socioeconomic status and client and
provider attitudes and beliefs are not readily mutable but may be altered with
interventions at the community level through programs, education, job preparation
and job placement. Client insurance status including public aid are frequently subject
to change. The interventions for the APN will focus on those factors which can be
targets for change such as client health status, community attitudes, presence of

primary care providers, and knowledge deficits.

54

55

Nursing literature refers to the roles of the APN as expert practitioner,
educator, consultant, and researcher (Hamric, 1989; Snyder & Yen, 1995; Sparacino
& Cooper, 1990). When developing nursing interventions using the DM, APN roles
which are most appropriate for discussion include the APN as practitioner, educator,
and researcher. The interventions hopefully can influence the variables in the DM
and decrease the non-urgent use of the HED. The nursing interventions will be
related to primary, secondary, and tertiary prevention.

W

The APN as practitioner can affect change in numerous variables of
interventions. Interventions at the primary level of prevention include an increase in
the number of APNs as practitioners in primary care settings. The recommendations
of the AAN (1993) include expanding the number of APNs practicing in areas of
undersupply to match the supply and demand for access to basic health care services.
By increasing the number of APNs as primary care providers, the immutable
variables of client demographics and socioeconomic status are indirectly affected in
terms of offering services to provide care for at-risk populations such as minorities,
those with chronic health problems, and the un-insured and underinsured (AAN,
1993). Thus, while client demographics and client SES cannot be changed, these
variables are impacted by the availability of providers in terms of numbers.

Also at the primary level, an APN as provider of primary care impacts client
health status when engaging in health promotion, illness prevention, and chronic
illness management activities. Included in health status are clients who present with

psycho-social problems; APNs are well-suited to serve as case managers, counselors

56

and collaborators with the health care team for clients with psycho-social problems.
The APN in the primary care setting will encounter many clients whose health care
needs include psycho-social components and as the first contact of care and case
manager, the APN can increase the overall health status of the client (Naegle, 1993).
The APN can also impact clients by developing longitudinal relationships with the
clients which can result in improved health outcomes. Additionally, the APN can
affect the attitudes of clients and other providers by presenting themselves as positive
role models.

An issue of concern with the APN as practitioner is that of reimbursement.
The AAN (1993) recommends the adoption of reimbursement reforms to ensure
standard fees for specific procedures and care regardless of provider type or setting.
With adoption of reimbursement reforms for primary care services, the APN can
deliver care in the home or in other community settings including health centers. To
promote the use of the APN in practice in primary care, removal of barriers to
practice at the state and federal levels through legislative action is recommended
(AAN, 1993). The APN nwds to be both a leader and a role model by actively
participating in local and state organizations which in turn should strengthen future
changes and prevent barriers to practice. Every APN must be a member of his or her
local, state, and national nursing organization to maintain involvement in issues of
concern.

Also at a primary level of prevention are measures to encourage financial aid
for the un-insured or underinsured. legislation regarding qualification of individuals

for public aid, the initiation of a co-payment to share the responsibility of cost of

57

care, and the funding of community health centers are all aimed at increasing primary
care services for individuals without insurance or those who are underinsured. APNs
can affect legislative acts as registered voters, as members of their nursing
organizations, and as public speakers.

The APN can affect funding in his/her client advocate role by securing
assistance for health care needs for the indigent. One example in which an APN can
affect funding is to communicate with local services and organizations to convey the
needs of special populations who may not receive adequate health care. In the
author’s own practice setting, a migrant health center, clients are very rarely able to
receive adequate services for referral services such as radiographs, dental care, or
ophthalmic care because of their inability to pay. Several community groups,
working through the county health department, were able to secure state funds to
assist the migrant to afford needed services. Throughout the past five years, the
funds had not been used to the maximum because of the mechanisms for securing
assistance and language differences, so at the present time allotment of vouchers for
special needs has been given to the health center manager to avoid an extra step in the
process of securing help for the migrant. This type of client advocacy is one role
component in which an APN can affect the health status and community awareness of
need.

The variables in the DM at a secondary level which can be influenced by an
APN in practice include availability, acceptability, and accommodation to health care
services. An increase in the number of APNs as providers of primary care can

increase the availability and accommodation of primary care services. Acceptability

58

of services can be addressed by an increase in APNs who may be of like culture or
have a better understanding of the cultural beliefs of a client (AAN, 1993).

The APN, acting as a gatekeeper, can affect the use of the HED for non-
urgent health problems. With an increase in APNs as primary care providers in many
settings, APNs are able to share responsibility for call services and act as gatekeepers
in managed care situations. As a gatekeeper, the APN can directly affect access to
the HED at the time the problem occurs.

The gatekeeper phenomenon has extended into phone triage of health cere
services and is another variable in the DM. A recent article in the Wall Smt
Mpg; (February 5, 1997) described how experienced nurses function in health care
organizations performing phone triage. As an example, Access Health Inc. in Denver
has a staff of 90 registered nurses manning shifts behind a desktop computer.
According to the author (Anders, 1997), approximately 35 million Americans now
have access to phone-triage lines. The function of the nurse is to elicit information
regarding the symptoms for which a client calls and, by following established
protocols, give competent advice. According to industry estimates, approximately 2 %
of the callers are actually steered to the HED with a savings of $2 in reduced
emergency-room use on each dollar spent on an advice line (Anders, 1997). While
most triage phone services are staffed with registered nurses specially trained in the
area, the APN in primary care acts in the same manner when on-call for a group

practice with managed care clientele, and must make similar decisions.

59
The APN as Educator

The APN as educator can effect change at all three levels of prevention. The
educator role is a fundamental component of the APN which distinguishes the APN
from other health care professionals (Sparacino & C00per, 1990). Education is
directed towards not only the client, but also families, groups, and communities. As
an intervention in the DM, education is applicable to all three levels of prevention.

The APN as educator can affect the variable of health status at a primary
prevention level with client education focused on health promotion and prevention.
By increasing emphasis on health promotion and prevention, long-term negative health
effects can be reduced, thus improving the general health status of the client. With an
increase in the number of APNs in practice as health educators and expert assessors,
clients can become aware of health risks at a younger age. An example would be the
recognition of those at risk for hypertension and diabetes, in which early recognition
can lead to improved health outcomes by means of client education and promotion of
healthier lifestyles.

Also at the primary level, an APN can educate community members about the
role of the APN as a provider of primary health care. Community education is
directed at marketing the APN by means of public speaking to various groups in the
health care and broader community. Community education by APNs organized as a
professional group can have a significant impact. In the author’s own community,
several leaders within the community hospital assisted area APNs to form such a
group. The APN professional group has as one of its goals marketing the APN

within the community. Suggestions for marketing the APN in the community include:

60

writing articles on the role of the APN for local newspapers and hospital newsletters,
encouraging the APN to speak publicly at local organizations and clubs, local
television programs which focus on health care issues, and local talk radio programs
whose audience includes many of the community’s elderly population.

At a secondary level of prevention, that is, at the time a client decides to
access the HED, an APN can influence the DM variable, illness perception of the
client, through client education. An APN, by definition of the role, is adept at
educating the client about illness. For example, this can include instruction on fever
care to avoid unnecessary visits to the HED and instruction on distinguishing between
signs and symptoms of minor illnesses which can be cared for at home or at a
primary care visit, as opposed to the signs and symptoms which indicate the need for
urgent care at the HED.

Also at the secondary level of intervention, the APN as educator can affect
client knowledge of available services as a gatekeeper. The gatekeeper, already
discussed, functions as the first contact of care. The APN as the first contact of care
in the primary care setting can educate the client on the appropriate use of the HED
versus appropriate use of the primary care setting. When the APN functions as a
gatekeeper by taking call as a primary care provider in a health maintenance
organization, the APN acts as an educator in addition to preauthorizing a visit to the
HED. For example, when a mother with a febrile child calls the service, the APN
can do the following: solicit information on patient signs and symptoms, inquire as to
the use of any home care such as Tylenol administration, offer patient education on

fever care, and refer to the HED if indicated.

61

The APN as educator at the tertiary level can affect client knowledge of access
to services at the time the HED has been used by a client with a non-urgent health
problem. Alternatively, an APN can educate the client seen in the primary care
setting as to appropriate utilization of the HED, and increase client awareness of the
availability, accessibility, and the accommodation to his/her primary care setting as
well as how to access one’s own primary care provider. For instance, many times a
client cannot make a prompt office visit because of office mechanisms such as busy
phone lines or receptionist perception of need for care. The APN can instruct the
client to request to speak to the office nurse who has the ability to fit a client into a
provider’s schedule.

At the tertiary level of prevention, education of the client to appropriate access
to future primary care services is essential. If the DM is adopted by professionals in
the HED, a greater understanding of client use of the HED may be used as an
educational opportunity at the time of service. If the client does not have a regular
provider, available primary care services in the community may be recommended by
the HED staff. Knowledge of the role of the APN in primary care by the HED
personnel can promote vulnerable populations to access care from a primary care
provider such as an APN.

The APN as Remher

The DM can serve as a framework for future research. An APN can influence
the non-urgent use of the HED with participation in research aimed at specific factors
in the DM. Participation in research can take place at many levels for the APN:

principal investigator, project director, co-investigator, or consultant (Hampton &

62

Snyder, 1995). The APN can target research to all three levels of intervention. At
the primary level there is a deficit of information in the area of health outcomes and
the APN. With the focus in the present health care system aimed at increasing the
number of primary care providers such as the APN in a community, research is
needed concerning the effects such an increase has on the utilization of the HED.

Research at the secondary level of prevention is needed in the area of access to
existing services. Several surveys were reviewed in this project for future use in the
research of access to primary care services, or the access to the HED by clients with
non—urgent health care problems. These surveys can be used to focus on specific
variables in the DM. For example, research can focus on the relationship between
insurance status and use of the HED, on the impact of HED use after an increase in
the number of APNs as primary care providers, or on client satisfaction.

Research at the tertiary level has been demonstrated in several studies used in
this project (Pane et al., 1991; Young et al., 1996); however, none of the studies
were nursing-oriented. The APN possesses knowledge about the research process
(Hampton & Snyder, 1995) and is qualified to provide input in using research as a
means to evaluate access to primary care services. The APN as practitioner of
primary care and trained in research techniques can integrate these two roles to affect

evaluation and change in the non-urgent use of the HED.

CHAPTER 6

CONCLUSION

The use of the HED for non-urgent health problems has been the subject of
numerous investigations in which various contributing factors were identified and
analyzed. However, no study or reference attempted to analyze the issue by
diagrammatic display. This project’s product was an initial schematic representation
of this issue through the utilization of the ”Web of Causation" model, a problem-
oriented model. The representation, the Dirkse Model for the Utilization of the HED
for Non-Urgent Health Problems, identifies factors from existing literature which
contribute to the use of the HED for non-urgent health problems. Application of this
model should assist the health care professional, including the APN, to develop
interventions which focus on specific factors essential to decreasing the use of the
HED for non—urgent health problems. The model can enable practitioners to identify
points for future interventions at the primary, secondary, and tertiary levels of
prevention. The model can also serve as a framework for future research in the area
of access barriers to primary care services.

This model is likely not complete and will need to be altered in the future as
further changes occur in the issue of access to primary care services. The

implications for the APN include but are not limited to the suggestions offered in this

63

64

project. With an increase in the numbers of APNs in both primary care settings and
the HED; it is hopeful that the DM will encourage APNs and other health
professionals to examine and evaluate factors and interventions concerning the

utilization of the HED for non-urgent health problems.

LIST OF REFERENCES

LIST OF REFERENCES

Adler, N. E., Boyce, W. T., Chesney, M. A., Folkman, S., & Syme, S. L.
(1993). Socioeconomic inequalities in health: No easy solution. lonrnal pf

American Mfiigl Asmian'dn, 2Q9(24), 3140-3145.

Aiken, L. H., Lewis, C. E., Craig, J., Mendenhall, R. C., Blendon, R. J., &
Rogers, D. E. (1979). The contribution of specialists to the delivery of primary

care. New England .19an pf Medieine, 300(24), 1363-1370.

American Academy of Nursing. (1993). Wand

W. (Working paper). Washington DC: American Nurses
Association.

Anders, G. (1997, February 4). Telephone triage: How nurses take calls and
control the care of patients from afar. The Wall Smt lgumal, pp. A1, A4, A6.

Andersen, R., & Aday, L. A. (1978). Access to medical care in the US:
Realized and potential. Mgllealganedfifl), 533-546.

Andren, K. G., & Rosenqvist, U. (1985). Heavy users of an emergency
department: Psycho-social and medical characteristics, other health care contacts and
the effect of a hospital social worker intervention. Smial Science and Medieine,
2_1(7), 761—770.

Andren, K. G., & Rosenqvist, U. (1987). Heavy users of an emergency

department: A two year follow-up study. W(D,
825—83 1.

Baker, D. W., Stevens, C. D., Brook, R. H. (1994). Regular source of
ambulatory care and medical care utilization by patients presenting to a public hospital

emergency department. rn f Ameri M i A i 'n 271(24),
1909-1912.

Berk, M. L., Schur, C. L., & Cantor, J. C. (1995). Ability to obtain health

care: Recent estimates from the Robert Wood Johnson Foundation National Access to
Care Survey. Health Affairs, 14(3), 139—146.

65

66

Blendon, R. J. Aiken, L. H., Freeman, H. E., & Corey, C. R. (1986)
Access to medical care for black and white Americans. lpu mal pf Amerign Medieal

AWOL 278- 281.

Buesching, D. P., Jablonowski, A., Vesta, E., Silts, W., Runge, C., Lund,
J. , & Porter, R. (1985). Inappropriate emergency department visits. Amalspf
Em r n M i ine 14(7), 672-676.

Calnan, M. (1984). The functions of the hospital emergency department: A
study of patient demand. Journal pf Emergeney Mfl'eine, 2, 57-63.

Clark, M. J. (1996). Nursing in the Community. (2nd ed.). Stamford, CT:
Appleton & Lange.

Collado, C. B. (1992). Primary health care: A continuing challenge.
Nursing 3; Health Cm, 13(8), 408-413.

Epstein, A. M., Stern, R. S., Tognetti, J., Begg, C. B., Hartley, R. M.,
Cumella, E., Jr., & Ayanian, J. Z. (1988). The association of patients’
socioeconomic characteristics with the length of hospital stay and hospital charges

within diagnosis-related groups. Th N w En l rn f M i inc 1 (24),
1579—1585.

Frey, L., Schmide, J., Derksen, D. J, & Skipper, B. (1994). A rural
emergency department WJQ 1). 38-42

Gadomski, A. M., Perkis, V., Horton, L., Cross, 8., & Stanton, B. (1995).
Diverting managed care Medicaid patients from pediatric emergency department use.

Pfiiatries, 95(2), 170—178.

Grumbach, K., Keane, D., & Bindman, A. (1993). Primary care and public

emergency department overcrowding. Ameriean leumal pf belie Hedda, 83(3),
372-378.

Haddy, R. I., Schmaler, M. E., & Epting, R. J. (1987). Nonemergency
emergency room use in patients with and without primary care physicians. lnurnal pf

Family Practig, 24(4), 389-392.

Hampton, J. K., & Snyder, M. (1995). Research. In M. Snyder & M. P.
Mirr (Eds.), Adv Practice N rsin : A G id Professi n Dev 1 m 11 (pp.
229-240). New York: Springer Publishing Co.

Hamric, A. B. (1989). History and overview of the CNS role. In A. B.
Hamric & J. A. Spross (Eds.), Th Clini N r ' '
(pp. 3-18). Philadelphia, PA: W. B. Saunders Co.

67

Hansagi, H., Carlsson, B., & Brismar, B. (1992). The urgency of care need

and patient satisfaction at a hospital emergency department. HealtnganeA/lmage
Review 17(2), 71—75.

 

Hilditch, J. R. (1980). Changes in hospital emergency department use
associated with increased family physician availability. qurnal pf Family meme,
1_1_(l), 91—96.

Hurley, R. E., Freund, D. A., & Taylor, D. E. (1989). Emergency room
use and primary care case management: Evidence from four Medicaid demonstration
programs. Am ri rn f li H th 7 (7), 843-846.

Institute of Medicine. (1993). M. Millman, (Ed.). mirage-41mm
Ameriea. Washington, DC: National Academy Press, pp. 19-44.

Kasper, J. D. ( 1987). The importance of type of usual source of care for
children’s physician access and expenditures. W6), 386-398.

Krieger, N. (1994). Epidemiology and the web of causation: Has anyone
seen the spider? Social Seienee Mgieine, 32(7), 887-903.

Lehman, C. A. (1995). Risk factors for pressure ulcers in spinal cord injury.
Seientifie Nursing, 12(4), 110—114.

Lozon, M. (1997, February 16). Prince brings the doctor to work. The
Helland Sentinel, pp. C1, C2.

Mayefsky, J. H., El-Shinaway, Y., & Kelleher, P. (1991). Families who seek

care for the common cold in a pediatric emergency department. Jettrnalmnam
mm), 933—934.

The Medicaid Access Study Group. (1994). Access of Medicaid recipients to
outpatient care. New England leurnal of Mfiicine, 339(20), 1426-1430.

Merriam-Webster’s collegiate dictionary (10th ed.). (1995). Springfield, MA:
Merriam-Webster, Inc.

Millis, J. S. (1977). Primary care: Definition of, and access to. . . .
Nursing Qutledk, 25(7), 443-445.

Naegle, M. A. (1993). Incorporating primary care concepts into psychiatric
nursing roles. In M. D. Mezey & D. O. McGivem (Eds.), Nurses, Nurg
maitioners (pp. 232-250). New York: Springer Publishing Co.

68

Oot-Giromini, B. A. (1993). Pressure ulcer prevalence, incidence and
associated risk factors in the community. Dguuitus, 6(5), 24-32.

Pane, G. A., Farner, M. C., & Salness, K. A. (1991). Health care access
problems of medically indigent emergency department walk—in patients. Annals pf

Emergeney Medieine, 20(7), 730-733.

Parboosingh, E. J ., & Larsen, D. E. (1987). Factors influencing frequency
and appropriateness of utilization of the emergency room by the elderly. Mfideal

M02), 1139-1147.

Purdie, F. R. J., Honigman, B., & Rosen, P. (1981). The chronic
emergency department patient. Annals pf Emergeney Mmieine, 19(6), 298-301.

Rask, K. J., Williams, M. V., Parker, R. M., & McNagny, S. E. (1994).
Obstacles predicting lack of a regular provider and delays in seeking care for patients

atanurban public hospital. m f Am ri M i A i 'n 71(24),
1931-1933.

St. Peter, R. F., Newacheck, P. W., & Halfon, N. (1992). Access to care

for poor children: Separate and unequal? Jdumal pf Ameriean Mfiieal Assgugian'pn,
M00), 2760-2764.

Selby, J. V., Fireman, B. H., & Swain, B. E. (1996). Effect of a co-
payment on use of the emergency department in a health maintenance organization.
N w En l urn fM i in 4(10), 635—641.

Shesser, R., Kirsch, T., Smith, J ., & Hirsch, R. (1991). An analysis of
emergency department use by patients with minor illness. AnnalLQLIirnetgeney
Medieine, 20(7), 743-748.

Snyder, M., & Yen, M. (1995). Characteristics of the advanced practice

nurse. In M. Snyder & M. P. Mirr (Eds.), Advmgfl Pnaeu'ee Nursing; A Guide tn
Prefessional Develppment (pp. 3-12). New York: Springer Publishing Co.

Sparacino, P. S. A., & Cooper, D. M. (1990). The role components. In P.
S. A. Sparacino, D. M. Cooper, & P. A. Minarik (Eds.), The Clinifl Nurse
Smlalist (pp. 11-42). Norwalk, CT: Appleton & Lange.

Spradley. B. W. (1990). WWW

Glenview, IL: Scott, Foresman and Company.

Starfield, B. (1992). Primag Que; gzpneept, evaluatien, md mliey. New
York: Oxford University Press.

69

Thomas, J., & Penchansky, R. (1984). Relating satisfaction with access.
Mfiigl glare, 22(2), 553-568.

Young, G. P., Wagner, M. B., Kellermann, A. L., Ellis, J., & Bouley, D.
(1996). Ambulatory visits to hospital emergency departments: Patterns and reasons

for use. .19qu pf Ameriean Medin Assmiatinn, 276(6), 460-465.

arr-r‘r“m

NIH/Wll(HI/WWII}WIN/HIlllflI/HIIM/l/l/ll .

 

02427 1243