-;k>

.‘ I‘-

..‘ {y

4-“...
. . A.
. .

-~L

 

it”?
v; .
afié“

'v. '1‘
>- "i

. ‘ "V, .
L‘...3‘.a.t,u.—' '<

~

'

ԤM#*w

3"} ‘

 

no

 

    

:11 .. ‘ - '7_- »- -;. 7.". ,__ .. “u. -j.' p‘.‘ 1
{ > ‘ ‘.._ ,- . -_ jq‘-:‘ ..,__'_>. ‘7 ‘ '.'>

“ { I " ‘. ’ .'-_ -..,, ..‘ » . ‘ A‘ ..‘... W'a-fi :‘kf: I ’1'.""

9 . ‘ . 7 . .., - - - ‘ ., - - g.

. ‘ - , , .

j’l ' 7 l ..- ' L -

v' M. - _ ‘ .

. ‘ A 7 - v >-

:. ;~ ,. »7

 

  
   
    
 

     

V n» o. .
o _— _ __ .. .
._., . - .-.- _.
V ' "
on - - _; . , _
. . A ' ‘ :' . ’- —
_ ‘ -
h" ’ v ‘ ~ 7 ~~ "
, -u. '~" ' .55 Z. -— '__
o . D —
J , - ‘ _ .. — J..,.
‘2': ‘ v ‘ . ' ' ' V V' a'
t; . " . I 7 . 'A -- ‘ ‘
‘ ‘ ‘- ' '- ' ‘ ‘ 't' 1‘4; v.
1-, _..- - - .' ' ' H "
[Lip 3; : ..__ 7 ___. o'. 1 , L).
r ’4' L.‘ . -_’ _, 1.» H.
n” v - ~ , - . - A
’v. .‘— ' - c . -—l _ ' A F.
, A . , ' . — " ' '
‘7- v, ‘ . - - “tr F .—
I"." V w. ‘ - . 'v y o "‘ ‘ H - .
on >’ ‘ ' - ‘ .d— A '
,-. ;.. ; .. :1- - ,J a“ -r.:=
I ‘ ' . fl- -'
2‘5- ' ‘ 7' ' ’ : ' ' - .J ,
_-J " ,‘p ’ l ‘_ (- . _'__A . ‘ d ‘- I

 

Em,

1::- 50

~ rests.

w ,9.) ‘ 1'1

u u Imam-lit
-

- Diem-II

;' '-¢__ '5. '
,..;-~-.-... Ii
fizmzm

Had -‘u-‘u-

,

I 0
~ «.1 .. '1'35‘?‘ wage: cm!"
can. i Viv-—

 

 

 

This is to certify that the

dissertation entitled

Broken Health Screening Appointments
Among Low Income Families

presented by

T. Richard Currier

has been accepted towards fulfillment
of the requirements for

degreein_§gc_igl_scj_ence

Jan

//

Major professor V

Date February 7, I986

MS U i: an Willem! Action/Equal Opportunity Inflation 0-12771

 

 

 

MSU

LIBRARIES
:—

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

 

 

 

BROKEN
HEALTH SCREENING APPOINTMENTS

AMONG LOW INCOME FAMILIES

by

Richard Currier

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Social Science

1986

finial/(0:33;

ABSTRACT

BROKEN HEALTH SCREENING APPOINTMENTS
AMONG LOW INCOME FAMILIES

by

Richard Currier

Missed appointments for health care create costly ineffi-
ciencies for medical providers, threaten the validity of
research efforts, and often subject the person involved

to unnecessary morbidity. Many studies have been conducted
on appointment keeping behavior of patients to ascertain
predictors of persons who fail to appear and who, reportedly,
average 15 percent to 33 percent of all scheduled appoint-
ments. Most studies focus on demographic factors. Few
studies take into account variables related to patient be-
liefs, social behavioral and perceptual characteristics,
aspects of the disease, its therapy, patient-provider in-
teraction, and environment and organizational features of

the providing facility. The present study attempts to exam-
ine the relationships of a large number of variables included
in previous studies believed to be correlates of appoint-
ment keeping behavior. The data were collected in a compre-
hensive survey conducted on eligible Medicaid persons in
Michigan who have been scheduled for a health screening ap—
pointment. The survey should identify correlatives of ap-
pointment keeping behavior. To accomplish this, a cluster

analysis was used to form clusters of variables that

Richard Currier

illuminate patient attributes relative to appointment
keeping.

Few findings of other studies have been confirmed in this
research. However, one result is clear, namely, the shorter
the time between scheduling and appointment, the more likely
it is that an appointment will be kept. For the most part,
the 81 variables measured showed little correlation with
scheduling outcomes. It seems reasonable to conclude that,
for future research on appointment keeping, investigators
should explore the contribution of specific factors to

appointment keeping behavior.

DEDICATION

to

Fran Gram

who is the music to these words

ii

ACKNOWLEDGEMENTS

The example and encouragement of George W.
Fairweather will long be remembered. Six
years ago, at Ocean's Edge, when fainter
hearts asked, why? He said, "Why not?"

For his direction in a course that was long
and sometimes stormy, I owe much to Professor
Fairweather.

A special thanks is due to Delrose Komosinski
who provided timely and expert typing of this

manuscript.

iii

I.

II.

III.

Iv.

TABLE OF CONTENTS

INTRODUCTION

Background. . . . . . . . . . . .
Assumptions . . . . . .

BACKGROUND OF THE PROBLEM

History of Health Screening . .
History of EPSDT in the United States
Need for Health Screening Among The
Poor. . . . . . . . . . . . . . .
Major Stumbling Block for Success

of EPSDT. . . . . . . . . . . . .

A Review of Previous Studies on
Failure to Keep Appointments. . . . .

Specific Hypotheses . . . . . . . . .
METHODOLOGY

Sampling. . . . . . . . .

Scoring and Field Testing of. Survey

Instruments . . . . . . . . . . . . .

Method of Analysis. . . . . . . . . .

RESULTS

Test of Hypotheses. . . . . . . . . .
Relationship Among Measures . . . . .
Cluster Description . . . . . . . . .

DISCUSSION . . . . . . . . . . . . . .

APPENDICES
A. Letter of Invitation. . . . . . .
B. Client Questionnaire. . . . . . .

C. Client's Opinion Questionnaire.
D. Medical Outreach Worker
Questionnaire . . . . . . . . . .

BIBLIOGRAPHY . . . . . . . . . . . . . .

iv

22
25
27
49
53
66
67

69

97
104
120
148
149
152
156

158

LIST OF TABLES

Conditions Currently Recommended for
Screening Programs . . . . . . . . . . . . . . . . .

Inventory of EPSDT Screening Procedure as
Typically Used in State Programs . . . . . . . . . . .

Distance of Patient From Clinic. . . . . . . . . . .

Test of Patient Demographic Characteristics
and Appointment Keeping. . . . . . . . . .

Distribution of Responses on Each Demographic
Measure by Percent . . . . . . . . . . . . . . . . .

Test of Patient Related Social Behavior or
Perception Variables and Appointment Keeping .

Distribution of Responses on Each Social Behavior
or Perception Measure by Percent . . . . . . . .

Test of Provider Related Characteristics
and Appointment Keeping. . . . . . . . . . . . . . . .

Distribution of Responses on Provider Related
Characteristics by Percent . . . . . . . . . . . . . .

Test of Characteristics of the Disease
and Appointment Keeping. . . . . . . . . . . . . . .

Distribution of Responses to Disease
Characteristics by Percent . . . . . . . . . . . . . .

Test of Characteristics of the Therapeutic
Regimen and Appointment Keeping. . . . . . . . . . . .

Distribution of Responses to Therapeutic
Regimen Characteristics by Percent . . . . . . . . . .

Test of Patient-Therapist Interaction
Characteristics and Appointment Keeping. . . . . . . .

Distribution of Responses to Patient-Therapist
Interaction Characteristics by Percent . . . . . . . .

16

40

71

72

73

74

77

78

79

80

83

84

86

87

Test of Facility Accessibility and
Appointment Keeping. . . . . . .

Distribution of Responses to Facility
Accessibility Characteristics by Percent

Test of Facility and Administrative Process
Characteristics and Appointment Keeping. . . .

Distribution of Responses to Facility/Adminis-
trative Process Characteristics by Percent .

Test of Characteristics of the Environment
and Appointment Keeping. . . . . . . . . . .

Distribution of Responses to Characteristics
of the Environment by Percent. . . . . .

Variables Found to be Significantly Correlated
With Appointment Keeping at the .05 Level of
Confidence . . . . . . . . . . . . . . . . .
Variables Found to be Significantly Correlated
With Appointment Keeping at the .10 Level of
Confidence . . . . . . . . . . . . . . .

Nine Clusters of Scale or Variable Scores.

Correlation Between Oblique Cluster Domains. .

vi

. . . 88
. 88
. 91
. 92
94

. 95

. . . 97
. 98

. . 103
105

I. INTRODUCTION

Background

 

Today emphasis in health care is shifting increasingly
toward prevention. This is due largely to the growing aware—
ness in the general public that the major sources of illness
are damaging personal behavior, e.g., smoking or lack of
exercise, and the environment in which we live. To a certain
extent, we have control over, or at least power to modify,
both sources of ill health in that we can change our behavior
or improve our environment. Another important development
fostering preventive orientation to health care delivery is
advancement made in science whereby it is possible to detect
defects and disease in early stages of development when re-
medial care is most likely to be effective.

Escalating costs in correcting health problems is, per-
haps, the strongest force for raising the consciousness of
people towards the need for prevention, early detection, and
treatment. Government is faced with these costs in providing
health care for low income families. The cost of providing
comprehensive medical care to the medically indigent can be
controlled and even reduced if there is effective cooperation
between providers of services and recipients. The need for
this cooperation is illustrated most strikingly in a national
health screening program supported by federal and state funds
for children of low income families. In this study, we will

briefly look at the nature of screening, its history, its

1

current form in the Early and Periodic Screening Diagnosis
and Treatment (EPDST) Program, and then focus, primarily,
on a major stumbling block in achieving the goals of mass
screening, namely, failure to keep scheduled appointments.

Assumptions

 

There is a vast sea of literature relevant to the
problem of missed appointments for health care. This re—
search will attempt to look at, comprehensively, the myriad
of variables examined by other researchers in order to iden-
tify those variables associated with appointment keeping be-
havior. To achieve the goal of this study, certain assump—
tions are necessary.

It is assumed that the attitudes and behavior regarding
one's own health care will be extended to one's attitudes
and behavior toward health care for the children in one's
care. Very little research has been done on this issue.
Jayne Linley (1984) found that the mother's view of health
and the health care system ultimately determines the kind
of health care the child receives. Furthermore, the health
behavior of a caregiver is defined by the culture. In most
instances, poor health habits, undue risk taking in health
matters, delay in seeking health care, and disregard to me-
dical advice are direct reflections of social attitudes
(Suchman 1970). Becker, et a1, (1977) showed that the
health beliefs of the mother are critical in utilization of
preventive care for young children. These findings have

important implications in the health treatment of children.

The health care provider must take time to include the care-
giver in the planning process for health care of the child,
promote caregiver interactions, elicit caregivers reviews
about the child, provide to the caregiver an understanding
of the health needs of the child and, in general, provide

encouragement.

II. BACKGROUND OF THE PROBLEM

History of Health Screening

 

From ancient times, efforts were made to screen out
members of a society as a method of prevention. Isolation,
banishment, and even death were effective in achieving this
purpose. Leprosy, for example, was effectively controlled
in Europe, in part, by isolating lepers in leprosaries,
which, by the 13th century, could be found everywhere through-
out Europe (Winslow 1923).

Late 18th and early 19th century progress focused on
environmental sanitation, as it was gradually realized that
somehow poverty and filth were associated with disease. This
development was stimulated, primarily, by writings of John
Howard (1773) on British prisons, by Anthony Cooper's (1802)
writings on child laborers in British factories, and by Ed-
win Chadwick's writing on the conditions of the laboring
population of Britain in 1842. Britain's Sanitary Act of
1886 followed by similar legislation in America (Winslow
1923) was the result of this new awareness.

The modern age of preventive medicine must be credited
to Louis Pasteur. His studies of fermentation ultimately
led to an understanding of the relationship between micro-
organism, environment, and host, thereby placing preventive
medicine on a truly scientific basis. In the 1880's, vac-
cination was developed for cholera, anthrox, rabies, diph-

theria, typhoid, and smallpox. Koch described the tubercle
4

bacillus and his theories became the cornerstone of micro-
biology (Winslow 1923).

With these historic developments, preventive medicine
could expand from public health, with its emphasis on en-
vironmental efforts, to the early diagnosis and treatment
of individual persons. The new understanding, also, pro-
vided a foundation for future possibilities of screening
large numbers of people, based on the idea of early de-
tection of clinical and laboratory signs before onset of
clinical symptoms.

Public Health reached its Golden Age during 1890 —

1910 because of the application of scientific discoveries

to the prevention of disease. Malaria, yellow fever, bu-
bonic plague, typhoid, typhus, cholera, diphtheria, small-
pox, pertussis, and tetanus were brought under control.

More importantly, the success lead to the replacement of a
feeling of helplessness to one of great optimism. Armed
with new understanding, we were in a position to gain consi-
derable control over our health and well-being.

In 1909, the National Committee for Mental Hygiene
(MCMH) was founded by Clifford Beers. The remarkable achieve-
ments of preventive medicine and public health provided the
philosophical and methodological foundation for this organi-
zation. MCMH predicted that mental disorders would be con-
quered, just like the infectious diseases, if only the Ameri-
can people could be trained to recognize early signs of ment-

al disease (MCMH 1912). It was assumed that what worked for

the disease of the body would also work for those of the
mind.

The introduction of science into the realm of preven-
tive health led to pre-occupation with measuring things,
charting progress, and generally moving beyond speculation
into the realm of quantitative data. This new technology
of measurement was soon applied to gauging intelligence
and predicting behavior. Early diagnosis and treatment of
the criminal and the juvenile delinquent became a priority
item with the advent of large numbers of immigrants to ur—
ban ghettos (Russell 1912, p. 189). The First World War
institutionalized the practice of measuring mind and be-
havior. Thousands of soldiers were tested for the purpose
of military classification (Brigham 1923). The testing of
intelligence became generally accepted so that IQ became a
household word. About this time, also, the eugenics move-
ment developed and adopted the practice of screening.

In the 1920's, there arose the child guidance movement.
This movement served as the major vehicle by which juvenile
delinquency was to be controlled. The child guidance move-‘
ment saw the converging of the three new disciplines of
psychoanalysis, juvenile court case work, and psychological
assessment. The child psychiatrist, social worker, and
psychologist formed an interdisciplinary team that became
an arm of the juvenile court (Healy 1948). Freud's emphasis
on early childhood was entirely consistent with the preven-

tive goals of Adolph Meyer and other founders of the mental

hygiene movement who, themselves, were crucial figures in
organizing the child guidance movement (Stevenson 1948).
The American Foundations (especially the Commonwealth Fund
and the Rockefeller Foundation) became the primary backers
of the new delinquency prevention efforts (Karier 1972).
The goal was a child guidance clinic in every community as
the best weapon against juvenile delinquency. It was deemed
vastly preferable to identify and treat a would-be delinquent
rather than wait until the person became a hardened criminal.

In the 1930's and 1940's, the practice of measuring
skulls or earlobes as well as the widespread belief in gene-
tic factors in criminality and deviance prediction fell into
disrepute. Despite this screening and other brave efforts,
there was a gradual waning hope. The techniques of predic—
tion never materialized; juvenile deviance seemed, if any-
thing, to be growing despite the spread of child guidance
clinics. It became apparent that, notwithstanding the hopes
of the mental hygiene movement and the child guidance move-
ment, nothing comparable to the glorious revolution in pub—
lic health was on the horizon. Accordingly, sights were set
more realistically and greater availability of services to
children and families became the goal. Delinquency no longer
dominated the concerns of mental health screening. It was
believed that psychonalysis helped many, and attention was
increasingly turned away from the large number of poor.

After World War II, with the influx of federal dollars,

the National Institute of Mental Health (NIMN) was organized

in 1949 and, ultimately, the Community Mental Health Centers
Act of 1963 formalized the relationship between federal dol—
lars and psychiatric programs. This gave rise to new inter-
est for a public health model for prevention of mental prob-
lems. If the 19th century attributed madness to the lack of
discipline, hard work, and the other virtues of rural America,
and the early 20th century looked for bad genes, bad hygiene,
and lack of education, by the mid-20th century there was a
concern, more and more, of early treatment of biochemically
based mental disease.

In more recent years, schools of thought in public health
made more precise a definition of health screening, health
problems to be screened for, and criteria for health screen-
ing tests.

Definition of Health Screening

 

The United States Commission on Chronic Illness, in 1957,
proposed a definition of health screening as follows:

"The presumptive identification of unrecognized
disease or defects by the application of tests,
examinations, or other procedures which can be
applied rapidly. Screening tests sort out, ap—
parently, well persons who probably do have the
disease. A screening test is not intended to be
diagnostic. Persons with positive or suspicious
findings must be referred to their physicians for
diagnosis and necessary treatment."

Health Problems to be Screened.

 

Since 1951, scientists have recommended screening for
an increasing greater number of conditions. A few of the
pediatric conditions suggested for screening (Camp 1957,

p. 9), are listed in Table I.

TABLE I

Conditions Currently Recommended For Screening Programs

 

Hearing

Vision

Speech

Language

Development

School Readiness

Lead Poisoning

Anemia

Sex Chromosome Abnormalities
Congenital Dislocation of Hip
Rheumatic Heart Disease
Inguinal Hernia

Congenital Heart Disease
Dental Problems

Apgar

Bacteriuria

Tuberculosis

Venereal Disease

Color Blindness

Phenylktetonuria

Maple Syrup Urine Disease

Hypercholesterolemia

Hyperlipoproteinemia

Wilson's Disease

Calactosemia

Hereditary Angioneurotic
Edema

Cretinism

Mellituria

Succinylchonilesterase
Deficiency

Glucose-6—Phosphate Dehy-
drogenase Deficiency

Gargoylism

Tay-Sachs Disease

Learning Disorders

 

Criteria for Health Screening Tests

 

Screening seeks to identify a disease or developmental

delay at an early state or prior to that point in time when

treatment is less effective.

The time should ideally be

prior to signs and symptoms that would normally be noticed

by the individual or guardian since, usually, diagnosis is

sought after signs begin to appear.

A careful distinction should be made between health

screening and health services.

The former is an attempt to

find health problems, while the latter is remedial in nature.

To meet the objectives of screening, efforts must generally

meet certain criteria such as those outlined by William

Frankenburg (1978).

The criteria are as follows:

10

The Condition Is Potentially Serious.

The seriousness takes into account such factors as
remedial costs, if delayed; loss of earning power in the
future; spread of disease, if not checked; permanency of
the damage; human suffering involved; and the like.
There is a point when the cost of locating a malady out-
weighs the screening investments. A balance needs to be
struck between screening and treatment.

The screening procedures used can effectively iden-
tify persons at risk. Most conditions related to health
follow a continuum with many shades of gray. It is quite
useless to screen for conditions that cannot be identi-
fied specifically enough to warrant remedial care.

Treatment Is More Effective With Early Intervention Than
If Delivered When The Condition Becomes Obvious.

Treating a child for the muscle imbalance in the
eye (amblyopia) after the age of five is ineffective.
Intervention prior to that age can prevent blindness
and may result in restoring normal sight to the eye.
This is an example of where a screening test is criti-
cal for identifying the individuals needing early in-
tervention.

The Condition Is Treatable Or At Least Controllable.

Since health screening is justified as a means of
enhancing the health of the individual if a dangerous
condition is uncovered, but no treatment is available,

it is quite obvious that the very purpose of screening

11

is defeated. Some may question, as an example, the
value of screening for sickle cell anemia since there
is no known remedy for this condition. Since sickle
cell is genetically based, the objective of screening
in this case is not to treat but to control by provid-
ing counseling.

The Disease Should Be Relatively Prevalent.

The disease of smallpox is now so rare that time,
expense, and effort to find cases of it are relatively
fruitless. The rarer the disease, the greater are the
resources needed to find it. Therefore, a balance needs
to be established between the prevalence of a disease,
and the amount of resources available to find it.

The Procedure Does Not Cause Undue Harm To The Individ-
ual Tested.

Invasion of privacy and stigmatization can easily
occur on finding sickle cell anemia. Job opportunities
may be limited or promotions denied as a result of the
disease. Provision must be made to prevent harm to the
individual in screening, especially since clients are
aggressively sought after in the screening process rather
than the reverse, that is, when a client seeks health
care for a known and felt health problem.

Resources Are Available For Remedial Care When A Problem
Is Found.

This requirement for justifying a health screening
relates to the previous requirement regarding avoiding

harm to the individual being screened. If resources

12

are not available to the individual being screened, the
anxiety of the person is raised by uncovering the health
problem.

Cost For Both Discovery And Remedy Should Take Into Ac-
count Both Monetary And Human Misery Dimensions.

Monetary costs include collecting specimens/infor—
mation, testing, equipment, maintenance, supplies, ad-
ministration, follow-up, record-keeping, diagnosis,
and treatment to name some of the areas. The investment
of human misery in the discovery and treatment process
is to be weighed against fiscal considerations.

Screening tests are to be characterized by accept-
ability, reliability, and validity. The acceptability
of a screening test refers to the consensus among health
professionals, and the general public as well. Tests
that are painful or embarrassing are less likely to be
accepted by the public. Reliability refers to consist-
ency of getting the same results from a test when per-
formed by different peOple. Validity refers to the ac-
curacy of the test in uncovering a specific health con-
dition that needs care. It is not expected that a sub-
sequent diagnosis will agree with findings from a screen-
ing process as testing involved in diagnosing a problem
are much more elaborate and specific than would be used
in a screening program. The problem is not with the
false positives, but with the false negatives in which

case a health problem goes undetected in the screening.

13

History of EPSDT in the United States

 

The Early and Periodic Screening, Diagnosis and Treat-
ment (EPSDT) Program, with its roots in the screening efforts
in the past as reviewed above, blossomed in the liberal at-
mosphere of the 1960's. The civil rights movement, and the
ghetto uprisings left, in their wake, a vast array of anti-
poverty programs designed to lead to the "Great Society".
Unfortunately, there was often a tendency to "blame the
victims" (Ryan 1972). Various forms of deviance (school
failure or dropout, family breakdown, violence, drug abuse,
and the like) were conceputalized as indications of some
type of deficiency or deprivation syndrome of poor peOple.
The remedy usually involved doing something to the needy,
aimed at improving their lives in someway.

High among the objectives of the "Great Society" was
good health to poor children. The Department of Health
Education and Welfare (DHEW) formulated the Early and Perio-
dic Screening, Diagnosis, and Treatment (EPSDT) program, and
submitted it to President Johnson who signed it into law
in January, 1968. The program was designed as an amendment
to Title XIX of the Social Security Act through which pay-
ments were made for medical services delivered to the poor.
The EPSDT amendment went beyond a mere payment system for
medical care requiring states to provide "such early and
periodic screening and diagnosis of individuals who are
eligible under the plan and are under the age of 21 to as-

certain their physical or mental defects and such health

14

care, treatment, and other measures to correct or amelior-
ate defects and chronic conditions discovered thereby, as

may be provided in regulations of the Secretary." (Public

Law 90-248, Sec. 302a)

States are required to provide screening and certain
other types of diagnosis and treatment for hearing, dental,
and vision problems even if these are not already included
in the Medicaid State Plan for Medicaid eligible individuals.
Persons under the age of 21 also are to receive all mandated
Title XIX services apart from EPSDT, including inpatient
and outpatient hospitalization, physician services, labora-
tory and x-ray services; and family planning. They may also
receive other Optional services under the State Medicaid
plan. The unique aspects of EPSDT include:

- Arrangements for screening when they are not otherwise
available.

- Specific standards for screening.

- A systematic attempt to integrate children into health
screening, diagnosis, and treatment through outreach,
referral, and follow-up,

- Health education.

- Provision for diagnosis and treatment of hearing, dental,
and vision problems if they are not otherwise included
in the State plan.

The EPSDT program can be briefly conceptualized in the fol-

lowing manner:

Goal: To improve the health status of Medicaid-
eligible children under the age of 21.

Objective: To prevent chronic disabling illness.

To increase treatment where appropriate.

15

To change the attitudes of recipients in
the area of health care.

Sub-objectives: To introduce recipients into the health
care system as early as possible.

To establish continuity of health care.

To establish a comprehensive evaluative
base for health care.

To increase the accessibility and avail-
ability of health care systems.

To increase recipient awareness of health
measures and encourage their use as appro-
priate.
The principal assumptions on which the EPSDT program is based
are that:

- Health can be improved through medical intervention.

- Medical intervention is best done through the early
diagnosis of problems.

- Early diagnosis is best done through preventive screen-
ing.

- Preventive screening is best done through a government
program.

- Appropriate government programs are not now available.

- If available, this program would be utilized by potential
clients.

— Problems uncovered would be treated rapidly.
- Treated problems would be corrected successfully.

— Medicaid-eligible children are most in need of health
improvement.

- Medicaid-eligible children are without alternative
delivery systems.

- Medicaid-eligible children would use this special
government program.

16

sumo Hmucov ou mmmoom ousmmm ou Hammm a“ surname“
ma moua>uom Hmuaov mo uamcanoo m .mouawoooua
uaoaummuu mam Ufiumocwmfiv Hmusov poms HHH3 sou
Icafiso Ham umoaHm umsu mmuUHvoum on new uH moafiw

.mwm mo memo» moo
akucm3u 0m nuufim

.mwm mapfimmoa umoaaumm ecu um ommomfiw
manmuco>oua aoum vmuumuoua ma vHHnu >uo>o umsu
magma“ cu meow ma msumum coaummficsaaw so xomso <

.owm mo mummm woo
IxucoSu cu nuufim

.COfiumcHeuooo use unmem>os mo coaum>uomno cam
.ousmmmua cooan .cowuooomcfi ocuiouiwmmc mumaoaoo m
mmvnauca coaumcasmxo one .uumawm: uo\v:m mmsnm
eaano mo mocmva>m mcfinoausa .vmooam>ow coon m>mn
momma wcficomuom eumvcmum oc nuan3 now mamas

noun Luann: mam mommmmwv mmosu um>oomwv wmmum
wcficmmuum ecu maam: cofiumcfimeo Hmofimzsa one

.owm mo mumm» oco
Izucm3u cu :uufim

.nuamm: ou vmumaou mmofiuomua nonuo cam .muwpmz
Hmcofiufiuuac m3oa>mu .mamanouo nuamo: maow>oua mam
mmnwuommv .mumu nuamm: msofi>muu m.:u:0% no vafino
m mo maamoua m mmnfi>oua uH .msumum nuamoa wag
Immommm Ga amum umuam mam ma zuOuman Hmofivme one

.mwm mo mpmm» moo
Ixucoau cu :uufim

 

uwmumme mHm>oa ow< "mmomuam

mEmumoum oumum CH

unmaumowe
mmmmmaa Hmucon

AmfimmSuummv swaou
wafiaoonz .mmasz

.AmHHmnamv mmHmmmz

:mEuou .Amaoonsmv
mmammmz .owHom
.Amscmuoav 3mm

-xuom .aauaeusamn

waHvSHUCa msumum
coaummficsaaH

mcoaumefiamxm
Hmoammnm

zuoumfim Hmowwoz

 

"pom wcficompum

com: mHamogozh mm mouswmooum wcwcmouum Hommm mo mucuco>cH

N mqm<e

l7

.owm mo mumm% mco
Ihucm3u ou fiuufim

.mwm mo mummm mco
Izucosu cu nuuwm

.mwm mo
mwmoz xfim Ou Luufim

.owm mo mumom oco
Izucmsu cu nuufim

.owm wo mums» mco
Ixucmau Ou Luuam

.mwm mo mummx
occlxucoau ou xwm

 

"vmummh mHm>mg mw<

.mwmamv zoapfix unoCmEHoa Ou wmoa and mcofiuoomcw
uumuu xumcfius wouooumvc: mm maouaahm on usn
.mGOfiuoowca uomuu zumafiu: o>m£ 0:3 mamsvfi>wv
Ida uwfiucovfi ou wmsfiamxo ma coafioomm mafia: <

.oowuuomcw mfimoasoumnau um>oumfiw
ou mum: ma >uH>wufimcmm cflasuuonau How ummu m
.ufimmoua ma mwmoasouonau oumc3 mCOfiumasaom cH

.ucmaaoao>mv Hmoamhna no .HmCOHuoEo
.Hmofiwoaousoc .Hmowwoaozohmo ca owmuo>m osu Eouw
ummwwv maucmuamacmam 0:3 couvawno xwwucona cu
com: ma Aucmaaoam>mw Hmcowuocsm cam uwcmwuo no
mcoumoafie cmcfimow wo msumu SH mmmumoua m.wafino
ecu mo Hmmfimuoam cmv acoammommm Hmucmanam>mQ

.uomawm:

no :owuwuuscuovc: .maaemxm pom .Suaouw HmEHoa
cuwa mumwumucfi :ofin3 mcofiuwvcou no mmmomfie mmwu
lemma mam: Ou vmnwfioa cam venomous mum convaanu

.ucmEo>ma£um Hmcoaumosvo mam mafia Hmfiuom uamnu
now: mummumuaa Ou unmaoammsm wcaummn voosvou
o>mn 0:3 convafinu >mau¢mvfi ou voxomco ma wcfiummm

.vawco ecu mo cowumusvm mam
mam ucmEQon>me ozu Sofia ououpouca vasoo Lows:
mucoEufimmEa Hmsmfi> uumumv cu cocHmem mum mmmm

.ucosumouu cam coaumsam>o
ofivofiuoo u0m umwucoc mnu ou xomp vaano mnu women
Ou cam .euoomu amazon usuamsuoa m nmfianmumm ou

Homomusm

A.e.ucouV N mmm<e

mwwafiumuomm

>ufi>auamcmm

caflaonmnse

ucmEmon>oo

ucoEmmmmm<
nu3ouu

wcfiumom

mEmHnoum mzm

 

"mom wcficomuom

18

.omm mo mnmom oco
Ixucozu cu sunnm

.mwm wo
mnmoz o>nw On one

.owm mo mnmmm oco
imucoau on :nnnm

 

"woumme mHo>mA ow<

.c0nuuavonomn unonm mconmnomv coEnOHGn mxme Ewan
mam: cu wcnaomsnoo cam conumsnowcn ao>nw onm
Hamu maxunm auna manomm .mSOuQE%m mo wonamn ecu
now pmnnowon mnm masons Hamo maxonm £nn3 manomm
.mnEmcm Hamo maxonm rename mn Snow mno>om m cam

.unmnu Hams oaxonm voaamo mn Snow page < .mHaomn

cmmcmnnmunwmz 6cm xomam wcoEm venom mHHou wooHn
won man we conunpcoo onumamw m mn Hams maxonm

.wcncomnoa vmmH ecu mo monsom wan menu

cmo menu umnu om mHmnonwmo gunmen unansa unmam
Ou cam wcncomnoa puma Eonm nnmov can nunann
immnw ucm>mna on cmmmommm mn Ho>oH umoa wooam

.moonuomna HmGOnunnusc noon
Eonw unsmmn zme gonna mnaocm >ucononmowlconn

mwnucmmn On wmumamcm new camnv mn waqsmm moons ¢

Homomnsm

A.e.uc00v N mam<n

nnmo anxunm

conumnomnm
vmoq

mHEmc<

 

"now wcncomnom

19

The testing procedure, purpose, and age levels tested
as commonly used in the EPSDT program are indicated in Table
2 (above).

As previously mentioned, the EPSDT program was not a
sudden development but had roots in the mass screening de—
velopments in the past. From the standpoint of government
involvement with its vast resources of billions of dollars,
more specific steps of development can be identified. EPSDT
reaches back through maternal and child health activities
to 1935. In that year, the Social Security Act provided
for cash payments to families with dependent children. This
action of the federal government forms the corner stone of
government involvement in future social programs. Families
were paid cash directly to allow them to seek medical care.
Many used the money for other needs and health care was
sought only at a time of crisis. This act also provided for
the Crippled Childrens Program through which states received
money to locate children with a variety of disabilities and
provide medical, surgical, and corrective services.

During the Second World War, the Emergency Maternity and
Infant Care program was enacted to provide for maternal and
infant care of servicemen's dependents. Families of service-
men moved to small towns near military bases. This put a
heavy strain on local health care systems. In what was the
largest public health undertaking at the time, the federal
government set up a system of preventive and treatment pro-

grams to care for the health of wives and children of

20

servicemen. Following the war in 1950, the Social Security
Act was amended to provide federal funds to states to help
meet the cost of medical care for the low income families.
Ten years later, the Kerr-Mills Act provided the first pro-
gram of comprehensive health care for older people. During
the years around 1960, a great deal of attention was focused
on the difficulties encountered by older people. The Kerr-
Mills Act attempted to provide health care for older Ameri-
cans whose income was too high for public assistance but too
low to afford health care. While many states did not imple—
ment this program, a model of comprehensive health care to a
population in need was created that became the blueprint for
Medicaid and its EPSDT component. In 1965, Medicare and
Medicaid (Titles XVIII and XIX of the Social Security Act)
financed health care for the aged and eligible poor. Medi-
caid is the parent program for EPSDT.

In the context of this increasing social welfare involve—
ment of the federal government in the lives of the elderly,
mothers, and children, the amendment to Title XIX of the
Social Security Act (EPSDT) came about in 1967 to provide
comprehensive health services for all Medicaid eligibles
under the age of 21 years. Because states did not heed the
new law, five years later states were required to implement
EPSDT or face a penalty of one percent cut of the federal
share of their Aid to Families with Dependent Children (AFDC)
budgets for every quarter in which they failed to comply

with the EPSDT regulations.

21

Anne-Marie Foltz, who has probably earned the title of
historian for EPSDT, notes that the EPSDT amendment slipped
through Congress late in 1967 with very little comment.

DHEW did not formulate regulations until late in 1971 with
a stipulation that full implementation was not required
until mid-1973 (Foltz 1982, p. 1).

States responded to the federal law almost as slowly as
DHEW. Foltz notes that, five years after the regulations
were issued, only two million children out of an eligible
population of twelve million were screened. The cost of
EPSDT was minimal as federal social programs generally go.
As late as 1976, health screening costs were only $47 mil-
lion or less than 0.3 percent of Medicaid expenditures (DHEW
1979). The states were slow to respond because of the un—
known costs that would result from treating health problems
of millions of children found during the mass screening
program. Although the federal government paid half of the
cost of treatment, a sizeable amount would be left to the
state. Perhaps another partial explanation for the slow
implementation of EPSDT is the traditional attitude of the
American people to resist government involvement in their
lives. It is not our purpose to address the issue of why
the United States Government has become so involved. For
a discussion of this, see the writings of Peter Morris (1969),
Daniel Moynihan (1969), and Frances Piven (1971). Due to
slow implementation, Congress, in 1973, roundly scolded

DHEW and the states (U. S. Comptroller General 1975).

22

DHEW was accused of "mismanagement" by a House Subcommittee
(U. 8. Congress 1976).

EPSDT blossomed with a number of other social welfare
programs in the 1960's known as the era of the Great Society.
A host of scientists, sociologists, economists, and lawyers
have written about the shortcomings of the Great Society
(e.g. Karen Davis (1972), Charles Haar (1975), Jeffrey
Pressman (1973), Gilbert Steinen (1971), and Eli Ginzberg
(1972). EPSDT, however, should not be thought of as simply
one of the social programs initiated by the Great Society
because, as outlined above, the EPSDT program has roots
reaching back many years. .

Need for Health Screening Among the Poor

 

Insight into the problems of not showing up is of
special urgency in combating the health problems of the
poor through comprehensive screening, since the poor need
special attention because of the higher incidences of health
problems. International statistics on infant deaths show
that infant mortality rate in the United States is worse
than in 14 other countries (Demographic Yearbook of the
United Nations 1973). After infancy, contrast in mortali-
ty figures remain. Among children aged 1 to 4 years,
minority children die at a rate of 70 percent higher than
whites. And, in the 5 to 9 age group, minority children
die at a rate of 40 percent higher than white children
(National Center for Health Statistics 1975). Special

health services help to reverse racial discrimination of

23

the past 35 percent of the nation's non-white families fall
below the poverty line, and almost another 35 percent exist
at a near poverty level. For the white population, only 10
percent are below the poverty level and 10 percent are near
the poverty level (HEW Report 1967). Figures on how many
American mothers die in childbirth show that three times as
many non—white as white mothers die (National Center for
Health Statistics 1974). Deschin (1967) demonstrated a
positive correlation between indigent maternity and high
ratio of prematurity, perinatal mortality, drug addiction,
and serious housing violations. Many low income mothers
appear to be locked into a low health status as they fail
to respond to health care when it is made available. In a
study of pregnant women from five socio—economic classes,
Gallagher (1961) discovered that the mothers in the lowest
economic class are the least likely to receive prenatal
care, are more likely to have their babies at home, and
are less likely to seek immunization shots and check-ups
for their infants. Gallagher explains this situation is
due to inconvenient health care facilities, lack of motiva-
tion, and stringent regulations.

Statistics cited provide an inferential argument link-
ing poverty and health status. There is debate in the li—
terature regarding the precise relationship between poverty
and health. It may be useful to focus on how low socio-
economic status affects health.

Epidemological studies point out that the entire range

24

of medical pathology seems to be affected adversely by low
socio-economic status (Morris 1969). In answer to the ques-
tion, why this is so, Pratt (1977) explains that, ..."one of
the mechanisms through which poverty adversely affects health
is a deficient pattern of personal health care which includes
deficiencies in personal health maintenance practices, use
of medical services, health knowledge, and health supportive
equipment in the home".‘ Poor nutrition affects health
status, especially in regard to pregnancy outcome and the
incidence of pregnancy complications, according to a study
conducted by Singer (1968). Earlier pregnancy and a greater
frequency of pregnancy are found among low income women
and are found also to mean greater risk for mother and in-
fant (Birch 1970). Physical crowding occurs more frequently
among low income groups. This may have significant effect
on health status of the people so affected in that it pro-
motes infectious disease and poorer environments conducive
to lead poisoning, poor quality/unsanitary food, unsafe
water, accidents, and injuries. Damage caused by these
factors may result in a lifetime of neurological and neuro-
psychiatric damage (North, 1973). Every year, thousands
of children suffer from lead paint poisoning. In the course
of a year, 300 to 400 children die from it, and 6,000 more
suffer irreversible brain damage (Green 1976).

Perhaps the Watts area of Los Angeles exemplifies what
is meant by a poor environment. This area contains only 17

percent of the city's population but, in category after

25

category, it harbors nearly 50 percent of the city's ills.
It had 48.5 percent of amoebic infections, 42 percent of
food poisoning, 44.8 percent of whooping cough, 39 percent
of epilepsy, 42.8 percent of rheumatic fever, 44.6 percent
of dysentary, 47 percent of venereal disease, 36 percent
of meningitis, and 65 percent of reported tuberculin re-
actors. The death rate in Watts was 22.3 percent higher
than the remainder of the city (Hurley 1969). The incident
of tuberculosis is four times higher in Watts than in the
rest of Los Angeles County (Birch 1970). Watts is not an
isolated example. In a Chicano community in California,
children display a whole range of vivid contrasts with
national averages: four times as much amoebic dysentary,
twice as much measles, mumps, and tuberculosis, and 1.4
times as much hepatitis (Urban Health Study 1973). Child-
ren reared in such environments are likely to have ill
health that will become cyclical.

ngor Stumbling Block for Success of EPSDT

 

This researcher, with many years of experience with the
EPSDT program, finds that, when the screening tests and pur-
pose for each are explained clearly (as indicated in Table
2) to a prospective child caregiver, ninety percent respond
affirmatively to scheduling a screening appointment. Keep-
ing the appointment, however, is a different issue. Not
showing up for appointments averages around 35 percent.

The problem of missed appointments is a major stumbling

block to the health status of millions of eligible children.

26

The individual involved suffers damage by way of unchecked
health problems that may be difficult or impossible to solve
with long delay. The provider of care also is hurt because
the time set aside for an appointment often is not effective-
ly utilized. Finally, the community itself suffers because
a resource available for the betterment of its members is
not adequately used.

Where studied, the missed appointment problem has usu-
ally been in the context of treating chronic conditions that
require prolonged treatment. The extent of missing appoint-
ments can be inferred from the fact that, in general psy-
chiatric clinics, 20 percent to 57 percent of the patients
fail to keep appointments after the initial visit (Blenkner
1954, Dodd 1971, Overall and Aronson 1963). Similarly, in
group psychotherapy from 35 percent to 50 percent of the
patients fail to attend meetings (Berne 1955, Sethne and
Harrington 1971). The picture is even less encouraging in
the outpatient treatment of alcoholism, in which from 52
percent to 75 percent of the patients fail to show for con-
tinued treatment (Baekeland, Lundwell and Shanahan 1975,
Blane and Meyers 1964). Similarly, 37 percent to 60.2 per-
cent of tuberculosis patients fail to follow through on
treatment (J. Newman and Spares 1956, Wilmer 1956). In
the equally life-threatening condition of hypertension,
from 20 percent to 50 percent of the patients do not show
for treatment recommended (Armstrong, et. a1. 1962).

Thus, it is clear that failure to keep appointments

27

to follow through even on known and, at times, life-threaten-
ing situations is a major problem throughout the health care
industry. Such a state of affairs is of particular concern
at the present time, which is marked by a simultaneous repeat
increase in curative and preventive know-how and costs. It
seems important for the effective and efficient use of these
resources to be able to distinguish the person likely to keep
appointment for health services from those who will not. It
is equally important to identify the clinic and treatment
features thacpromotethe not showing up behavior on the part
of the client so that they can be changed.

It appears sufficient evidence exists to justify a spe-
cial national effort to provide health services to low income
families. For the past 10 years, Congress has, in fact, au—
thorized such an endeavor through the EPSDT program. The
objectives of this program are to prevent chronic and dis-
abling illness; to change the attitudes of recipients in
the area of healthcare; to introduce recipients into the
health care system as early as possible; to establish con-
tinuity of health care; and to increase recipient awareness
of health measures and encourage their use as appropriate
(Michigan EPSDT Annual Report 1980). These goals are un-
attainable for millions if appointments are not kept.

A Review of Previous Studies on Failure to Keep Appointments

 

The purpose of this present study is to provide a road
map through the mountains of studies conducted on the issue

of keeping medical appointments. Previously, studies have

28

used a fragmented approach in that a limited number of
variables were considered in an effort to assess their
impact on appointment keeping behavior. This present study
will look, comprehensively, at all the factors included in
other studies to assess the association of these variables
with failure to keep appointments.

For the sake of clarity and ease of management in
conducting a literature review of previous studies, vari-
ables examined by researchers will be broken down into
eight general categories suggested byHaynes (1976). These
include features of the patient (demographic and social
behavior); features of the medical provider; features of
the disease or reason for appointment; features of the
patient/provider interaction; features of the therapeutic
regimen; features of the medical facility and adminstrative
process; features of access to the facility; and environ-
mental features.

Features of the Patient

 

Demographic data on patient drOputs are available in a
large number of studies but are often inconsistent and unen—
lightening. It appears on the whole, however, that younger
age correlates with higher rates of broken appointments in
a variety of settings (Jonas 1971, Shronick 1977, Hertz
1977, Hurtado 1973). Delk (1975) found that the mean age
of those who keptappointments was six years greater than
that of patients who dropped out of treatments with over

61 percent of discontinuers being 25 years of age or

29

younger. Studies have found that patients 45 years of age

or older generally have lower rates of missed appointments
(Gates 1976). Age is not a good predictor in the pediatric
population, although some studies show a higher failure

rate when the patient is less than one year old. Low socio-
economic status correlates with broken appointments in
studies conducted by Alpert (1964), Badgley (1961), Hoenig
(1966), and Fiester (1975) as does educational level. Level
of education is shown to influence appointment keeping be-
havior. Stine, et a1 (1968) showed a significantly increased
incidence of failed appointments in patients who did not
complete high school compared to a group of high school gradu-
ates. When age, education, and socio-economic status are
accounted for, race is probably unimportant (Delk 1975,

Adler 1962). However, widely desperate conclusions have been
reached on the issue of race as it relates to missed appoint-
ments. Three studies suggest that non-white populations have
a significantly increased rate of missed appointments (Alpert
1964, Badgley 1961, Jonas 1971). In a more recent analysis
of this issue, Hertz and Stamps (1977) found that low income
patients and patients from ethnic minorities are more likely
to break appointments. They reviewed over 3,000 patient
visits to a Model City's Health Center which serves, primar-
ily, low income groups and welfare recipients. Hertz and
Stamps found that difference in the kept appointment rates

of different ethnic groups does not reflect a true ethnic

30

difference in health behavior. Craig (1976) notes that low
income, inner city black patients are, at least, as capable
as other patient groups of sustained therapeutic contact.
In view of this, clinics that serve such patients and set,
as their goals, episodic, crisis—oriented services may be
underscoring people who could benefit from a more long-term
therapeutic experience.

Patient demographic variables of gender, occupational
status, marital status, and religion appear to have little
correlation with attendance behavior (Baekeland 1975, Gates
1976, Dervin 1978). In sum, age, education, and socio-econo-
mic status are probably the only consistently important demo-
graphic influences on appointment keeping behavior.

Fewer studies are available that examine the social be-
havior characteristics of the patient, but trends can be
identified. Patients perceptions of the seriousness and sus-
ceptibility of a disease and belief in the efficiency of
therapy appear to be an important correlate of compliance
with medical recommendations in general, including appoint-
ment keeping behavior (Tash 1960, Kegeles 1973, Haefner 1970).
Furthermore, the correlations found, and some experimental
results, suggest that effective interventions can be based
on altering these beliefs. For example, efforts to increase
the patient's perception of the threat posed by his/her ill-
ness have been helpful in childhood obesity (Becker 1977).
Controversy exists, however, regarding the chronology of

these beliefs and whether they precede health behavior or

31

develop simultaneously (Taylor 1979). The patient's know-
ledge of his/her disease and therapy appear to correlate with
appointment keeping (Tagliacozzo 1978, Caldwell 1970), and
educational efforts can improve broken appointment rates
(Glowgow 1970, Shmarak 1971). Such efforts might focus es-
pecially on patients who have dropped out in the past, as
these are likely to repeat the behavior (Baekeland 1975).
While education helps to modify a person's behavior in re-
gard to appointment keeping, it does not suggest that action
will follow on other health related behavior. Personal prac—
tices, such as dietary patterns and other long-established
habits, were not markedly influenced by changes in belief
induced by education (Haefner 1970). Probably habits and
long-established behaviorial patterns engage many motives
that include, but go beyond, health care. Altering a person's
beliefs about health may be sufficient to change actions

that are largely motivated by health matters but will, usual-
ly, be insufficient to alter behaviors that simultaneously
satisfy a variety of motives (Haefner 1970).

Features of the Provider

 

Most studies focus on the characteristics of the patient
in predicting appointment keeping behavior, however, charac-
teristics of the provider are also important predictors.
Hoenig (1966) found that female therapists treating female
patients experienced higher appointment keeping rates than
other sex pairings. In general, male therapists, in the

psychiatric literature, are more likely to lose their

32

patient (Baekeland 1975). Age of the clinician is only
weakly correlated with appointment keeping with kept ap—
pointments favoring the older health care practitioner
(Hurtado 1973). Data comparing physicians assistants or
nurse practitioners with physicians are few. However, in
studies where practitioners or physicians assistants were
allocated randomly to providers, rates of missed appoint-
ments and droputs were as good or better for the mid-level
practitioner as for physicians (Spitzer 1974, Bessman 1974,
Burnip 1976).

Attitude and behavior of the provider of care is an
even higher predictor of appointment keeping behavior than
the socio-economic status of the patient. Although looked
at principally, but not exclusively, in the context of out-
patient psychotherapy, important findings have been reported
by Baum 1966, Sethna 1971, Dinnen 1971, and Howard 1970.
According to their findings, some of the characteristics
of the therapist who is apt to drive his/her patient out of
treatment include the following: ethnocentrism; unconcern
for, or dislike of, or boredom with the patient; and ne—
gative feelings about the use of medication and/or reluc-
tance to give it. The therapist is also apt to be made
moody, to instruct his charge inadequately (especially with
regard to the use of medication), to cancel appointments,
to be permissive, introverted, and detached. The kind of
person to whom this therapist is most likely to give less

attention to is the lower socio-economic status patient.

33

It is, of course, difficult to know how much a patient's
propensity to miss an appointment is simply the result of
attitudes he/she brings, or how much it is a reaction to
boredom, incomprehension, and dislike on the part of the
practitioner.

Of the variables which have been found to be signifi-
cantly related to a patient's decision to keep appointments,
change of provider of care appears to be of paramount im-
portance (Delk 1975). This holds true for patients involved
with psychiatric treatment in which good interpersonal re-
lationships with the patient must be developed and maintain-
ed to be effective. Findings such as this in the psychia-
tric literature may not hold for patients seeking care from
other medical specialties. In cases where provider replace-
ment is unavoidable, the impact on the patient can be soften—
ed somewhat by strengthening those elements in the treatment
program that enhance the perception of continuity in the eyes
of the patient. The provider also can ease the situation by
taking time to study the patient's case history and consult
with his/her perceptor before initial contact with a patient.

Features of the Disease

 

Some features of the patient's malady may have a bearing
on clinic attendance. Patients with a psychiatric diagnosis
appear to have higher rates of failed appointments than those
without a psychiatric diagnosis (Alpert 1964, Delk 1975).
Rubenstein (1976) found adult asthmatics to have a higher

rate of failed appointments and late arrivals than patients

34

with other allergic diseases. The numbers, however, were
very small, and the series was uncontrolled for many impor-
tant factors. John Walsh, et. a1. (1967) studied 12,364
visits to outpatient clinics. His study tracked appoint—
ment keeping behavior for a variety of specific health
services. These include health care related to general
services: medicine, dental, pediatrics, perinatal, oph-
thalmology, orthopedic, general surgery, otolaryngology,
urology, dermatology, tuberculosis, and Hansen's disease.
He found that the more specialized a service was, the more
likely the patient would keep an appointment. Thus, the
highest rate of missed appointments was for general medical
services (25 percent), pediatrics (10 percent), general
surgery (9 percent), and dispensation of medication (15
percent). In contrast, the highest kept appointments were
for such specific services as Hansen's disease, tuberculosis,
urology, and dermatology for which the missed appointments
were, overall, 1 percent, 2 percent, and 3 percent, respec-
tively.

Walsh found that nearly one in every ten patients can-
cels or fails to keep his/her appointment for one reason or
another. For every patient who cancels an appointment, three
fail to cancel and fail to keep their appointments.

Duration of illness was found to be unimportant (Badgley
1961, Click 1965), but patients with chronic disease have

been found to have fewer broken appointments than patients

35

with acute illness (Hurtado 1973, Shah 1977). Also of
importance is the presence of symptoms or a specific con-
cern. Sackett (1976) reports that patients with a speci-
fic complaint have a compliance with appointments between
70 to 87 percent, compared with studies of patients without
a specific complaint with an appointment compliance of 47
percent. Related to this finding, appointments for screen-
ing or diagnosis are broken more often than those for physi—
cal examination as follow—up of illness (Gottlieb 1962);
patients with severe functional impairments do not keep
their appointments as faithfully as patients with intermedi—
ate degrees of impairment (Glowgow 1973); and patients capable
of full time work do not keep their appointments as well as
those with intermediate degrees of impairment (Glowgow 1973).
Features of the Therapeutic Regimen

Features of the therapeutic regimen may also have an
influence on missing appointments. When medication is pre-
scribed, patients sustain higher rates of appointment keeping
(Dodd, J. 1971). Craig (1976), onthe other hand, found a
more complex effect. Although a significantly large prOpor-
tion of non—psychotic patients receiving no medication
dropped out before the fourth session, one-half of the fifty
patients who remained in treatment also received no medica-
tion. The interpretation, according to Craig, might be that,
although the prescription of medication exerts an important
influence on the non-psychotic patient's early contacts with

a clinic, this influence diminishes in importance as a

36

relationship develops between patient and therapist. Such
an interpretation is congruent with observations of placebo
effect and is consistent with Frank's (1968) observation
that, in brief, psychotherapy, symptomatic improvement in
the first four visits tends to be nonspecific, but that im-
provement beyond the fourth week seems to depend on the em-
ergence of a therapeutic system embracing identifiable char—
acteristics of the patient, his/her therapist, and, perhaps,
other participants in his/her life. In view of findings
such as these, the appropriatness of the use of medication
or placebo as part of the initial therapeutic contract with
medically oriented patients may need to be reconsidered.
This is especially true since evidence suggests that overt
use of a placebo (i.e. with the patient's full knowledge

and consent) seems to have a powerful therapeutic effect
when patient expectations are congruent with its use (Park
1968).

Cost, duration, and side effects of therapy correlate
negatively with appointment keeping behavior (Rickels 1968),
Winkelman 1964). Compliance with medication regimen, how-
ever, does not necessarily correlate with clinic attendance.
More study is needed on this issue, because researchers focus
on clinic attenders rather than both attenders and non-atten-
ders (Hayes 1976).

Features of the Patient—Therapist Interaction

 

Very little data are available on the question of pa-

tient—therapist interaction as predictor of appointment

37

keeping behavior. Psychiatric literature has dealt with

this issue to some extent. Patient satisfaction with the
visit, therapist or clinic, not surprisingly, does correlate
with lower appointment failure rates (Alpert 1964, Haynes
1976), and specific attempts at patient education can im-
prove failure rates (Glogow 1970). The latter finding,
however, has not been consistent, and it may be that sub-
groups of patients respond differently to educational efforts
(Tagliacozzo 1974).

Feister (1975) investigated the joint interaction of
patient input (demographic and pre—therapy expectations),
therapist input (demographic), and therapy process (patient
perspective) as related to the outcome of early psychotherapy
termination. He found evidence that the process of dropping
out operates differently at separate community mental health
centers. Even though no overall differences in dropout
rates were observed at the two centers, the existence of
different therapist input, patient input, and therapy process
variables indicates the significance of the setting in ex-
plaining the droput phenomenon. The two settings examined
were a hospital-based center, and a community mental health
center. Feister (1975) suggests the difference in therapist
input and therapy process found at these two settings are
much more important than patient input in understanding
patient dropout behavior. He found, for example, that the
community mental health center consisted of state workers

who were older, more experienced, upper-class background

38

and with traditional training.

In regard to satisfaction, older adolescents and fe-
males were found to express the greatest satisfaction (Litt
1984). This may reflect the fact that contraceptive patients
fall into this category. Patients who make their own appoint—
ments are more likely to both keep their appointment and be
satisfied with the services received (Litt 1984). Irwin, et.
al (1981) found, however, that adolescents whose parents made
the appointment are exceptions to Litt's finding, as in this
case, the adolescent is more likely to keep the appointment.
Furthermore, adolescents who had a good self-image, and
those scoring high on the test of personal freedom, were
more satisfied with their care than those lacking these
characteristics. Young males showed the lowest satisfaction
in health services received. Satisfaction is assocated with
length of time for waiting, thoroughness of treatment and
explanation of medical problems (Litt 1984).

The types of services to be offered at a visit, and
the patient's expectations of a visit may both be signifi—
cant factors (Baekeland 1975). A successful method to
reduce broken appointments in psychiatry clinic utilized a
patient interview before the onset of actual therapy to
communicate to the patient realistic expectations of the
process and outcome of his/her therapy (Hoehn-Sarie 1964).
Barry (1984) made use of this find by designing a study
in which, prior to induction into the health care system,

a patient is exposed to an educational process. The

39

purpose of the education session was to prepare the patient,
by way of altering erroneous and unrealistic expectations
which, if left unaddressed, can lead to patient frustration
and subsequent noncompliance. Four hundred sixty patients
were each randomly assigned to one of three groups. One
group viewed a 20 minute videotape introducing the clinic,
its staff, and services, and how to utilize the staff during
and outside office hours. A second experimental group re-
ceived the same information in pamphlet form. The control
group received no information about clinic functions, ex-
cept that which was requested by the patient. Eleven months
after onset of the study, all patient charts were reviewed.
Compared with both the no-treatment control group, and the
pamphlet experimental group significantly fewer new patients
viewing the induction videotape missed subsequent scheduled
appointments (P .025). This same group had a significantly
lower number of missed appointments during the study period
(P .05).

Access to the Facility Features

 

To some extent, factors related to access have been
studied. Distance from the clinic has not, contrary to
expectations, been found to affect appointment keeping
(Feister 1974, Bigger 1976). In 1962, Hoenig, et al,
found almost one-fifth of new outpatient appointments in
the psychiatric department of a Manchester hospital were
not attended. Hoenig looked at the distance frOm the

clinic as a possible influencing variable. His findings

40

are as follows in Table 3.

TABLE 3

Distance of Patient From Clinic

Distance in

 

 

 

Miles Psychiatric Patients Neurological
Attenders Non-Attenders Non-Attenders
Less than 1 16 11% 17 11%
1-3 51 34% 61 41%
4-5 29 19% 31 21% 43 45%
6-8 17 11% 12 8%
8-9 10 7% 9 6%
9+ 27 18% 20 13% 52 55%

This table shows there is remarkably little difference
between two psychiatric groups in the distance of the pa-
tient's home from the clinic. However, when the neurological
non-attenders were compared with the psychiatric non-attend-
ers, it was found that, in the psychiatric group, only 13 per-
cent lived more than 9 miles away, while among the neurologi-
cal group, 55 percent did so. Availability of transportation
has not been carefully studied, but is occasionally cited by
patients in surveys of failed appointments (Alpert 1964).

The presence of a telephone in the home, through which a
patient has easier access to the providers, has shown some
predictive value for clinic attendance even without telephon-
ed reminder systems (Hansen 1953, Shepard 1976, White 1967).

Lack of a telephone can also be a factor in a patient's

41

failure to cancel appointments he/she cannot keep (Walsh
1967).

Features of Facility and Administrative Process

 

Most of the traditional approaches to the problem of
appointment keeping behavior have ignored organizational
factors that may be implicated in differentially high broken
appointment rates. Rather, the conclusion that low—income
patients break appointments more frequently than others has
become accepted in such a way that, in many articles dealing
with care delivery to lower income patients, an implicit
assumption is made that the broken appointment rate will be
high. This alleged tendency of low-income patients to break
appointments is explained in a variety of ways including
ethnic background, low education levels, cultural barriers,
low harmony in family relationship, social disorganization,
and other factors related to urban living. Too few re—
searchers have carefullyanalyzed the relationship between
the type of health care most often delivered to these low-
income population groups, and the facility's impaCt upon
broken appointments. The most traditional appointment me-
chanism for the large hospital-based outpatient clinic that
serves a low-income population is the block method, which
acts as a further confounding variable of higher broken ap-
pointment rates, and which only serves to further support
the belief related to the appointment keeping behavior of
low-income patients.

Features of the facility and its administrative and

42

scheduling procedures have an enormous impact on attendance
rates and are important because they can be manipulated by
providers to improve attendance. Studies often omit facili-
ty and administrative process features when reporting an
appointment failure, but facility related features may be,
at least, if not more, important than patient factors.
Clinic waiting time, for example, is important in studies
where it was examined (Badgley 1961, Finnirty 1973). Wait-
ing time is a function of the scheduling system, of patient
lateness for appointments and physician lateness. Wolkon
(1972) suggests that longer waiting times frustrated a
client's dependency and affiliation needs. 0n the other
hand, it may be that clients with longer waiting times be-
fore their first clinic appointments resolved their crisis
on their own. More research is needed to clarify what hap-
pens during waiting time.

The issue of appointment scheduling to minimize the
occurrence of missed appointments is almost entirely neg-
lected in the health care delivery and management litera-
ture. The status quo is one of individualization based
on local patient habits and physician or business manage-
ment personnel preference. More research is needed to
devise some broad guidelines that can be effectively in-
corporated into a wide variety of practice types.

One of the few research efforts in this area was un-
dertaken by Shonick and Klein (1977). They proposed ap-

pointing an expected number of patients for a given clinic

43

session. The expected number is calculated from the sum of
probability of 0.80 for keeping the appointment; then these
five would total four patients expected to show (5 x 0.80).
For an additional group of patients each with a probability
of showing of 0.90, seven additional appointments could be
made to yield an additional 6.3 expected patients (7 x 0.90)
resulting in a total of 10.3 expected kept appointments out
of a total twelve actually scheduled. The key to this system
is the capability to establish, with some accuracy, a pro-
bability that any given patient will fail to keep an appoint-
ment. Whether or not this can be done in a rapid, cost ef-
fective manner remains to be additionally established.

The above method essentially overbooks appointments to
compensate for the probable number of missed appointments
at a time when other patients are predicted to fail to show.
Simply overappointing by the same gross percentage as that
of expected missed appointments results in patient visits

at a time when there is no appointment vacated by a failure,

thus considerably lengthening patient waiting time. For
overappointing to be most effective, it must be scheduled
at those times during which the highest probability of a
missed appointment occurs.

Similarly, the practice of allowing for walk—in patients
or patients without appointments at the same rate as missed
appointments is inefficient. Walk-in appointments and

missed appointments cannot and, generally, do not occur at

44

the same time. Thus, this practice leads to excess patient
waiting times on some days and idle physcian time on others.

Another scheduling variation, used in recent years, is
the "modified wave system". This system schedules several
patients at the beginning of each hour and half—hour, then
leaves frequent gaps during the ensuing 30 to 60 minutes for
walk-in patients, or for patient visits that are longer than
expected. In addition, if any given patient fails to keep
anzmpointment, there is always another patient present to
reduce idle time. For example, if one is seeing patients
every 10 minutes, then two patients could be scheduled at
9:00, none at 9:10, and one patient at 9:20. Some have
critized this system because of the built-in waiting time
for patients scheduled at the beginning of the hour. How-
ever, if excessive initial overbooking is avoided, waiting
time should not be inordinate and considerable operating
flexibility is achieved.

A last suggestion is the elimination of automatic
reappointing of patients who have failed to keep an appoint-
ment. This practice is common in university ambulatory
care clinics and, as shown by Hofmann and Rockart (1969),
leads to inordinately high rates of broken appointments.

If a patient misses an appointment, notification of the
failure and a request to reschedule a visit, if desired,
should be sent. The responsibility for rescheduling is
then placed upon the patient with the hope that the patient-

initiated appointment is more likely to be kept.

45

Practical and theoretical considerations in scheduling
were reviewed also by Rockart and Hofmann (1961) who demon-
strated the effectiveness of individual vs. "block" schedu—
ling in reducing broken appointment rates. Both physician
and patient were found to be more punctual under the individ-
ual appointment system. Reduction of waiting time by this
method has been demonstrated by a number of researchers
(Johnson 1968, Rosenblut 1972). Where this system is used
in clinics serving low-income families, broken appointment
rates compare favorably with rates reported from other set-
tings, and the researchers suggest that, in many cases, the
style of medical care delivery may prove to be more important
than demographic variables in influencing appointment fail-
ure rates (Finnirty 1973, Hertz 1977). Referral source has
some effect on attendance with referrals from emergency
rooms showing lower appointment keeping rates, and those
from a specific referring provider showing higher rates
(Hofmann 1969, Shah 1977). Time between scheduling and the
appointment has an influence. In general, the longer the
interval, the greater the number of broken appointments
(Hagerman 1978, Nazarian 1974). This appears to be especial-
ly true with regard to follow-up appointments after mass
health screening (Glogow 1973). On the other hand, patients
with very frequent appointments are more likely to miss at
least some of them (Delk 1975). The use of home visits or
personal contact has been successfully used to minimize

broken appointments (Curry 1968, Hildebrandt 1975). In

46

at least some settings, the provision of neighborhood
clinics in familiar surroundings with a highly involved
staff has resulted in dramatic reduction of broken appoint-
ments in comparison with previously used central facility
(Curry 1968).

One of the most studied administrative procedures is
that of reminding the patient, either by mail or phone, as
a method of reducing broken appointments. Reminding clients
has been most successful in reducing no shows. Reductions
in broken appointment rates on the order of 20 percent have
been reported (Schroeder 1973, Nazarian 1974, Levy 1977).
Mailed reminders appear to be generally effective as tele-
phone reminders. Shepard and Mosely (1976) provided cost
estimates in their evaluation which showed the mailed notice
to be less expensive than telephone reminders. Levy and
Claravall (1977) found telephone reminders to be effective
when the interval between appointments was greater than two
weeks, but not if the interval was less than two weeks. In
contrast, the use of retrospective reminders, i.e., notifi-
cation of patients after missed appointments to ascertain
their reasons for absence and to offer a new appointment,
has been ineffective (Hurtado 1973).

Features of the Environment

 

Weather (excluding extremely severe conditions such as
heavy snowfall) probably has very little influence on ap-

pointment keeping (Badgley 1961), Hurtado 1973). Family

47

size shows a fairly consistent relationship with appoint-
ments with large families having more broken appointments
(Alpert 1964), Badgley 1961). The presence of small child-
ren may be especially relevant in explaining broken appoint-
ments (Alpert 1964). The time of day and day of the week
have a minor, if any, role in appointment failure rates
(Gates 1976, Oppenheim 1979). The expectation of friends
may exert an influence (Gray 1966), however, family stabil+
ity as recorded by a variety of measures appears much more
to influence appointment keeping rates (Alpert 1964, Stine
1968). Stability measures used included such variables as
marital status, length of time on a job, frequency of moving,
and the like. The less socially stable 3 patient was, the
more likely he/she was to miss an appointment. Occupation-
al stability appeared to be a more consistent predictor of
missing appointments for treatment of alcohol and drug ad-
diction than in other kinds of patients. Occupational and
residential stability appear to carry more weight than does
marital stability. The poor ability of the unstable person
or patient to form ties with others (he/she tends to be
single, separated, or divorced) is reflected in the weakness
of his/her attachment to his/her therapist, and, thus, such
a person is more inclined to miss appointments (Quatrone
1973). Further investigation is needed to identify cause

of the instability or isolation. Is it the result of per-
sonality characteristics (e.g., those associated often with

marital disruption), or is it because of lower socio-economic

48

status. It may, in fact, be a combination of these vari-
ables.

While we direct our attention to the issue of account-
ing for features that are related to appointment failures
for a balanced view of the problem, we should be aware that
it cannot be assumed that the dropout gets no benefit from
treatment. Similarly, it cannot even be assumed that the
patient who does not show up for treatment is a total loss.
Many patients on a waiting list for psychiatric treatment
improve without specific treatment. Endicott and Endicott
(1963) found that such patients are better educated and
rate lower on depression and hostility and higher on self-
esteem, adaptation to reality, defense, and ego strengths
than those who fail to improve.

Despite spontaneous improvement and entry or re-entry
into treatment, on the average, thedropout seems to do worse
than his/her counterpart who perseveres in treatments (Ren-
ton 1963). Elopers from a general psychiatric hospital were
found to be unimproved at a six months follow-up while a
control group of patients discharged with medical advice
fared significantly better (Pam 1973). Similarly, among
alcoholics, both in inpatient and outpatient settings,
dropouts have a worse outcome than program completers (Bowen
1968, Tomsovic 1970). The situation is far worse for heroin
addicts. Patients who complete hospital detoxification are
about three times as likely to be drug free a year later as

those who do not finish it (Raynes 1972).

49

Research ontfluunawho miss appointments or dropout of
treatment has been primarily concerned with psychiatric pro-
blems, alcoholism, or drug abuse. Further, research is ne-
cessary to establish the consequences of missed appointments
and dropout of treatment for other health problems.

Specific Hypotheses

 

Based on findings from studies reviewed, and the many
years of experience in the field of health screening on the
part of this researcher, there should be significant positive
correlation between appointment keeping behavior and the
following variables:

1. Appointment keeping behavior will have a signficant
positive correlation with the patient's age, educa-
tion, socio-economic status, the occupational status
of his/her father and income, but no significant re-
lationship with gender, race, religion, place of dwel-
ling, language, or place of birth.

Furthermore, the following social behaviors and
perceptions of the patient will have a significant
positive correlation with appointment keeping, namely,
belief of disease as serious; belief of personal sus-
ceptibility to specific diseases; belief of suscepti-
bility to disease in general; belief regarding effi-
cacy of treatment; belief regarding belief in thera-
pist's ability; knowledge of disease and therapy; level
of intelligence; previous use of facility; previous

dropout; general attitude toward health professionals;

50

drug dependence; and ownership of a thermometer. There
will be a significant negative correlation between ap-
pointment keeping and belief regarding therapy as pain-
ful, belief in supernatural causes of disease, and being
head of the household.

There will be a significant positive correlation between
appointment keeping, and the following characteristics
of the provider of care, namely, age, years from comple-
tion of training, continuity of service, attitude/behavior,
and patient load. There will be no relationship with the
race or gender of the provider.

There will be a significant positive correlation between
appointment keeping, and the following characteristics
of the disease and diagnosis, severity, duration, pre-
vious therapy, symptoms, clinical improvement, urgency
of appointment, consequences of missed appointment,

and functional impairment caused by the disease.

There will be a significant positive correlation between
appointment keeping, and the following characteristics
of the therapeutic regimen, namely, whether or not medi—
cation is prescribed, dosage, and prescription of a
diet. There will be a significant negative correlation
between appointment keeping, and the duration of the
treatment, cost, side effects, and degree of behavioral
change required.

There will be a significant positive correlation between

51

appointment keeping, and the following characteristics
of patient-therapist interaction, namely, staisfaction
with visit to clinic, expectations met, comprehensive-
ness of services offered during the visit, patient
education, and level of supervision.

There will be a significant positive correlation between
appointment keeping, and the following characteristics
of access to the health care facility, namely, availabil-
ity of transportation and the telephone in the home, but
a negative correlation with distance of home from the
clinic, and indirect costs.

There will be a significant positive correlation between
appointment keeping, and the following characteristics
of the facility/administrative process, namely, clinic
waiting time, referral sources, referral to specific
provider vs. referral to clinic, time for scheduling
appointment, patient-staff ratio, scheduling method
(block vs. individual time), availability of parking,
frequency of appointments, use of prospective reminders,
and restrospective reminder, but a negative correlation
with scheduling errors.

There will be a significant positive correlation between
appointment keeping, and the following characteristics
of the environment, namely, influence of family and
friends, family stability, family size, time of day of

appointment, day of week of appointment, illness in

52

family, performance of children in school, and presence
of both parents in the home, but a negative correlation

with weather and time on public assistance.

III. METHODOLOGY

Sampling

Selection of a Sample.

The State of Michigan maintains a Client Information
System (CIS) on all families and individuals enrolled in
public assistance programs. Current policy on periodicity
for health screening recommends that each child receive a
comprehensive assessment every two years, or every six
months if a child is under one year of age. In the CIS
master file, a record is kept on birth date and health
screening history of approximately one half million child-
ren and youths below twenty-one years of age.

Each month a list of children to be screened is sent
to the local screening agencies throughout the State. The
computer is programmed with the following criteria for the
selection of the names that appear on the lists:

— Select any recipient never having been screened and
belonging to a new CIS case.

- Select any recipient never having been screened and
currently at a recommended screening age.

- Select any recipient less than one year old and not
having a screening during the past six months.

- Select any recipient less than one year old and having
their last screening six months ago.

- Select any recipient not having a screening in the
past two years.

— Select any recipient not having a screening in the

past two years and is currently at a recommended
age for screening.

53

54

- Select any recipient older than five months, has never
been screened, and was notified for a screening one
year ago.

— Select any recipient less than one year old at last
screening, the last screening was one year or more
in the past, and was last notified for a screening
one year ago.

- Select any recipient one year or more at the last
screening, the last screening was two years ago,
and one year has passed since the last notification
for screening.

From this list, an outreach worker in each of five regions
in the State randomly selected seventy families to be includ—
ed in this study for a statewide sample total of 350 parti-
cipants. If a family was found to be ineligible for screen-
ing, then another name was randomly drawn. There are a
variety of reasons why a family that may have fallen into
the sample is ineligible. The computer list is generated
in the middle of the month, and many families listed lose
their eligibility at the end of the month. Case load changes
on an average of 30 percent each month so that a sizeable
number become ineligible for Medicaid benefits. Screening
appointments cannot be arranged quickly enough to guarantee
service before the end of the month for all the families
listed. Thus, about 30 percent contacted are, in fact, not
eligible for screening.

It is also not uncommon that a family is listed as due
for screening when, in fact, the family was screened a few
months before the date listed for contacting. This may be

due to leaving a public assistance program and then return-

ing to the rolls. The computer automatically prints out

55

new cases without regard for screening history. Further-
more, a family may be listed as eligible because the health
screening report never was entered into the computerized
case file of the family. In other cases, a family may have
already left the county so that contact was impossible. Low-
income families have a history of mobility. In some cases,
the address given in the client's CIS file is that of a
friend or relative who receives and transmits to the family
the monthly check. In some cases, it is impossible to find
or contact, directly, the family listed on the monthly com—
puter printout. These are examples where it becomes neces—
sary to randomly select another family to reach the quota
of seventy clients per region.

The sample was drawn from a period so as not to include
either August or December. Statewide statistics indicate
that there is virtually no fluctuation in the ratio of ap-
pointments not kept from month to month except in August
and December. During these two months, for the past ten
years, notable fluctuation in the ratio of appointment keep-
ing occurs. In August, families are much more likely to
keep their appointments probably because of their concern
in preparing children for school. Michigan law requires
that immunizations be up to date before a child may attend
school. In December, families are less likely to keep their
appointment. This is probably due to the increased activit-
ies occurring during this holiday season. Excluding these

two months, broken appointments average almost 37 percent

56

of appointments made with a fluctuation of a plus or minus
three percentage points.

Training of Interviewers

 

After an agreement was received from the health officers
in each of the agencies chosen, this researcher met with and
trained surveyors in-each region. The surveyors were intro-
duced to the project by presentation of a general overview
of the study as follows:

Persons who make appointments and fail to keep
them cause disruption to the delivery of health
services and sometimes damage to themselves be-
cause of untreated health problems. The disrup-
tion to health services is costly, because time
set aside for a patient by a trained health pro—
fessional is not effectively used. In the EPSDT
program, a missed appointment translates into a
wasted $70-$80. In making an appointment, it is
important for the outreach worker to know whether
or not a person is going to show for screening.
Understanding factors that point to a successful
scheduling outcome is the goal of this survey.

In this study, a very small sample of the thou-
sands of families on public assistance will be
surveyed to obtain insight in the thinking of
the group as a whole. Furthermore, information
gained will be important to administrators in
arranging health services to achieve maximum
effect and efficiency. To achieve these goals,
it is important for the surveyors to follow
specific guidelines, so that the opinions of

the persons surveyed come through with as little
static (caused by the surveyor) as possible, and
the information gained is genuinely useful for
future planning of health services.

Following the introduction, the surveyors received spe-
cific instructions about procedures for sampling, record
management, scheduling, and questionnaire administration.

A practicum on administration of the questionnaire was held

among the surveyors to insure consistency.

57

Sampling:

Each month, the surveyors received duplicate lists of
families to be contacted for screening for the given month.
The surveyors took one copy, cut it so that one family ap-
pears on one slip of paper, placed the slips into a box,
and, finally, drew out seventy names. This list became the
sample for the study. The surveyors drew more names ac—
cording to the criteria noted above, under "Selection of
a Sample", until the five regions in the State had a valid
sample of seventy names. A letter (Appendix A) was sent
inviting the family to participate in the health screening,
along with a Client Questionnaire (Appendix B) to be filled
out. Included in the letter was a brochure explaining
health screening. At least three days after the letter was
sent, the surveyors called the client to schedule the family
for a health screening and to administer the Client's Opinion
Survey (Appendix C). When the Opinion Survey was completed,
the surveyor asked for the numbers checked or circled by the
client on the Client Questionnaire that had been sent along
with the letter inviting the client to participate in the
screening. Finally, the surveyors filled out the Medical
Outreach Worker Questionnaire (Appendix D). The surveyors
then highlighted the name of the surveyed family on the
screening schedule so that, depending on the response re-
ceived, the family was placed into one of four groups,
namely:

1. Those who refused to participate.

58

2. Those who accepted an appointment and did not keep it
without prior notice.

3. Those who accepted an appointment and called to cancel
or change it.

4. Those who accepted an appointment and kept it.
Scheduling Procedures:

Scheduling procedures took place before any question-
naires were filled out. Attention of the client was called
to the brochure on health screening contained in the letter.
The series of tests to be given were explained, and the pur-
pose of the screening was identified as an effort to detect
health problems in children so that health care can be given
when a health problem was still in its early stages. The
client was then asked whether or not he/she would like to
schedule the children for a screening. If the client declin-
ed to do so, the scheduling procedures ended. If the client
was willing, then the client was asked what day would be
convenient. When the client selected a day, he/she was
asked whether a morning or afternoon appointment would be
preferred. A schedule was made according to the response
received. The client was then offered transportation, and
it was arranged if requested. Whether or not the family
signed up for screening, the interviewers proceeded to the
administration of the questionnaire.

Questionnaire Administration:

The surveyors used the letter, previously sent as a

guide, and reviewed with the client the salient points

contained in the letter, namely:

59

- An opinion survey is being conducted Of a few families
on their feelings about health care services.

- The purpose of the survey is to find ways to improve
health services to your children.

- It will take about ten minutes for the interview.
After eliciting from the client his/her willingness to
cooperate, the explanation continued:

— The survey is about free health screening available to
your children.

- Your answers to the questions on the survey will be con-
fidential, and your name will not be put on the question-
naire.

- Your answers will not be shared with, and have nothing
to do with, the Department Of Social Services, and they
will not affect your eligibility for assistance or Medi-
caid in any way.

- The information given is for statistical purposes only.

- This effort is an attempt to hear the client side of
health services set up for Medicaid recipients like
yourself.

The surveyors then explained to the client that he/she
will make a series of statements. TO each, the client was
asked to respond with his/her Opinion as "Always", "Often",
"Sometimes", "Rarely", and "Never", as the case may be.

The interviewer read each statement carefully to the client

and repeated it, if requested to do so. The interviewer

avoided rephrasing, expanding, or commenting on statements
read to the client. If the client was puzzled, the problem
was Often solved by rereading the question more slowly.

Answers given by the client were check Off by the interview-

er.

At the conclusion of the opinion survey, the outreach

60

worker asked the client to read Off the numbers Of answers
checked or circled in the Client Questionnaire (Appendix B).
The worker waited for the client to finish if this question-
naire was not completed, and then took down the responses.
This ended the interviewing process.

A second visit was made by this researcher to all re-
gions while the data was being collected to insure conform-
ity tO instructions. Surveyors raised questions about fine
detail of the instructions indicating every effort was being
made to follow procedures as previously outlined. As an
example, in Delta County, one of the families refused to
participate, in which case the surveyor drew another name
to complete the requested sample of seventy.

Scope of the Study:

Although the EPSDT is a nationwide program, this study
focused only on Michigan. The population of this state is
concentrated in the Southeast portion, as nearly half Of
all persons in Michigan live in the counties located in the
Southeast section of the State, and more than 30 percent
live in urban settings. As one moves Northwest across the
State, the counties become, generally, more and more rural.

For the purpose Of managing health related programs
in the 45 health jurisdictions, the Michigan Department Of
Public Health divided the state into five regions. All
counties in the Upper Peninsula are included in the First
Region. The Second Region includes thirty-five counties

located in the northeast area of the Lower Peninsula.

61

Eighteen counties, beginning with Manistee and moving south
along the west border, constitute the Third Region. The
Fourth Region includes the counties of Washtenaw, Wayne,
and Monroe. The remaining counties, Shiawassee, Genesee,
Livingston, Oakland, Macomb, and St. Clair make up the
Fifth Region. In order to reflect a statewide scope, a
public health agency from each of the Regions was selected
to be included in this study. The selection of each agency
was based on which jurisdiction was closest to the mean in
terms Of the percentage of eligible families participating
in the health screening. In Region 11, Grand Traverse and
Alpena health jurisdictions had the same level Of partici-
pation. In a random drawing, Grand Traverse was picked.
The health jurisdictions chosen in each Region are as

follows:

Region I — Delta Menominee

Region II - Grand Traverse/Lellanau/Benzie
Region III - Barry-Eaton

Region IV - Monroe

Region V - St. Clair

Type of Participants:
The study includes only low-income families in Michigan
that have been certified as eligible for Medicaid. The de-

scription of the population eligible to receive EPSDT screen-

ing is best taken from the Michigan Department of Social Ser-
vices guidelines for determining the Medicaid eligibility.
The salient features of that determination process are as

follows:

62

l. The family resides in Michigan. This requirement is
taken to mean the family lives within the state and,
at least, one member is looking for work or the family
intends to remain indefintely. If the family leaves
Michigan, verification Of intent to return is necessary.
Institutionalized persons automatically meet the resi-
dency requirements.

2. The client must be a citizen Of the United States.
Under this provision, tourists, visitors, or persons
on a student visa are not qualified.

3. For the EPSDT program within Medicaid, the client must
be below 21 years of age.

4. For the EPSDT program within Medicaid, deprivation may
be grounds for eligibility. Deprived here refers to
lack Of parental care or support through absence, death,
incapacity, or unemployment of a legal parent.

5. All children receiving eligibility for EPSDT must be
living with a specified relative.

6. To be eligible for EPSDT, a child must be living in a
"suitable home". Unsuitability is determined by the
juvenile court.

7. Young adults, ages 18 through 20 years, must be a full-
time student in high school or in an equivalent level
Of vocational or technical training, and can be reason-
ably expected tO complete high school or vocational/
technical school before reaching age 19. A person is
considered still attending school if absence is due to
illness, convalescence, or family emergency.

The fundamental basis for determining eligibility, how-
ever, is income. Definition Of the eligible population for
EPSDT screening is usually those who fall below the poverty
line. The current designation of poverty level in the United
States refers to a non-farm family of four members with an
income less than 3,600 dollars with a 500 increment for each
additional member. Under this criteria, about 10 percent

of the population and 14 percent of the children in the

United States are below the poverty level (White House

63

Conference on Children, Washington 1970).
Variables Included in This Study:

As mentioned in the Introduction, researchers have found
there are a myriad of factors that influence the behavior of
a person regarding the keeping of an appointment for health
services. Using Haynes (1976) classification, these factors
are grouped into general categories reflecting eight features.
The hypothesis Of this study is that all of these selected
features will have a significant correlation with a client's
appointment keeping behavior for a comprehensive health
screening in the EPSDT program. The eight behavior features
in this study together with variables to be measured under
each are as follows:

1. Patient Features:
Patient features include both demographic and social
behavioral variables.

A. Demographic Variables:

Age Race

Sex Religion
Education Urban vs. Rural
Socio-economic Status Language

Occupational Status of Father Place Of Birth
Income
B. Social Behavior and Perception Variables:
Belief of Diease as Serious
Belief Of Personal Susceptibility to Specific Diseases

Belief of Susceptibility to Disease in General

64

Belief Regarding Efficacy of Treatment
Belief in Supernatural Causes

Belief in Therapist's Ability

Knowledge of Disease and Therapy
Intelligence

Previous Use Of Facility

Previous Dropout

General Attitude Toward Health Professionals
Drug Dependence

Head of Household

Ownership of a Thermometer (proxy for health
"concern")

Provider Features:

Age

Race

Years From Completion Of Training
Gender

Attitudes and Behavior

Patient Load

Features of Disease:

Diagnosis

Severity

Duration

Previous Therapy

Symptoms ("Complaint" vs "Lanthanic")
Clinical Improvement

Urgency Of Appointment

65

Consequences of Missed Appointment
Functional Impairment

Features of Therapeutic Regimen:

Medication

Whether or Not Medication Prescribed
Duration

Costs

Side Effects

Dosage

Prescription of Diet

Degree of Behavioral Change Required
Features Of Patient-Therapist Interaction:
Patients Satisfaction With Visit to Clinic
Patients Expectations Met

Comprehensiveness of Services Offered During the Visit
Patient Education

Level Of Supervision

Access Features:

Distance From Clinic

Indirect Costs

Availability of Transportation

Telephone in the Home
Features of Facility/Administrative Process:
Clinic Waiting Time

Referral Source
Referral to Specific Provider vs. Referral to Clinic

Time From Scheduling to Appointment

66

Patient-staff Ratio
Scheduling Method (Block vs. Individual Time)
Availability of Parking
Use of Prospective Reminders
Retrospective Reminders
Scheduling Errors

8. Features of Environment:
Weather
Influence of Family and Friends
Family Stability
Family Size
Time of Day of Appointment
Day of Week Of Appointment
Illness in Family
Performance of Children in School
Presence of Both Parents in the Home
Time on Public Assistance

Scoring and Field Testing of Survey Instrument

 

Scoring:

The survey instrument was designed to assign the high—
est score (5) to reflect the most favorable response for an
item. As an example, if a respondent indicated transporta-
tion is always available, a.scoreof five (5) would be as-
signed. If, on the other hand, a respondent indicated
that no transportation is available, a score of one (1)
would be assigned. The outcomes for program participation

are scored from the lowest level of refusal to accept

67

health screening to the highest score for those who accept
an appointment and keep it. Participants in this study are
divided into fourgnoups:

1. Those who refuse to participate.

2. Those who accept an appointment and do not keep it
without previous notification.

3. Those who accept an appointment and call into cancel
or change it.

4. Those who accept an appointment and keep it.
Field Testing of Survey Instrument:

Prior to conducting the research, a field testing of
the survey instrument was conducted. Twenty-five interviews
were conducted in which two persons scored the responses.
The scores Obtained indicated a rater reliability Of 98 per—
cent. Some questions, following the field test, were shor-
tened to increase sharpness and clarity. An instrument re-
liability was tested by administering the questionnaire to
twenty-five individuals and re—administering it in an hour
or more later to see if response to the questions had changed.
The reliability Of the instrument was found to be 97 percent.

Method Of Analysis

 

The data were cluster analyzed according to the Tryon
and Bailey method (1970). This analysis serves to describe
general interrelationships among the variables included in
the study. A cluster analysis requires a number Of varia-
bles, the correlations (Pearson product moment) among them,
and the clusters. The nature Of the clusters is immaterial.

The variables ought to be in standard form, i.e. with a

68

mean of 0 and a standard deviation of 1 since the results
are in terms of correlations. While refusing or accepting
an appointment, changing it or not changing it, or keeping
or not keeping an appointment are dichotomous variables,
there is an underlying continuous scale with a low score
(1) of referring an appointment to a high score (4) of
accepting and keeping an appointment. A tetrachordal
correlation, consequently, can be computed and the cor-
relation is an estimate Of a Pearson product moment cor—

relation.

IV. RESULTS

The results Of this study will be presented in two
sections. The first section will consist of looking at
the various items used to measure the significance Of
eight features hypothesized as related to appointment
keeping behavior. The eight features divide the multiple
variables affecting appointment keeping behavior into eight
categories Of variables mentioned earlier, namely, those
related to the patient, providers of care, the disease,
the therapeutic regimen for treating the disease, patient-
therapist interaction, access to the clinic, facility/ad—
ministrative process, and the environment. The second
section includes the clusters formed by the measures used
in testing the hypotheses listed below.

Test of Hypotheses

 

The first hypothesis focused on patient related varia-
bles in terms of their correlations with appointment keeping
behavior. The first hypothesis stated that appointment keep-
ing behavior has a significant positive correlation with the
patient's demographic characteriStics Of age, education,
socio-economic status, occupational status Of father and
income, and that no snigificant relationship with gender,
race, religion, place of dwelling, language, or place of
birth would be found. The following characteristics Of the

patient will have a significant positive correlation with

69

70

appointment keeping, namely, belief Of disease as serious,
belief Of personal susceptibility to specific diseases,
belief of susceptibility to disease in general, belief
in thereapist's ability, knowledge Of disease and therapy,
intelligence, previous use of facility, previous dropout,
general attitude toward health professionals, drug depend-
ence, and ownership Of a thermometer. There will be a
significant negative correlation between appointment keep-
ing and belief regarding therapy as painful, belief in super-
natural causes Of disease, and being head of a household.
The test Of hypothesis regarding demographic character-
istics and the correlation with appointment keeping is dis-
played in Table 4. The distribution, in percent, Of parti-
cipant's responses to each demographic item is given in

Table 5.

It was assumed that the Older a parent was the more
responsibility would be demonstrated in looking out for the
preventive health care needs Of their children insofar as
keeping medical appointments. However, Table 4 shows that
education and socio-economic status, that is, the highest
level of income reached, the level of skill attained by the
father, the level of income for 1984, had no significant
relationship with appointment keeping behavior.

Since only 7 percent Of the respondents were male and
3 percent Black, gender, and race are not adequately sampled
in this study. On the other hand, the hypothesis that

stated no significant relation exists between appointment

71

TABLE 4

Test of Patient Demographic
Characteristics and Appointment Keeping

Correlations With

Patient Related Measures Appointment Keeping
Age -.01
Sex .04
Level of Education -.07
Socio-economic Status .03
Occupation State Of Father .03
Level Of Income for 1984 —.06
Race — White —.00

Black .01

Other .03

Religion - Protestant -.14

Catholic .13

Other Religion .07

NO Religion -.05

Living in an Urban vs Rural Area -.00
Speaks Another Language Other Than

English .02

Born Outside of Michigan .01

N=336

keeping and gender, race, religion, place Of dwelling as
urban vs. rural, ability to speak a language other than
English, and place of birth is supported by findings in
this study.

Table 4 shows that demographic characteristics Of
the patient are of little or no help in prediciting suc—

cessful outcomes in appointment scheduling. There is a

 

50gpercent change that, out of eleven demographic measures
used, at least one would have been significant by chance

(Fairweather 1978). The correlations with these variables
are so low, however, that not one significant association

occurred.

72

TABLE 5

Distribution of Responses on Each
Demographic Measure by Percent

 

Demographic Measures

Response Distribution

 

Age (in years)
Gender

Level of Education

Socioeconomic Status
(Highest Yearly
Income)

Occupation of Father

Level of Income
for 1984

Race

Religion

Urban Versus Rural
Living

Speaks Foreign
Language

Born Outside of
Michigan

N-336

13-17 18-24 25-34 35-44 45+
3 33 43 16 6
Male Female
7 93
Eighth Some High Some College
Grade High School Col- Graduate
or Less School Grad. lege
6 27 44 18 5
Less
than 5,000 10,000 15,000
$5,000 9,999 14,999 19,999 20,000
35 32 14 5 10
Semi-
Unskilled skilled Skilled
19 33 46
Less
than 5,000 10,000 15,000
$5,000 9,000 14,999 19,999 20,000
44 37 11 2 3
White Black Other
89 3 8
Protestant Catholic Other No Reli-
gion
34 25 22 19
Within 1 Mile 1-5 5-10 11+
City From Miles Miles Miles
City
59 8 13 12 8
Yes No
12 88
Yes NO
76 24

73

TABLE 6

Test Of Patient Related Social Behavior or
Perception Variables and Appointment Keeping

Patient Social Behavior or Correlations With
Perception Related Measures Appointment Keeping
Belief Of Disease as Serious Threat .04

to Health

Belief Of Personal Susceptibility
to Specific Diseases .11

Belief of Susceptibility to Disease

in General .09
Belief Regarding Efficacy Of Treatment .02
Belief Regarding Therapy as Painful ‘ .04

Belief in Supernatural Causes of
Disease .07

Belief in Therapist's Ability to

Heal .02
Knowledge of Disease and Therapy -.01
Intelligence —.01
Previous Use Of Health Care Facility .10
Previous Dropout From Treatment Used

Previously .09
General Attitude Toward Health

Professionals .02
Drug Dependence —.04
Head of Household .03

Ownership of a Thermometer (proxy for
health "concern") -.03

N=336

Ommuz

 

 

m H 5 0H ow A:enoucou: nnHmO:
now mxonav nonoeoanone O HO ansmnoaao
OH m OH O OO OHonomnom mo Ommm
OH OH om a O mommmaxonm mOOH>mnm :H mm: CH wane

o H «N am mm mecmHnm mm mHchHmmom
Ionm nnHmmm Onmaoe meannnn< Hmnmcoo

we on HH H o mHmDOH>mnm
womb ncmanmonH aonm naomono maoH>onm
n s OH om we mnHHHOmm onmu nnHmom no em: mDOH>onm
m n mm ON NN ooeowHHHmncH owmno>< O>OO< mmm

NH an am NN an namnmgn can
Ommomna mo mwva3OcM mwmnm>< O>OO< mmm
n N we ON mH Hmom On mnHHHn< m.anOmnO£H CH MOHHom
oq OH ON O OH mmmomno mo mmmamo Hmnsnmcnmasm OH MOHHOm
mn en He m HH nsncnmn mm namnmzn manenammm nannmm
m H Hm Om mm ncoanmone mo mumunmmm wcHOnmwmm MOHHmm

m O OH on Ow Hmnmcou OH
OmmOmHa On mnHHHOHnaOOmOm mo OOHHom

as w w s om mamammna unnnumam
On hnHHHanamomsm Hmcomnmm mo meHOm

OH «H mm NH mH anmom On
nmonna muonnmm mm OmOOmHa mo OOHHom

nO>Oz nHmnmm mOanOeom cenwo m>m3H<

 

ncoonom up COHnsaHnanQ Omcoamom

monsmmoz OOHnOoonom no nOH>mnOm HmHOOm

 

ncoonmm zn onsmmoz connamonmm no nOH>mcom HmHOOm
comm co momcoammm mo OOHnannann

h MHm<H

75

The hypothesis regarding social behavior characteris-
tics Of the patient and the correlation with appointment
keeping is displayed in Table 6. The distribution, in per-
cent, Of participant's responses to each social behavior
measure is given in Table 7.

It was anticipated that parents who believed that they
or their children could easily become susceptible to diseases
in general, or to specific diseases (like heart trouble,
cancer, or diabetes), and who believed the consequences of
getting a disease would be serious would, most certainly,
be anxious to keep appointments for health servicest This
study found no such relationship, consequently, the original
hypothesis is not supported.

A patient's belief regarding the efficacy of treatment,
in general, or in a therapist's ability to heal did not cor-
relate positively with appointment keeping. NO relationship
whatever was found. Likewise, there was no significant cor-
relation with appointment keeping, and an above average in-
telligence or knowledge Of disease, and how to care for it.

Past practices of the patient are not helpful in pre-
dicting appointment keeping. The hypothesis that previous
use of a facility, previous keeping Of appointments, general
positive attitude toward health professionals, and prescrip—
tion of drugs in previous visits tO the doctor would have a
significant positive correlation with appointment keeping
was not at all supported. It was, also, assumed that parents

who owned a thermometer to check for fever in their children

76

would have a tendency toward preventive mode of behavior in
health concerns, generally. This hypothesis was also not
supported by the data.

A significant negative correlation was expected between
keeping appointments for health services and belief that
therapy would be painful, that sickness is a punishment from
God, and parents preoccupied with other duties as head of the.
household. This negative association was not found in this
data.

There were fifteen measures of social behavior charac-
teristics of the patient. Not even one was found to be sig-
nificant. This is surprising as one would expect one signi-
cant finding to occur by chance in just ten measures of a
feature (Fairweather 1978).

The second hypothesis Of this study projects a signi-
ficant positive correlation between appointment keeping and
certain characteristics of the health care provider, namely,
age, years since formal training was completed, attitude
toward the patient, and patient load. NO relationship was
projected for race or gender of the health care worker.

The hypothesis regarding provider characteristics and
the correlation with appointment keeping is displayed in
Table 8. The distribution, in percent, Of responses regard-
ing provider characteristics is given in Table 9.

Only one measure used to assess provider impact on
clients' appointment keeping behavior was significant.

The number of years a person has been on the job

77

TABLE 8

Test of Provider Related Characteristics
and Appointment Keeping

Provider Characteristics Correlation With
Measures Appointment Keeping
Age -.02
Race —.01
Years From Completion of Training -.01
Gender .04

Years Medical Worker Has Been on

the Job -.l6*
Attitude Toward Patient —.09
Number of Patients on Caseload -.01
N-336

*Significant at the .05 level of confidence

correlated negatively with appointment keeping. This finding
is exactly opposite to the original hypothesis. It may in—
dicate a burn—out condition that can result in providing
health care over the years to patients perceived to be un-
responsive.

A significant positive correlation between appointment
keeping and the provider's age, attitude toward the patient,
and patient load did not materialize. No significant cor-
relation was found in mese measures. The projection of no
significant relationship with race or gender, however, was
supported by the data. The measure is poor since 96 percent

of the providers were white and 99 percent were female.

78

Ommnz

 

NON NO NHN Noe NmH
+oom OONIOON mmHlomH OOHIQOH aono OmOHOmmo so mncmHnmm mo nonesz
No No NH NON NON
no>oz OHonmm moSHn conmo m%m3H< mnonmwamz Ouncoonom
Imeom Ocmez :H nOOHnmm Onmaoe Owannn<
Nmm NH
OHmEom OHmz nowcmw
NN No NON NOO NNm now
Ozn so comm mmm nmxnoz HOOHOOE mnmow
NOm No NH No NN
+m mum OIN NIH OOOH no
nmow Oco OOHOHmnH mo :OHnOHOEOu Eonm mnmow
Ne NOO
xOmHO Oanz Oomm
Nam NO NON NHO NO
+mm calms eelmm smimw «Numn Hansen cnv one
OOHnOOHnana omcommmm monsmmoz mOHannonomnmzo

mnmxnoz nommnnso HOOHOOZ

 

ncmonmm On monannOnOmnmzo OOnOHmm nOOH>onm
co momcoammm mo coHnannanQ

m MHm<H

79

TABLE 10

Test of Characteristics Of the Disease
and Appointment Keeping

Disease Characteristic Correlations With

Measures Appointment Keeping
Diagnosis Of the Disease .10
Severity of the Disease .11
Previous Therapy Received -.03

Symptoms of the Disease ("Complaint"

vs. "Lanthanic") -.09
Clinical Improvement With Treatment .08
Urgency of Appointment for Case .08

Seriousness of Consequences Of Missed

Appointment .05
Functional Impairment Caused by the
Disease .01
N=336

Out Of the eight measures to assess provider impact on
appointment keeping, only one was found to have minimum sig—
nificance. There is a 30 percent possibility that this
would occur simply by chance (Fairweather 1978). The find—
ing, therefore, is very weak.

In the third hypothesis of this study, it is suggested
that there is a significant positive correlation between
appointment keeping and the following characteristics of
the disease, namely, its diagnosis, severity, duration,
previous therapy, symptoms, clinical improvement, urgency
of appointment, consequences Of missed appointment, and

functional impairment caused by the disease. The

80

TABLE 11

Distribution of Responses to
Disease Characteristics by Percent

Disease Characteristic
Measures Response Distribution

Some-
Always Often times Rarely Never

Diagnosis of the
Disease is Physical
in Origin 42 16 26 7 8

Severity Of the Disease 17 15 18 14 35

Previous Therapy
Received 35 14 18 15 17

Symptoms of the
Disease ("Complaint
vs "Lantanic') 5 5 23 21 44

Clinical Improvement
With Treatment 45 27 22 3 2

Urgency of Appoint-
ment for Care 13 16 40 17 13

Seriousness of Con-
sequences Of Missed

Appointment 24 13 36 13 13
Functional Impairment

Caused by the Disease 2 4 l9 16 58
N=336

hypothesis regarding characteristics of the disease, and the
correlation with appointment keeping is displayed in Table 10.
The distribution, in percent, of responses regarding charac-
teristics of the disease is given in Table 11.

Illnesses that are physical in origin are, perhaps,

easier to recognize and effects Of treatment are easier to

81

observe than illnesses that are from emotional or mental
disturbances. Based on this assumption, it was hypothesized
that persons who dealt, primarily, with health problems diag—
nosed as physical in origin would be more likely to keep their
appointment for preventive health services. The data of this
study did not support this assumption. Furthermore, the
severity of a health condition, or the duration, had no de-
tectable influence on appointment keeping behavior, contrary
to our original hypothesis.

Persons who have had the habit of going to visit the
doctor in order to receive therapy in the past, or who
visit the doctor and undetected health problems are discover-
ed, were assumed to be more prone to keeping appointments
than those who did not go to the doctor as a rule, or have
not experienced the benefit of early detection and treatment
of a health problem. This has not been found to be true.
In this study, it appears that previous behavior patterns of
this nature are of no value in predicting the successful
outcome of scheduling an appointment.

Some people receive treatment at the medical clinic
and are able to observe improvement as a consequence of
the services. Others are inclined to seek appointment with
the doctor because the health condition is viewed to be
urgent and in need Of immediate attention. While supposed
to be persons who have had these experiences are more apt to
keep medical appointments than others, data from this study

does not support this contention.

82

Perhaps, even more surprising, persons who report that
there would be serious consequences if they missed medical
appointments, or those who are functionally impaired by a
health problem, are no more inclined by these factors to
keep agreed upon appointments than those not so affected.
None of the hypotheses regarding characteristics Of the
disease and their impact on appointment keeping were sup-
ported by the study. In taking eight measures to assess
the impact of characteristics of the disease on appointment
keeping behavior, there is a 30 percent possibility that,
at least, one Of the measures would be significant based
simply on chance (Fairweather 1978). This did not occur
and serves to underscore the lack of any association be—
tween appointment keeping and disease characteristics such
as origin, severity, duration, urgency, and the like.

In the fourth hypothesis Of this study, it is proposed
that there is a significant positive correlation between
appointment keeping and the following characteristics of
the therapeutic regimen, namely, whether or not medication
is prescribed, dosage, and prescription of a diet. Also,
that there is a significant negative correlation between
appointment keeping and the duration of the treatment, cost,
side effects, and degree of behavioral change required. The
hypothesis regarding characteristics of the therapeutic re-
gimen, and the correlation with appointment keeping is dis—
played in Table 12. The distribution, in percent, of re-

sponses regarding characteristics of the therapeutic regimen

Test

Therapeutic
Measures

83

TABLE 12

of Characteristics of the Therapeutic
Regimen and Appointment Keeping

Regimen Correlation With
Appointment Keeping

Whether or Not Medication is
Prescribed in the Treatment of

a Disease .02

Duration of the Treatment .07

Cost of the Treatment .13

Side Effects From the Treatment .18*
Dosage Needed for Treatment .13

Prescription of Diet for Treatment

of the Disease .10

Degree of Behavioral Change Required

in Following Therapeutic Regimen .18*

N=336

*Significant at the .05 level of confidence

is given in

In our
significant
keeping and

dosage, and

Table 13.

original hypothesis, it was pointed out that
positive correlation exists between appointment
prescription of medication that is at a low

recommendation of a particular diet in treating

a health condition. It was believed that these actions of

the doctor would not cause undue difficulty to the patient

and would likely give the patient a certain satisfaction

that something was being done to remedy a health problem.

This hypothesis was not supported by the data in this study.

84

TABLE 13

Distribution of Responses to
Therapeutic Regimen Characteristics by Percent

Thereaputic Regimen
Measures Response Distribution

Some-
Always Often times Rarely Never

Whether or Not Medica-
tion Prescribed in the _
Treatment of a Disease 10 12 45 21 10

Duration of the Treat-
ment in Terms Of Brevity 32 20 29 7 10

Cost of the Treatment
in Terms of Being Low 27 11 25 17 20

Side Effects From
the Treatment 22 16 31 12 19

Dosage Needed for

Treatment in Terms

of Being Low 29 16 37 9 8
Prescription Of Diet

for Treatment of the

Disease 24 2 9 6 57
Degree of Behavioral

Change Required in

Following Therapeutic

Regimen 32 18 27 12 9

N—336

Militating against a favorable outcome in scheduling an
appointment would, presumably, be items that would initiate
or cause considerable inconvenience to the patient, such as
the long duration of the treatment, high cost, unwanted side
effects, and serious disruption of daily routine. The hypo-
thesis about duration of the treatment, cost, and disruption

of routine was not supported by the data, however, there is

85

a significant negative correlation between appointment keep-
ing, and the incidence of side effects from previous treat-
ments. There is 30 percent chance that one significant find-
ing would occur coincidentally in seven measures on a factor
(Fairweather 1978). Consequently, it must be concluded that
chance might play a big role in this statistic.

In the fifth hypothesis, it is proposed that there is a
significant positive correlation between appointment keeping
and the following characteristics of patient-therapist in-
teraction, namely, satisfaction with visit to clinic, ex—
pectations met, comprehensiveness of services offered during
the visit, patient education, and level of superVision. The
hypothesis regarding patient—therapist interaction character—
istics, and the correlation with appointment keeping is dis-
played in Table 14. The distribution, in percent, of re-
sponses regarding patient—therapist interaction characteris—
tics is given in Table 15.

Most people attend a medical clinic where they receive
satisfaction with the services, fulfill expectations they
bring to the clinic, Obtain all types of services they need,
and receive a reasonable explanation from the health care
provider on what their health problem is and how the treat-
ment prescribed will be effective. Where all these condi-
tions are met, it would appear that appointments are more
likely to be kept. Our data indicates there is no signifi-
cant correlation between these patient-therapist interaction

characteristics just described and appointment keeping.

86

TABLE 14
Test of Patient-Therapist Interaction
Characteristics and Appointment Keeping
Patient—Therapist Correlations With

Interaction Measures Appointment Keeping

Patients Satisfaction With
Visit to Clinic .02

Patients Expectations Met at
the Clinic .01

Comprehensive Services Offered
During the Visit -.03

Patient Education Received

From Therapist .03

Therapist Follows Up Closely

on Patient .16*
N=336

*Significant at the .05 level of confidence

Only one measureshowed any significant association with ap-
pointment keeping and patient-therapist interaction. It
appears that, when the doctor takes particular interest in
a patient enough so as to follow—up in some way to insure
that the treatment regimen is effective, this person is more
inclined to keep medical appointments. There is a 20 per-
cent chance that, out of five measures taken to assses the
impact of a factor, one of the measures would be significant
(Fairweather 1978). Therefore, the above finding is deemed
to be a weak finding.

The sixth hypothesis suggests that there is a signifi-

cant positive correlation between appointment keeping and

87

TABLE 15

Distribution of Responses to Patient-Therapist
Interaction Characteristics by Percent

Patient-Therapist Response Distribution
Interaction Measures
Some-
Always Often times Rarely Never

Patients Satisfaction
With Visit to Clinic 55 23 14 3 2

Patients Expectations
Met at the Clinic 47 26 20 4 2

Comprehensive Services
Offered During the
Visit 48 23 23 3 2

Patient Education
Received From

Therapist 56 l9 l9 2 4
Therapist Follows-Up

Closely on Patient 35 17 25 10 11
N=336

the following characteristics of access to health care facil-
ity, namely, availability of transportation and telephone
in the home, but a negative correlation with distance of home
from the clinic and indirect care. The hypothesis regarding
accessibility of the health care facility, and the correla-
tion with appointment keeping is displayed in Table 16. The
distribution, by percent, of responses regarding facility
accessibility is given in Table 17.

The data supported the original hypothesis that pre-
dicted a significant positive correlation between the avail-
ability Of a telephone in the home and appointment keeping.

This is not surprising, because the availability of a

88

TABLE 16

Test of Facility Accessibility
and Appointment Keeping

Facility Accessibility Correlations With

Measures Appointment Keeping
Distance From Clinic -.12
Indirect Costs .17*
Availability of Transportation -.14
Telephone in the Home .18*

N=336

*Significant at the .05 level of confidence

TABLE 17

Distribution of Responses to Facility
Accessibility Characteristics by Percent

Facility Assess
Measures Response Distribution

One Mile
or Less 1-2 2-4 4-8 8+

Distance From Clinic
in Miles 15 12 17 16 40

Some-
Always Often times Rarely Never

Indirect Costs to

Patient 7% 9% 32% 10% 41%
Availability of

Transportation 16% 7% 24% 17% 35%
Telephone in the Home 28% 4% 6% 10% 52%

N=336

89

telephone makes it easier to remind clients of their ap-
pointment and, conversely, it is easier for the client to
call in to cancel or reschedule as needed.

The hypothesis regarding the availability of transpor-
tation, however, was not supported. It was assumed that,
if transportation was readily available, the patient would
be more inclined to keep prearranged appointments. It is
apparent from the data that other factors rather than easily
available transportation must be sought to predict schedu-
ling outcomes.

Distance between the home and the clinic did not prove
to be significant as originally hypothesized. However,
there was a significant positive correlation between suc—
cessful scheduling outcomes and reduction of indirect costs
to the patient. Although the Medicaid eligible family does
not have to pay for health care services received, there are
indirect costs such as paying for transportation, hiring a
baby sitter, eating lunch out, investment of effort to dress
the children, fill out forms, and, in general, manage the
many details to keep an appointment and follow-up as needed
with health care. The finding indicates that reduction of
these indirect costs correlates with more successful schedu-
ling outcomes. That two of the four scales used to measure
importance of facility accessibility were significant is an
important finding. This finding is significant at the .01
level of confidence.- It is clear that accessibility is an

important consideration in scheduling appointments.

90

In our seventh hypothesis, it is predicted that there
is a significant positive correlation between appointment
keeping and the following characteristics of the facility/
adminbtrative process, namely, clinic waiting time, refer-
ral source, referral to specific provider vs. referral to
clinic, time from scheduling to appointment, patient-staff
ratio, scheduling method (block vs. individual time),
availability of parking, use of prospective reminders, and
retrospective reminder, but a negative correlation with
scheduling errors.

The hypothesis regarding characteristics of the facili-
ty/administrative process and the correlation with appoint-
ment keeping is displayed in Table 18. The distribution,
in percent, of responses regarding characteristics Of the
facility/administrative process is given in Table 19.

The hypothesis predicting that patients will not keep
their appointments if, in the past, the doctor made a mis-
take in scheduling, e.g., giving a wrong time or day to
see the patient, was upheld by the data of this study. A
significant negative correlation of —.16 was found between
appointment keeping and incidence of errors.

Two other measures of the facility/administrative
process factor were also found to be significant as pre-
dicted in the original hypothesis. It has been found that
an adequate number of staff persons is important in at—
tracting people into the clinic. The most important find-

ing in the study involves the time factor between scheduling

91

TABLE 18

Test of Facility and Administrative Process
Characteristics and Appointment Keeping

Facility/Administrative Correlations With
Process Measures Appointment Keeping
Clinic Waiting Time .04
Referral Source for Selecting A Clinic -.10
Referral to Specific Provider vs.
Referral to Clinic .12
Patient-Staff Ratio .18*

Scheduling Method (block vs. individual

time) .15
Availability of Parking .12
Use of Prospective Reminders .01
Retrospective Reminders —.04
Scheduling Errors —.16*
Number of Days Between Scheduling

and Appointment .54*

N-336

*Significant at the .05 level of confidence

and the appointment itself. The shorter this time is the
more likely the client will keep the appointment. A cor—
relation Of .54 was found indicating a highly significant
association. Thus, the immediacy of a service or event, the
more likely participation will follow.

The hypothesis that waiting time in the clinic is an
important factor influencing appointment keeping did not

prove to be so. Likewise, no significant correlation was

92

TABLE 19

Distribution of Responses to Facility/
Administrative Process Characteristics by Percent

Facility/Aministrative
Process Measures Response Distribution

Some-
Always Often times Rarely Never

Clinic Waiting Time
in Terms of Short Time 23 20 41 9 7

Referral From Family

or Friends in Selecting

3 Clinic in Terms of

Importance 27 19 33 7 14

Preference for a

Referral to Specific

Provider vs. Referral

to Clinic 39 16 32 6 7

Specific Time Schedu-
ling an Appointment is

Set 57 13 16 6 6

Adequate Number of Staff 53 24 18 2 1

Availability Of Parking

Close to Clinic 69 17 10 1 1

Uses Doctor Prospective

Reminders 30 9 18 13 29

Doctor Uses Retrospective

Reminders 16 5 18 13 47

Scheduling Errors Made

by Doctor 32 18 15 7 26
5 Days 6-10 11-14 15+
or Less Days Days Days

Number of Days Between
Scheduling and Appoint-
ment 27 18 12 25

N=336

93

found with the method used in scheduling, such as referral
to a specific provider versus referral to a clinic, or
scheduling by block method versus assignment of a specific
time to an individual.

The convenience of a parking area close to the clinic
was found not to be a critical consideration. Prospective
reminders from the doctor or restrospective notices for
missed appointments also were of small consequence. Recom—
mendations from other family members, relatives, or friends
about a specific clinic did not prove to have significant
influence in keeping an appointment. These findings were
contrary to the original hypothesis.

Out of ten scales used to measure the impact of facili-
ty/administrative process, three were found to be signifi-
cant at the .05 level of confidence. To find three measures
out Of ten valid in measuring this factor is significant at
the .01 level of confidence (Fairweather 1978).

The last hypothesis of this study predicts that there
is a significant positive correlation between appointment
keeping and the following characteristics of the environ-
ment, namely, influence of family and friends, family stabil-
ity, familysize, time of day of appointment, day of week of
appointment, illness in family, performance of children in
school, and presence of both parents in the home, but a
negative correlation with weather (in terms of the patient's
own assessment as being good or bad), and time on public

assistance. The hypothesis regarding characteristics of the

94

environment and the correlation with appointment keeping is
displayed in Table 20. The distribution, in percent, of
responses regarding characteristics of the environment is

given in Table 21.

TABLE 20

Test of Characteristics of the
Environment and Appointment Keeping

Correlations With

Environment Measures Appointment Keeping
Weather -.05
Influence of Family and Friends .05
Family Stability -.02
Family Size .03
Time of Day for the Appointment -.02
Day of Week for the Appointment -.02
Illness in the Family .13
Performance Of Children in School .14
Time on Public Assistance -.04

N=336

The highest association between appointment keeping
behavior and items used to measure the influence of the
environment was found with the presence of illness in the
family, and the poor performance Of children in school.
The association, however, was not large enough to be sig-
nificant contrary to our original projection. There was

even a far less association with items hypothesized as

95

OOOuz

 

ON OH HH OH OO
+m OIO OIN NIH H cane
mmOH mnmow :H mocmanmm< OHHOOO so many
OH OH ON OH ON
OOO\Hnm .mn:£H wammcvoz .mmsH .coz mncmancnoma< now xoo3 wo Ono
N «N OH ON OH
2m q 2O SO SO z< OH
nonm< «IN NINH NHIOH onowmm mncOEnOHOOO< now One mo OEHH
N O NN Om OH
+0 OIN Olm «In NIH mnegEmz no nmnaaz NO aunm annemn
ON NH NH HN ON
+OH NHIOH OIN OIO mnmo» oEOm man OH OHHno anOHO
muc man no manmn an Nunnnnmnm NHHEan
ON NH ON O «N noom mH Hoonom CH connHHno
mo OOcmEnOmnom cosz nOnOOO m On mOOO
n O NN an sq annsmn men
CH mmOcHHH mH Onoce cos: nonoon m On mooo
«H N OO OH NN OHOHHU m OcHnOOHOO CH mcoonnm
Ocm NHHEmm mo mocmaHmcH osn mo mocmnnanH
nm>mz mHmnmm mOEHnOEOO :OnOo mNm3H<

 

coHnOOHnanO mmcoamom

mmnommoz newsconH>cm

 

ncoonmm

NO newscOnH>cm Onn mo
mOHannOnomnmso On momcoammm mo :OHnannanO

HN MHm<H

96

significant, namely, family stability measured in terms of
the oldest child living at home, family size, and time of
day or week of the appointment. The length of time spent

on public assistance and bad weather conditions were assumed
to be important negative factors leading to broken appoint-
ments. The study found that these variables have no sig-
nificant influence on appointment keeping excluding severe
weather conditions.

None of the nine scales for measuring the influence
of environment were significant. There is a 30 percent
chance that, with this number of scales,at least one would
be found to be significant (Fairweather 1978). This sta-
tistic underscores the independence of environmental factors
and appointment keeping behavior.

In concluding this section on testing of the eight
hypotheses, it should be noted that, out of eight cate-
gories of variables, namely, those related to the patient,
providers of care, the disease, the therapeutic regimen,
patient-therapist interaction, access to clinic, facility/
administrative process, and the environment, five contained
at least one scale that was significant. Of these five,
only access to the facility and facility/administrative pro-
cess had enough significant scales to be considered important
measures of appointment keeping behavior. Table 22 lists
variables found to be significantly correlated with appoint-
ment keeping. Taken as a whole, the most important contri-

bution of this study is in terms of what is not related to

97

TABLE 22
Variables Found to be Significantly
Correlated With Appointment Keeping
at the .05 Level of Confidence

Correlations With

Variables Appointment Keeping
Years Medical Worker Has Been

on the Job -.16

Side Effects From the Treatment .18

Degree of Behavioral Change Required

in Following Therapeutic Regimen .18
Therapist Follows Up Closely on

Patient .16
Indirect Costs .17
Telephone in the Home .18
Patient-Staff Ratio .18
Scheduling Errors -.16

Number of Days Between Scheduling

and Appointment .54
appointment keeping. The few significant scales found, how-
ever do provide some clue to predicting successful schedu-
ling, especially in regard to the element of time between
scheduling and appointment.

Relationship Among Measures

 

In the second portion of this analysis, we will look
at the relationship among the measures used in testing the
original hypothesis. This second level analysis involves
conducting a cluster analysis according to the B. C. Tryon
and Bailey (1970) method. As noted in the previous chapter,

in a cluster analysis of variables using the.B. C. Tryon

98

TABLE 23
Variables Found to be Significantly
Correlated With Appointment Keeping
at the .10 Level of Confidence

Correlation With
Variables Appointment Keeping

Years Medical Worker Has Been
on the Job -.16

Side Effects From the Treatment .18

Degree of Behavioral Change Required

in Following Therapeutic Regimen .18
Therapist Follows Up Closely

on Patient .16
Indirect Costs .17
Telephone in the Home .18
Patient-Staff Ratio .18
Scheduling Errors -.16
Number of Days Between Scheduling

and Appointment .54
Previous Contact With Clinic -.14
Home Close to Clinic -.12
Transportation Available -.14
Doctor Prescribes a Diet .10
Missed Appointments Previously -.09
Avoids the Health Department .10
Tends to Stay Well .09
Tends Toward a Particular Sickness .11

Physical vs. Mental Cause of
Sickness .10

Sickness Tends to be Serious .11

Low Cost for Care .13

99

TABLE 23 (cont'd.)

Correlation With

Variables Appointment Keeping
Dosage Prescribed is Low .13
Recommendation From the Family -.10
Appointment With a Specifid Doctor .12
Appointment at a Specific Time .15
Parking Close to Clinic .12
Visits the Same Doctor .13

Difficulty in School Results in Visit

With Doctor .14
Attitude of Outreach Worker .09
Time of Day for the Appointment —.13
Protestant .14
Catholic .13

Out of the 81 variables included in this study, only
9 were found to be significant at the .05 level of confi-
dence. When looked at from the .10 level Of confidence,
32 variables were found to be significant as indicated in

Table 23 (above).

100

approach, composite groups of similar variables are objec-
tively formed rather than based on preconceived notions.
These groups will be formed so that variables assigned to
each cluster are similar within themselves but different
from other groups. The analysis is based on the degree of
collinearity among the definers of the group. "Generally,
collinearity is defined by the line graph of the correlation
coefficients of two variables with all the variables in the
study, their correlation profiles. Collinear variables have
the same profile of correlations. . . Clusters of collinear
variables have two objective characteristics of similarity:
they correlate positively with each other, and they follow
the same pattern of correlations with other variables. They
also are objectively different from other clusters of col-
linear variables because their common correlation profiles
have a different shape from that of other clusters" (Tryon
and Bailey 1970, p. 47). Measure was also made on the re-
lational strength between clusters. This was done by comput-
ing the interdomain correlations.

When a cluster analysis was done to discover correla-
tions among the many variables used in this study, forty-
eight or over half of the variables were eliminated because
their commonalties were too trivial (that is, less than .20)
to be meaningful. These variables include the following:

- gender of the client
- ability to speak a foreign language

- born in the state of Michigan

101

time of the day for the appointment

day during the week of the appointment
occupation of the father

frequency of visiting a medical clinic

distance from home to the clinic

location of home in terms of urban or rural
number of years on public assistance
availability of transportation

availability of phone in the home

prescription of medication at the time of clinic visit
previous record of missing medical appointments
previous use of the clinic

tendency to stay well

seriousness of the sickness

effectiveness of treatment

painfulness of treatment

belief in reward for good behavior

belief in doctor's ability to heal

being head of the household

ownership of a thermometer

physical causation of sickness

improvement after following doctor's instruction
perceived seriousness of missing a medical appointment
prescription of weak medication

treatment time is short

cost of treatment is low

close check kept by the doctor

102

- recommendation from family regarding a clinic

- Doctor calls to remind client of appointment

- weather conditions

- influence of family regarding keeping an appointment
- policy of going to the doctor in time of sickness

- policy of taking child to doctor when doing poorly in
school

- education of the outreach worker
- gender of the outreach worker

- race of the outreach worker

- caseload of the outreach worker

- attitude of the outreach worker in terms of making
health screening required

- time of day for the screening appointment
- previous history of screening
- religion of client
— race of the client

This does not mean that these variables are not impor-
tant but only that, because of their independence, they can-
not be used to measure the same phenomenon as the clusters.
The remaining variables form into nine independent clusters
which are listed in Table 24 that follows. To further ap-
preciate the findings of this cluster analysis, it is ne-
cessary also to-examine the interdomain correlations. Table

25 contains the correlations between oblique cluster domains.

Cluster

103

TABLE 24

Nine Clusters of Scale or Variable Scores

Cluster I - Participation in Health Screening

1.

2.

The client is willing to accept and keep

a screening appointment.

The client keeps an appointment when fewer
days intervene between scheduling and
appointment.

Cluster II — Health Care Provider and Patient

N
o

b

Information

The expectation of the client are met.
Client is satisfied with the medical clinic.
Client receives all the services needed at
the clinic.

Waiting time for services is short.

Health professional takes time to explain

a condition and how to care for it.

Cluster III — Client's Intellectual Ability

1.
2.

The client has a high IQ.
The client has a knowledge of sicknesses
and how to care for them.

Cluster IV - Maturity of the Client

H

Clients tend to be 25-34 years of age or older.

The oldest child in the family tends to be
13 years of age or older.

The size of the family is generally 3-4
members.

The highest income for a given year is less
than $10,000.

Cluster V - Health Status of Client

1.

N
0

Client does not have a tendency to a parti—
cular sickness, e.g., heart problems, cancer,
diabetes, or the like.

Past sicknesses have not been severe.

Client lives closer to the large urban
center of the state.

Treatment in the past for health problems

did not produce side effects.

Little change in routine resulted from
treatment ordered by a doctor.

Loading

0.77

104

TABLE 24 (cont'd.)

Cluster Loading
6. The doctor is acurate in scheduling an
appointment. 0.47
7. Sickness does not require immediate
attention. 0.46
8. The doctor does not prescribe a diet as a
treatment. 0.45
9. The origin of the sickness is through
physical causes. 0.35
Cluster VI - Experience of Health Care Workers

1. The health care worker tends to be 34-55

years of age. 0.71
2. Years on the job are less than three years. 0.70
Cluster VII — Financial Resources

1. The highest income for a given year is

less than $10,000. 0.64
2. Income for 1984 is less than $10,000. 0.63
3. Time on assistance tends to be short. 0.37

Cluster VIII - Therapeutic Regimen

l. Treatment for a health problem was received

in the past. 0.63
2. Drugs were needed in the treatment. 0.56
3. Health problem was unknown prior to visit

with a doctor. 0.56
4. Mobility was hindered by health problem. 0.50
5. Sicknesses in the past tend to be long. 0.42

Cluster IX — Availability of Health Care Provider

1. There is adequate staff to provide health care. 0.64
2. A specific time is set aside for the patient

by the doctor. 0.45
3. A specific doctor is made available as desired. 0.3

Cluster Description

 

Cluster I - Participation in Health Screening
Only one variable out of the 81 included in this study
had a similar collinearity with the major outcome variable,

namely, participation in preventive health care screening.

105

oo.H

ma.

NH.

MH.I

NH.

0H.

Ho.

mocmvfimcoo mo Ho>oH

I «mm. HH. Ho. k
50.: NN.I no.
mo. «we. «N.I
co. Ho. 00.
mo. 0H.I *
«H.I
m c m

ow.

ON.I

mo.

co.l

mm.

mo.l

OH.

kmm.

mH.I

no.

mm.l

ma.

qo.l

Ho.

Ho.

oo.H

mo. ago um ucmuawaawum«

umvfi>oum mo
suaaaamflfim><

amsfiwom
ofiusommuonh

moousomwm Hmfio
Iamswm m.uaofiHo

mocm
lawmaxm m.uwxu03

moumum nuammm

unmaau
osu mo huwusumz

sufiflun<
HmsuomaawuaH

:ofiuomumucH
ucowumm\umvw>oum

wcficmmuom nuHmmm
CH coaumoaofiuumm

.xH

.HHH>

.HH>

.H>

.>H

.HHH

.HH

.H

mumumsao

mCHman umumsao mscfiaso :moBumm mcowumamuuoo

mm mum<H

106

This variable pertains to the time element between schedu-
ling and appointment. The client is more likely to keep

an appointment when fewer days intervene from the time of
scheduling. The cluster is called "Participation in Health
Screening" because this is the pivotal variable that best
defines the cluster. In general, what this analysis says

is that the best approach to promoting preventive health care
screening among low-income families is whatever works, and
the worker is advised to get the family in as quickly as possi—
ble because what worked to get the agreement for participa-
tion from the family will not work for long.

Not only do the other 79 variables fail to have a simi-
lar collinearity with the first cluster, but, also, none of
the other cluster domains, with the exception of the ninth
cluster, have any significant correlation with Cluster I,
namely, "Participation in Health Screening".

Cluster I has an extremely small (.01) association with
provider/patient interaction (Cluster II). The second clus-
ter includes such seemingly important measures as the extent
to which patient's expectations are met, satisfaction with
services received, desire for comprehensive services, length
of time spent waiting for services, and education received
regarding a health condition and how to care for it. The
lack of correlation here is something of a surprise.

The association is even less (.01) between program par-
ticipation and the client's intellectual ability as defined

in Cluster III. It appears that higher intellectual skills

107

or even above average knowledge of sicknesses and how to
care for them are not useful in predicting those who will
participate in a preventive health program such as health
screening.

The finding, just described, in associating Cluster I
with Cluster III receives greater affirmation in the associ-
ation between Cluster I and Cluster IV. The correlation
with this cluster is negative (-.04). It appears that the
maturity of the client in terms of chronological age of the
parent and children, size of family, and level of income have
no significant association with appointment keeping and bre-
vity of time between scheduling and appointment.

A very slight negative correlation (-.04) exists be-
tween Cluster IV described as the "Maturity of the Client"
and program participation. The lack of meaningful associa-
tion between program participation and the maturity of the
client signifies, in effect, that older parents who have
older children are no more inclined to keep their appoint-
ments for health screening than younger families. Further-
more, larger families are no more responsive than smaller
ones. Families with higher incomes do not keep appointments
more so than those with lower incomes.

A much higher association (.15) exists between the
health status and health screening participation. This
association, however, is too small to be significant. The
important items included in Cluster V is that the client does

not have a tendency to a particular sickness, e.g., heart

108

problems, cancer, diabetes, and the like. A host of other
variables serve to further define the measuring of health
status. These include severity of past sicknesses, proximi-
ty to urban centers, lack of side effects from previous
treatments received, minimal change in routine due to treat-
ment regimens, care of doctors in scheduling appointments,
lack of urgency for attention to the health problems, lack
of need for a diet in treating the condition and cause of
the disease is physical rather than mental or emotional.

The participation in health screening, as defined in
Cluster 1, is associated with younger workers who are re-
latively new on the job. These two variables make up the
sixth cluster which has a negative correlation (-.25) with
Cluster I, however, is not large enough to have significance.

There is very little association (.07) between a
client's financial resources as defined in Cluster VII and
program participation. The amount of income for 1984, or
the brevity of time the family is on public assistance do
not, as a cluster, score associate meaningfully with scores
on the first cluster.

Cluster I has some association (.15) with Cluster VIII.
The correlation, however, is negative. It is of interest
to note not only the lack of significant association with
participation in health screening, but, also, that the
association is negative for the variables that compose
Cluster VIII, namely, frequency of receiving treatment for

health problems in the past, frequency of drugs used

109

for treating a health problem, knowing the existence of a
health problem prior to visiting with a doctor, the impair-
ment of mobility due to the illness, and the length of time
for mobility.

Only one cluster was found to have a significant cor-
relation (.28) with Cluster I, "Participation in Health
Screening". This cluster is called the "Availability of the
Provider" and includes the adequacy of staff available at
the health clinic, setting aside a specific time for the
patient, and making an appointment with a specific doctor.
The association indicates the importance of having adequate
staff to give personal attention to clients as soon as possi-
ble to insure program participation. The finding clearly
constitutes a common sense approach to insure the success
of any program.

Cluster II - Health Care Provider and Patient Interaction

The title assigned to Cluster II is "Health Care Pro-
vider and Patient Interaction" because the cluster is prin-
cipally defined by "meeting the expectations of the client"
at the clinic as the pivotal variable and, further defined,
by the level of client satisfaction and comprehensiveness
of services. 0f less importance, but also included in this
cluster, are variables related to the waiting time for ser-
vices, and the care providers take to explain what the health
problem is and what the treatment for it will do.

Cluster II is one of two that remains intact out of

the eight theoretical clusters with which we began this

110

study. The only variable in the original cluster that lacked
collinearity was the variable regarding close follow-up on
the patient by the doctor.

Cluster II had a small, but insignificant, relationship
with Cluster III. The interaction between provider and
patient as defined in Cluster II had little or no bearing on
the level of the client's intelligence or how much the client
knew about sickness or how to care for it. The same holds
for "Maturity of the Client" asdefinedin Cluster IV. Age
of client, size of family, age of children, or income do not
have association with provider/patient interaction.

There is, however, a significant association between
Cluster II and the health status of the client as defined
in Cluster V. Thus, it follows that when a client's expec—
tations are met, he/she is satisfied with the medical clinic,
receives all the services needed, waits for a short time for
health care, and receives adequate explanation from the
doctor about his/her health problem and what the treatment
will do. Furthermore, this same client does not have a ten—
dency to a particular sickness, e.g., heart problems, cancer,
diabetes, and the like, has not had severe sickness in the
past, lives closer to large urban centers, experienced few
side effects from past treatments, has had little change in
routine due to treatment ordered by the doctor, has little
need for immediate attention, is not placed on a diet, ex-
perienced few, if any, errors in scheduling by the doctor,

and suffers from health problems due to physical causes

111

rather than mental or emotional. This finding is what one
would have expected since a less frequent contact with the
health care system, because of a low morbidity rate, would
likely result in less friction because of unmet needs and,
consequently, greater satisfaction with health services
generally.

Cluster II has a slight negative association with Clus-
ter VI, "Worker's Experience". Because the correlation is
insignificant, nothing can be concluded between provider/
client interaction as defined in Cluster II, and the age
or years of experience of the health care worker. The same
must be said regarding client's financial resources as de—
fined in Cluster VII and provider/patient interaction. The
correlation between these clusters is .08. The level of
income or time on public assistance has no significant associ-
ation with provider/patient interaction factors.

There is a negative correlation (—.20) between Cluster
II and Cluster VIII, but the association is not large enough
to be significant. The relationship was, however, larger
than with most of the other clusters. Provider/patient in-
teraction, as defined in Cluster II, is somewhat associated
negatively with clients who received treatment for health
problems in the past, who needed drugs in the treatment
regimen, who were unaware of health problems prior to visit-
ing the doctor, who had long sicknesses in the past, or whose
mobility was hindered by health problems.

Cluster II, "Provider/Patient Interaction", had its

112

strongest association (.26) with Cluster IX, "Availabil—
ity of the Provider". This association is significant at
the .05 level of confidence. Thus, when a client's expecta-
tions are met, he/she is satisfied with the medical clinic,
receives all the services needed, waits a short time for
health care, and receives adequate explanation from the
doctor about his/her health problems and what the treatment
will do. This same client experiences adequacy of staff at
the doctor's office and is able to receive services from a
specific doctor at a specific time.
Cluster III - Client's Intellectual Ability

This cluster was named "Client's Intellectual Ability"
because the defining pivotal variable is based on the in-
tellectual level of the client. A further definer variable
is the knowledge the client has regarding sicknesses and
how to care for them. No other variable had a collinearity
similar to these two. The correlations of this cluster with
the eight others were too small to be significant. The
highest correlation (-.24) occurred with Cluster VII, "Cli-
ent's Financial Resources". The level of intelligence and
knowledge of how to care for sicknesses correlated negatively
with level of income and time spent on assistance. The
association, however, may be one of chance rather than of
predictive value due to its lack of significant correlation.

Although Cluster III had low association with the other
clusters, generally, there was a negative correlation of

-.14 with Cluster IV, "Maturity of the Client", and a -.10

113

with the health status of the client as defined in Cluster

V. Furthermore, a trivial correlation of less than .05 was
found between the Client's Intellectual Ability and the
Worker's Experience (Cluster VI), the Therapeutic Regimen
(Cluster VIII), and the Availability of the Provider (Cluster
IX).

Cluster IV - Maturity of the Client

This cluster is so named because of the age of the client
is the defining variable. The age of the oldest child in
the family serves as further definition of this cluster.

Two other scales were found to have similar collinearity,
namely, the size of the family, and the highest yearly in-
come received.

The highest correlation that occurred among the cluster
domains was found between Cluster IV and Cluster VII, "Client
Financial Resources". The clients who tend to be older and
have older children also had a higher income in 1984, and
tend to stay on public assistance for a shorter period of
time. The only surprise, perhaps, is the larger size of
families, however, the median size of the household in the
sample population consisted of three to four members.

Cluster IV had a negative but reasonably high correla—
tion (—22) with Cluster VIII, "Therapeutic Regimen". Older
clients with older children and larger families with higher
incomes tend to have received treatment in the past, used
drugs for health reasons, have health problems unknown prior

to a visit with the doctor, have mobility impaired because

114

of a health problem, and tended to have longer bouts of
illness. The association, however, is not strong enough
to be significant.

Cluster IV had only trivial relationship with Cluster
V (Health Status), Cluster VI (Worker's Experience), and
Cluster IX (Availability of the Provider). As mentioned
regarding client's intellectual ability, so also client's
maturity are not at all associated with health status of
the client as defined in Cluster V.
Cluster V — Health Status of Client

This cluster, comprised of nine scales, is the largest
of the nine clusters. Its name is determined by its pivotal
variable, namely, the client's tendency to a particular
sickness, e.g., heart problems, cancer, diabetes, and the
like. This cluster is further defined by the two variables
of tendency to less severe sicknesses, living in more urban
settings, and lack of side effects from treatment regimens
in the past. The collinearity of five other variables were
similar to this cluster and, consequently, were included.
These variables are, namely, little change in routine result-
ing from treatment ordered by the doctor, accuracy in schedu-
ling appointments, sickness needs less immediate attention,
prescription of a diet as treatment, and origin of the sick-
ness from physical rather than from emotion or mental causes.

Cluster V had a significant correlation (.33) with
Cluster IX, "Availability of the Provider". This finding

illustrates that when adequate staff is available and

115

personal attention given to the patient by way of getting
services from a specific doctor at a specific prearranged
time, the patient will not have a tendency toward a major
illness, nor severe bouts with sickness, will receive fewer
side effects from treatment, will have less disruption in
his/her life, will not require immediate attention for sick-
nesses which will, most likely, be fromphysical causes, and
will not end up on a diet to correct health problems. This
seems to be a very reasonable association of notions.

The association of Cluster V with Clusters VI, VII,
and VIII was trivial. Health status of the client is not
affected by the medical outreach worker's age or years on
the job, nor is it affected by the client's level of income
or years on public assistance.
Cluster VI - Experience of Health Care Workers

Cluster VI is comprised of only two scales. It has
a pivotal variable based on the age of the worker. A fur+
thur definer of this cluster is the number of years the
medical social worker has been on the job. This cluster had
no significant association with any of the other eight
clusters.
Cluster VII - Client's Financial Resources

The defining variable for Cluster VII is the highest
income a client received for any given year. The level of
income for 1984 serves also to define this cluster. Those

with higher incomes, as would be expected, also spent less

116

time on public assistance. This letter variable serves as
the third scale comprising this cluster.

Cluster VII had a negative correlation (-13) with
Cluster VIII, "Therapeutic Regimen". Persons with higher
incomes and less time on public assistance tended not to
have had treatment for health problems in the past, did not
need drugs for treatment of health problems, knew of health
problems before visiting with a doctor, tended not to have
mobility impaired by a health problem, and tended to have
brief episodes of illness. The correlation among these two
cluster domains, however, is not high enough to be signifi-
cant and, therefore, may be only of chance occurrence.

There is also a similar correlation (.12) of Cluster
VII with Cluster IX, however, it is in the opposite direc-
tion. Thus, persons with more financial resources tend to
receive health services from doctors with adequate staff,
have a specific appointment time, and a specific doctor.
This tendency is, however, to low to be considered statis-
tically significant.

Cluster VIII - Therapeutic Regimen

The defining variable for Cluster VIII is that "treat-
ment for a health problem was received in the past". Two
other variables serve to give further definition. These
are that drugs were needed in the treatment regimen, and
the health problem was unknown prior to a visit with a
doctor. Two other variables had collinearity similar to

this cluster, namely, mobility hindered by the health

117

problem, and tendency of the illness to be long. Clients

who have received treatment for health problems in the past,
who needed drugs in the treatment, who were unaware of a
health problem prior to visiting the doctor were impaired
intheir mobility by the health problem, and whose past ill—
ness were long in duration tend, also, to go to doctors who
have adequate staff and received personal attention as de-
fined in Cluster IX. The correlation (.13), however, be-
tween Cluster VIII and Cluster IX is too small to be sig-
nificant.

It is of interest to note that the "Therapeutic Regi-
men" Cluster is the second of two from the original eight
theoretical clusters that survived the cluster analysis.
Some of the variablesfrom the original cluster were elim-
inated because of lack of collinearity. These include
cost of treatment, side effects resulting from drugs taken,
level of dosage needed, prescription of a diet, and the
degree of mobility impaired by a health problem. These
variables, as mentioned earlier, were either eliminated
because of lack of commonalty, or were included in other
clusters.

Cluster IX - Availability of Health Care Provider

The adequacy of clinical staff serves to define this
last cluster. Assigning a specific time for an appoint-
ment verses block scheduling further defines the meaning
of this cluster. Included, also, is a third variable,

namely, assignment to a specific doctor rather than to a

118

clinic for health services.

Two of the variables in this cluster are from the
original cluster designated as "Facility and Administrative
Process". Other variables, from the original theoretical
cluster, that did not prove to be collinear with Cluster
IX, "Availability of the Health Care Provider" include
waiting time at the clinic, recommendation of the clinic
from family and friends, availability of parking, use of
prospective and retrospective members, scheduling errors,
and the number of days between scheduling and appointment.
What was formerly considered one of the large and more
important clusters has, in fact, become a very small clus-
ter of scales. However, it has been found, in its reduced
size, to be correlated with more of the other clusters
than any other.

Summarily, a sharp difference is evident between con-
ceptual clusters and those found statistically. Six of
the original eight clusters failed to survive as meaning-
ful categories of scales to give insight into appointment
keeping behavior. The original clusters eliminated includ-
ed patient features, provider features, features of the
disease, access features, features of the facility/adminis—
trative process, and features of the environment.

Out of the 15 social behavior and perception variables
originally thought to be important in predicting appointment
keeping, only four were retained as having sufficient com-

monalty with other variables to be included in the clusters.

119

These four are: belief in the susceptibility to parti-
cular diseases, knowledge of disease and therapy for them,
level of intelligence, and prescription of drugs by the
doctor. It appears that a lot of caution should be taken
in using social behavior factors in predicting scheduling
outcomes.

The same must be said for variables related to
access to the clinic. None of the variables originally
identified as important survived in statistically formed
clusters.

The statistically formed clusters focus our attention
on issues directly related to provision of health care
services. Cluster I suggests timely service; Cluster II
looks at patient satisfaction issues; Cluster V points
to the level of need for health services; Cluster VIII
reflects on previous use of the health care system; and
Cluster IX underscores need for the availability of ade-
quate staff for direct personal services to the patient.
The statistically formed clusters gave a basis for a
more realistically formed theoretical framework for un-
derstanding health related behavior of Medicaid eligible

clients, specifically in regard to appointment keeping.

V. DISCUSSION

For the sake of clarity due to the vast number of
studies on the appointment keeping issue, it is helpful
to use the original categories of variables suggested by
Haynes (1976) as a guide for developing a contrast between
findings of this study and those of earlier studies.
These categories include features of the patient (both
demographic and social behavior) features of the medical
provider, features of the disease or reasons for appoint-
ment, features of the patient/provider interaction, fea-
tures of the therapeutic regimen, features of the medical
facility and administrative process, features of access
to the facility and environmental features.

Demographic data on patient dropouts are available in
a large number of studies but often are inconsistent and
unenlightening. The present study supports this general
conclusion. A number of researchers (Jones 1971, Shonick
1977, Hurtado 1973) found higher rates of broken appoint-
ments among the younger age patients. This study found
a negative (-.01), although insignificant, correlation
with age level and appointment keeping.

In regard to education, Stine (1968) showed a signi-
ficently increased incidence of failed appointments in
patients who did not complete high school compared to a
group of high school graduates. This study shows a

negative correlation (-.07) between the level of education

120

121

and appointment keeping. It appears that the level of
education is not a predictor of health related behavior
in terms of this study.

Socio-economic status is an important predictor of
keeping or not keeping appointments according to a number
of researchers (Alpert 1964, Badgley 1961, Hoenig 1966).
Only a slight positive correlation (.03) could be detected
between socio-economic status and appointment keeping. It
was far from being large enough to be considered significant.

Researchers are sharply divided as to the importance
of race as predictor of appointment keeping. Three studies
suggest that non-White populations have a significantly
increased rate of missed appointments (Alpert 1964, Badgley
1961, Jones 1977). More recently, Hertz and Stamps (1977)
found that low-income patients and patients from ethnic
minorities are more likely to break appointments. Craig
(1976) found that low-income, inner city Black patients
are at least as capable as other patient groups of sustained
therapeutic contact. This study is inconclusive on the
issue since only 3 percent of the population included in
the study was Black.

Patient demographic variables of gender, occupation-
al status, and religion appear to have little or no effect
on attendance behavior according to Baekeland (1975),

Gates (1976), and Dervin (1978). This study concurs with

the findings of these researchers.

122

It is harder to measure the characteristics of a pa-
tient; fewer studies consequently have been done in this
area. Tash (1969), Kegeles (1973), and Haefner (1970)
found perceptions by the patient of the seriousness and
susceptibility of a disease and belief in the efficacy of
therapy has an important correlation with medical recommen-
dations in general, including appointment keeping behavior.
This study did not support that position. The highest cor-
relation among these three factors, just mentioned, was
found to be .11, namely, between belief in the seriousness
and the susceptibility of a disease in the efficacy of
treatment and appointment keeping. This is far from a
significant assocation.

Tagliacozzo (1970) suggests educational efforts can
improve the rate of broken appointments because there is a
correlation between the patient's knowledge of disease and
how to care for it. This study found a -.01 correlation
between knowledge of a disease and appointment keeping. This
is hardly enough of an indication to justify an educational
campaign. Haefner (1970) found that altering a person's
belief regarding health matters does not automatically trans-
late into changed behavior. Findings in this study strongly
suggests Haefner's remarks.

Most studies limit the investigation to variables that
pertain to the patient; few focus on provider related char-
acteristics. It has been found (Hoenig 1966) that female

therapists treating female patients experienced higher

123

appointment keeping rates than other sex pairings. Nothing
can be said to dispute his finding from later in his study
since only 1 percent of providers were male.

Age, according to Hurtado (1973), is weakly associated
with appointment keeping, favoring the older health care
practictioner. In this study, the tendency is in the op-
posite direction, but the association (-.02) is so small
that nothing can be concluded.

The attitudes of the provider has had important in-
fluence on keeping appointments according to Baum (1966),
Sethman (1971), and Dervin (1971). These studies measured
the influence of ethnocentrism, unconcern for, dislike of,
or boredom with the patient. In this study, the key at-
titudinal measure was the invoking of economic penalty for
families who did not participate in health screening by
way of withdrawing from them their eligibility for public
assistance. This attitude was viewed as one using force
to get compliance versus convincing a family to value health
screening and, consequently, to attend freely. Those who
took the educational sales approach were less effective as
there was a -.09 correlation with appointment keeping. The
association, however, was too small to permit this inference.

It seems a wise policy to make health screening op-
tional because the overall arm of EPSDT is to achieve pre-
vention as a life style. A comprehensive screening is only
a small part of this difficult goal. Use of force, may in

fact, be counter productive in the long run.

124

Characteristics of the disease provide little clue to
appointment keeping behavior. Duration of an illness was
found unimportant (Badgley 1971, Click 1965). An associa-
tion of .13 was found between appointment keeping and the
duration of the illness. This was not high enough to merit
a significant correlation. Findings in this study agrees
with that of other researchers regarding this issue.

Sackett (1976) reports that patients with a specific
health problem have a compliance with appointments between
70 to 78 percent, compared with studies of patients without
specific complaint who have an appointment compliance of
47 percent. No such finding is found in this study. Only
a .09 correlation could be found between appointment keep—
ing and previous knowledge of a health problem. This
association is too small to be meaningful.

Glowgow (1970) reported that patients with severe
functional impairments do not keep their appointments as
faithfully as patients with intermediate degrees of impair-
ment. This seems logical enough, but the association be-
tween impairment and appointment keeping in this study is
a meager .01 percent which is hardly significant. Impair-
ment should not be considered in predicting appointment
keeping behavior.

Researchers have examined variables related to the
therapeutic regimen to find clues to appointment keeping.
The effort has had limited success. Dodd (1971) found

that patients sustain higher rates of appointment keeping

125

when medication is prescribed. This study found only a .02
correlation. It, therefore, seems a waste of time to con-
centrate on those clients already on or who have been on

a drug therapy. Frank (1968) found that, initially, medi-
cation influenced appointment keeping but, over time, other
factors became significant for insuring continued fidelity
to appointments. Although this study found no significant
relationship between medication prescription and appointment
keeping, incentives given out at screening may have some
effect. Researchers feel that something tangible, even a
placebo, would be advisable to insure appointment keeping.
An incentive such as toothbrush, vitamins, food items, and
the like would be a sensible practice to try in spite of the
fact that a significant correlation was not found in this
study.

Cost, duration, and side effects have been found to be
significantly correlated negatively with appointment keep-
ing behavior (Rickels 1968, Winkelman 1964). The only item
in these three variables found to be significant is side
effects due to a treatment regimen. This study concurs with
other researchers concerning the negative correlation with
side effects. Increased cost of treatment has a respectable
.13 percent correlation with broken appointments, but this
is not high enough to be considered significant. Duration
of treatment achieved only .07 correlation with appointment

failures.

126

The literature available on patient-therapist inter-
action variables relative to appointment keeping has been
almost exclusively concerned with psychiatric services.

A cross study comparison of this study with the literature
is weak due to the very different situations. Alpert (1964)
and Haynes (1976) found a significant correlation between
satisfaction with visits to the doctor and appointment keep-
ing. Data in this present study showed an extremely small
(.02) correlation with satisfaction and screening attendance.
A similar low (.03) correlation existed between attendance
and education of the client given by the therapist. Glowgow
(1970), on the other hand, found that when the therapist
takes time to educate the patient concerning the helth pro-
blem, appointment keeping behavior improves. Tagliacozzo
(1974), as did this study, could not find evidence of this.

Meeting the expectations of the patient is significant-
ly correlated with appointment keeping according to Barry
(1984). He found that when time was taken to explain to
the patient exactly what to expect at the clinic, a signi—
ficantly higher number of patients kept their appointments.
This has not proven to be a significant factor in this study
as clinic attendance correlated only .01 percent with the
level of expectations being met.

A considerable number of studies focused on access to
the health care facility as a key factor. This present
study found no significant correlation with distance from

the patient's home to the clinic. In fact, what

127

correlation that existed is a negative association (-.12).
Thus the trend is: the closer the clinic is the less likely
an appointment will be kept. It could be that families
living at greater distances appreciate the service more
because health care services are not readily available.
This, of course, is guess work. Other researchers (Feister
1974, Bigner 1976), however, concur with this present study
concerning the existence of no correlation between distance
traveled and appointment keeping.

Very little study has been done on the availability of
transportation as a variable related to appointment keep—
ing. Alpert (1964) found a significant correlation. This
study showed a negative although insignificant correlation
(-.l4) between the availability of transportation and ap—
pointment keeping.

The presence of a telephone in the home is an important
means of access to the facility, not only to make appoint-
ments, initially, but, also, to change or cancel them. It
is, likewise, an important item for access from the clinic
to the family to remind the patient of a forthcoming appoint-
ment. All studies reviewed found the telephone to be a
critical element in a successful appointment keeping efforts.
This study concurs with other researchers on this issue.
With the breakup of the telephone system, and its accompany-
ing price increase, a greater number of homes of low-income
families are lcsing telephone services. This is not a

hOpeful trend for programs such as EPSDT.

128

Variables related to the facility and its administrative
procedures have been found to have an enormous impact on at—
tendance. The largest correlation found in this study per-
tain to the administrative process of scheduling appoint-
ments. It was found that a very high significant correla-
tion (.54) exists between appointment keeping and the number
of days that intervene between scheduling and the appoint-
ment. The longer the interval, the greater the number of
broken appointments. Other researchers (Hagerman 1978,
Nazarian 1974) found this association. This is especially
true with regard to follow-up appointments after mass health
screening (Glowgow 1973). It is consistent with a short
range perspective low-income families are often forced to
adopt because they do not have the resources for long range
planning. A relatively immediate appointment fits into
this style of living.

This study found two other variables to be significant
relative to facility/administrative process. These are:
the sufficiency of staff to provide services, and the fre-
quency of errors made in scheduling. Where there are ade-
quate number of staff persons to do the job and fewer errors
made, the more likely an appointment will be kept. It is,
consequently, counter productive to reduce staff because
scheduled patients are not keeping their appointments and,
as a result, more errors are made because the remaining
staff is put under greater stress in trying to carry the

load. The solution is to find more effective ways of

129

either getting the patient to come for the appointment or
taking the service to the patient. The use of home visits
or personal contact has been successfully used to minimize
broken appointments (Hildebrandt 1975). In at least some
settings, the provision of neighborhood clinics in familiar
surroundings with a highly involved staff has resulted in
reduction of broken appointments in comparison with pre-
viously used central facility (Curry 1968).

Although other researchers (Badgley 1961, Finnirty
(1973) found that waiting time was an important issue in
appointment keeping, this study is unable to corroborate
the finding. A correlation of only .04 existed between
waiting time and successful other scheduling outcome. To
increase professional staff utilization, Shonick and Klein
(1977) experimented with a "block time" scheduling system
where a number of patients are given the same time for an
appointment. This has the effect of extending the waiting
time for patients. The "block system" resulted in a marked-
ly improved appointment utilization level. The improved
utilization of appointment time is due to overbooking. Key
to this system is the capability to establish, with some
accuracy, a probability that any given patient will fail to
keep an appointment. Whether or not this can be done in a
rapid, cost effective manner remains to be established as
is evident from this present study. This study found a
small positive correlation with appointment keeping and

block scheduling (.15). Contrasted with this is a small

130

positive correlation (.12) scheduling success when a client
is referred to a specific doctor rather than to a clinic in
general. These correlations are not large enough to be
significant. This contradicts the findings of Rockart and
Hofmann (1961) who found appointment with a specific pro—
vider resulted not only in the patient being prompt for

the appointment, but the promptness of the provider was
also significantly improved. Finnirty (1973) found little
difference in appointment keeping patterns between low and
middle class income groups when personal attention was given
by way of providing a specific time with a specific pro—
vider. This study does not include middle income families,
but only the poorest of the poor. Findings regarding the
facility/administrative process, however, underscored the
need felt by low-income families for personal attention
rather herded in with the assumption that low-income pa-
tients break appointments more frequently than others due
to their ethnic background, low education levels, cultural
barriers, low harmony in family relationships, social dis—
organization, and other factors related to urban living.

A nearly universal practice in the Michigan EPSDT
program is calling or sending reminders to insure that
clients will keep their appointment. This present study
found only trivial relationship between prospective (.01)
and retrospective reminders (-.04). This practice is
among the most studied of the administrative procedures.

Levy (1977) reported a 20 percent reduction in appointment

131

failures through the use of prospective reminders. Mailed
reminders appear to be generally as effective as telephone
reminders. Extended use of prospective reminders appear

to be increasingly effective. The findings of this study
may reflect the result of using this practice over a period
of years and often with the same family. Use of retro-
spective reminders, i. e., notification of patients after
missed appointments to ascertain their reason for absence
and to offer a new appointment, has been ineffective
(Hurtado 1973).

At first thought, it would seem that people are great—
ly influenced by the weather in their behavior. Some
health workers report that on beautiful, sunny days atten-
dance at a health screening clinic is often very poor. In
contrast, on a day that a family would be expected to stay
home because of snow, cold, or rain, the clinic is often
filled to capacity. A number of medical social workers
have made this observation. One may conclude that weather
has an influence on the mood of individuals, but it is clear
that it does not determine behavior. This study, however,
found no such relationship (-.05) of significance between
appointment keeping and weather (barring severe conditions).
Other researchers (Badgley 1961, Hurtado 9173) found the
same.

In scheduling families for health screening, workers
often make Special effort to get larger families in be-

cause of the larger number of individuals involved.

132

According to Badgley (1961), large families miss their ap-
pointments much more frequently than smaller ones. There—
fore, the total number of screenings may not be improved
at all by concentrating on large families. This study
found virtually no (.03) correlation with family size and
appointment keeping habits. Alpert (1964) suggests that
presence of small children in the family are especially
relevant to broken appointments. No evidence supporting
this notion can be found in this present study.

The time of day or day in the week that an appointment
is scheduled has no meaningful correlation (-.02) with a
successful outcome. Oppenheim (1979) reported the same
finding. This finding should alert workers scheduling ap-
pointments that they cannot depend heavily on the influence
of a client's friends or neighbors to convince him/her of
the value of health screening. Each person is to be treated
individually and made to understand what is in health screen-
ing, and why it needs to be done. Furthermore, it does not
follow that, if a client is not attracted to a screening
clinic because of encouragement of friends or neighbors,
the client will also not be repealed by bad reports from
these same friends or neighbors.
Summary

Large industrial nations such as Britain, France, the
United States, and others are faced with an exploding cost
for health care. This is due to the increasing prevalence

of health problems because of increasing concentration of

133

people in urban areas, the environmental pollution resulting
from industrialization, stress due to rapid social change,
as well as other factors. Rapidly increasing costs are a
by-product of rapid advance in technologies that inevitably
accompanies the process of industrialization. New and
costly techniques and machines are invented that, at the
same time, raise expectations that improved health care

will be available to everyone.

Government, business, and the citizenry in general
are becoming increasingly aware that new technologies
cannot be translated immediately into better health services
to all. The "writing on the wall" is that there clearly
will be a limit to the availability of health services.
Increasingly, as unsavory as it may be, decisions will be
made as to who will continue in sickness or disability,
or be allowed to die, and who will be restored to a healthy
life. Already an estimated 20 percent of Americans have
little or no access to health care because they cannot
afford insurance. As pointed out in the Introduction of
this study, the incidence and prevalence of health problems
among the poor is much higher and often serves to lock in-
dividuals into a lifetime of poverty.

It was in this context that this study was undertaken.
Since focus on remedial health care is becoming increasingly
unrealistic, physically speaking, attention and available
resources must be redirected toward prevention. Health pro-

blems must be found early when treatment is more effective

134

and cheaper. It is precisely for this purpose that a
massive health screening program (EPSDT) was authorized

by Congress in 1967. The effort was directed at low—
income families as the most vulnerable segment of the popu—
lation. Out of the twelve million eligible, only two mil-
lion were screened (Dew 1979) in the early years of this
program.

In the many years of experience for this researcher,
when the health screening is explained adequately, most
families indicate a readiness to participate. However,
when the time for the screening arises, a large percent
(35 percent to 50 percent) fail to keep the appointment.
The problem of missed appointments is a major stumbling
block for impacting on the health status of millions of
eligible children. Not only does the client suffer be-
cause of undiscovered or unchecked health problems, but
also, the provider of care is hurt because the time set
aside for an appointment is not effectively utilized. It
is of critical importance that some light be cast on what
are the key factors that relate to keeping or not keeping
appointments.

This study has been an effort to provide this insight.
Of the eight clusters originally proposed as important
categories of variables to explain appointment keeping be-
havior, only some items in the cluster of variables related
to features of facility/administrative process, namely, an

adequate number of staff, specific time and specific

135

provider for the appointment were found to be significant.
The strongest correlation for appointment keeping was with
the time factor between scheduling and appointment. The
shorter the time, the better the outcome is. Generally
speaking, this study did not find much help for those seek-
ing answers by looking at a large array of variables in an
attempt to reduce broken appointments. Specific advice that
flows for this study is —- do not sit back and try to mani-
pulate people into the health screening clinc by prejudging
behavior according to theories or impressions regarding cer-
tain variables. Rather, get directly involved with the
families and do what needs to be done including educating/
directing the family but, also, change the system to meet
the needs. It seems clear to this researcher, both from
this study and from years of experience, people want per-
sonal and immediate attention. They want follow-up care
with continuity from health care workers that are profession-
al and in adequate numbers. Where families find this,
response will not be lacking.

Although there are a multitude of studies focused on
the appointment keeping problems, certain cautions should
be kept in mind in making comparisons of this study with
others. Bearing these cautions in mind, the reader will
be in a better position to interpret constructive findings
of this study versus those conducted by other researchers.

Most studies interested in appointment keeping be-

havior have been concerned about persons with psychiatric,

136

alcoholic, or drug related problems. People with such pro-
blems usually need to go through a long series of appoint-
ments to treat their chronic condition. Researchers are
anxious to discover, not only the variables affecting ini-
tial treatment of the health problem, but, often, what
causes the patient to drift away or stop treatment alto-
gether, expressed through the behavior of missing appoint-
ments. This study is concerned with the behavior of keeping
an initial appointment for health screening. That may or
may not result in multiple follow-up appointments for health
care due to problems uncovered in the screening tests. It
is very likely that this type of appointment cannot be equa-
ted with appointments that are serial in nature.

The subjects of this present study are the poorest of
the poor. As explained in the section on Methodology, the
income of families must be minimal to qualify for a meager
welfare grant. Many families receiving higher incomes are
not eligible for assistance but are still very much living
in poverty with little or no resources available for health
services. In contrast, most studies are conducted on a
cross section of the population in general. There may or
may not be a significant contrast between the behavior of
very low, low, and middle class patients relative to health
care. The contrasting findings of this study with others
may, in fact, be reflecting significantly different popu-

lations from which samples are drawn for study.

137

A third very important caution is to keep in mind that
the appointments for health screening studied are, basi-
cally, a preventive health care measure. We live in a
society that has traditionally practiced episodic health
care. When a problem becomes apparent to an individual,
then and only then does that individual make an effort to
seek out diagnosis and follow-up health care. Prevention
is comparatively new and few people, including educated and
of considerable income, are tuned into it. Knowledge of
prevention is not automatically translated into changed
behavior.

In this environment, the need for a comprehensive health
screening as a preventive measure is barely accepted. Agree-
ment for an appointment may be only one of convenience to
get a person off the telephone or away from the door step.

It is difficult, at best, to make an across the board com-
parison between appointment keeping between those who have
known, and often painfully felt health problems, and those
who have no inkling of a health condition. Corresponding

the lack of prevention in practice in the general population
is the fact that the people included in this study are forced
to live on a day-to-day survival level and, consequently,

are even less conditioned to the notion and practice of
prevention against some future possible health problem.

Finally, given this situation just explained, the
initiative for an appointment for a health screening is

nearly always taken by someone other than the family.

138

Usually this is done by by a field worker for the health
clinic. The client must be convinced of the value of the
health screening, and the response is, perhaps, based more
often to please the worker than on conviction as to the
worth of the screening procedures. In appointment keeping
studies, generally, the sample is shown from those who, for
the most part, take initiative on their own to contact the
doctor and, consequently, are subject to a minimal pressure
to keep an appointment.

The cautions, just discussed, give a clue about what
should be done in future studies. There may be important
differences between initial appointments and serial appoint-
ments for health care. Some attention (Walsh 1967) has been
given to appointment keeping for a variety of health ser-
vices, but the variable of initial versus serial appoint-
ment was not considered. This issue is a critical one in
a health screening program such as EPSDT which operates on
one time only appointments followed by a repeat two or more
years later.

More attention needs to be given on the issue of income
as determinant of behavior relative to appointment keeping,
specifically, or health behavior generally. Do we, in fact,
tend to blame the victim who, in this case, is poor because
he/she supposedly is irresponsive? Neglecting health care
is only evidence of this lack of responsibility. If those
better off behaved in the same irresponsible way, it is

because they are too busy to attend to everything. Are

139

there, in fact, differences in health related behavior in
the low and middle classes, and what are the prevailing
assumptions about such behavior?

More study needs to be conducted on preventive related
health behavior. This is especially urgent because, as
noted earlier, costs for health care is exploding along
with curative technology. Most health problems are not
from disease causing microorganisms, but stem from environ-
ent and behavior over which we have direct or indirect
control. Both knowledge of prevention and translation
of the knowledge into practice are important issues for
the future.

Most of the variables related to appointment keeping
included in this study have been less than helpful in
predicting scheduling outcomes. After turning over so
many stones, it is impossible not to notice issues that
are interesting, to say the least, and inviting for future
research. A review of some of these issues will be illus-
trative.

Stine (1968) found that the level of education is
significantly correlated with appointment keeping. In this
present study, no such relationship is to be found. At
issue here is the question: Does higher levels of educa-
tion inevitably lead to improvement in behavior in general,
or in practice of preventive health care specifically?
Information on this issue would provide some guide as to the

approach that should or should not be used for inducing

140

behavioral change regarding health matters.

Similar to the education issue is the leve of income.
Alpert (1964) found a significant correlation with the
level of income and appointment keeping. This present study
did not find this relationship. More study is needed on the
matter to find at what level of income no difference of be-
havior is detected if, indeed, there is a relationship.
Would low-income people, in fact, use additional money for
health improvement? It should be noted that a large per—
cent (35 percent to 50 percent) do not keep appointments
when, in fact, there are no out-of-pocket expense to the
family. Level of income, therefore, would not seem to be
a vital issue in this situation.

There is sharp division among researchers regarding
the significance of ethnic origin and appointment keeping.
Unfortunately, nothing can be learned from this study be-
cause only 3 percent of the sample was Black. Further,
research is needed to resolve the ethnic factor.

There are many studies on the demographic variables
and their association with appointment keeping, mainly
because these are easy to study. More attention should be
given to the socio—behavioral variables. These are diffi-
cult to study because they are very hard to measure in an
objective way. Furthermore, there does not appear to be a
clear connection between belief and behavioral patterns.

More information needs to be provided on this vital link.

141

In research alone on characteristics of the provider
and appointment keeping, the most puzzling finding is the
negative association between years on the job and success-
ful outcome of scheduling. This may reflect the presence
of burn out on a job that has a lot of built-in frustra-
tions. As an example, the worker is required, as a premise
of employment, to reach a certain quota of screenings per
month but the family is free to accept or refuse the program.
Furthermore, even after accepting, there is no penalty for
not keeping the appointment. Adding to this frustration
is the plight of the worker who, in some cases, have no
health insurance and, consequently, cannot avail themselves
of medical and dental services. Another source of frus-
tration results from the patronizing service low-income
families are often accorded by professionals in the health
care system who, sometimes, view low-income people as ir-
responsible. Workers scheduling families with such experi-
ence from professional health care workers receive the
brunt of the reaction. The scheduler is caught in a con-
flict over which he/she has no control. Further study is
needed to design mass screening programs to reduce the
number of built-in frustrations. Furthermore, a study
should be conducted on how tensions that cannot be elimina-
ted can be, at least, minimized so that burn out does not
occur.

Dodd (1971) found that prescription of a medication

results in a higher return of kept follow-up appointments.

142

While this study did not support this, more study should
be done on the use of incentives other than medicine to
insure fidelity to appointments. It would be most useful
to know what positive incentives, in terms of the amount
and method of distribution, would be most effective to
insure the success of a mass screening program.

Most of the available literature regarding patient-
therapist interaction is concerned only with psychiatric
services. This is understandable due to both the length
and complexity involved in a successful psychiatric treat—
ment regimen. Encounters with medical providers are,
generally, brief and, often, quite impersonal. This pre-
sent study does not find any significant correlation be-
tween variables related to the patient-therapist inter-
action and appointment keeping with one exception. It
was found that, when a provider did close personal follow—
up on a patient, that patient is more likely to keep his/
her future appointments. More study should be done on
variables related to patient-therapist interaction variables
in situations other than for psychiatric services. This
is an area that is still relatively unknown but of great
importance in guiding persons working in the health care
system.

This study found its highest correlation with appoint-
ment keeping between the brevity of time between schedules
and the appointment. Furthermore, sufficiency of staff and

accuracy of scheduling were also important. These findings

143

point to the importance of providing immediate and pro-
fessional service to insure compliance. A study should
be conducted as to how a clinic based service, such as
health screening, can be taken and the families for whom
it is intended so that the service can be immediately
available. As an example, experimental mobile clinics
can be set up in neighborhoodsto reduce access barriers
and provide more direct and personal attendance to eli-
gible families.

While attempts have been made to measure the influence
of 80 variables, an appointment keeping for health screening
among low-income families, the number of variables that
could be studied is by no means exhausted. The number of
influences that impact on a person's daily life are poten-
tially limitless. The question that needs to be asked is,
what practical gain can there be by expanding the variables
to an increasing number?

Suggestion was made earlier that, in measuring deter—
minants of appointment keeping behavior, further investi-
gation may be useful in such areas as serial appointments,
preventive versus curative behavior, medical versus psy-
chiatric appointments, and the like. However, one finding
in this study should be kept clearly in mind, namely, no
significant correlation exists between attitudes, behaviors,
and perceptions. This means, in effect, that researchers
are headed in the wrong direction to investigate attitudes

to predict behavior, to behavior to predict perceptions.

144

More concretely for administrators of health care programs,
effort aimed at attitude or perception change in order to
achieve a more desirable appointment keeping behavior will
probably be ineffective in achieving this goal. Appoint-
ment keeping behavior demands a multivariete approach that
deals with attitude, perception, and behavior. Central
concentration, if improved appointment keeping is the goal,
should be on behavioral changing issued directly.

This finding is in agreement with Leedom's (1980)
research on energy conservation behavior. This research
involved giving extensive information about the importance
of energy conservation in order to change the attitude of
the individuals involved. The researcher hypothesized
that, with a better knowledge concerning the consequences
of wasteful use of energy, change of behavior toward one
of conservation would surely follow. This did not occur.
Those directly involved in actitivities of energy conser-
vation, however, did, in fact, change their behavior in
this direction.

Lounsbury (1972) had similar findings. A study was
designed to see if attendants in two mental retardation
training schools could plan job training programs for
other attendants and whether attitude, personal history,
and job satisfaction variables were related to planning
activity. Lounsbury found training suggestion ratings
were not related to tenure and, similarily, not related

to attitude or job satisfaction.

145

Behavior, attitude, and perception are not only un-
related to each other but, also, unrelated as a variable
in different social settings. Thus, the behavior of an
individual may be changed through a program of activities
so that behavior regarding medical appointments greatly
improves. It does not follow that this person will also
improve in appointment keeping for school, jobs, or other
types of activities. Furthermore, if a person's attitude
toward a doctor is improved, it does not necessarily follow
that his/her attitude toward self or other professional
people will change. Tucker (1974) documented this in a
study on Methods of Teaching. Where attitudes of students
changed, they did so only in the specific area of attitudes
towards the teacher. The attitudes of the student toward
him/herself or towards other aspects of the environment
were not significantly affected. Furthermore, Tucker (1974)
confirmed the findings of Leedom and Lounsbury regarding
the correlation of behavior and attitude. Students who
improved their performance in studies did not change their
attitude.

Similar findings were reported by Fairweather (1964,
1969) in studies of community placement of persons with
mental illness. He found that, although participants of
his study indicated satisfaction with a community program,
a high level of performance did not follow, at least, in
regard to returning to the hospital and sustained employ-

ment. Patients expressed gratification at being in the

146

community, but these attitudes were unrelated to whether or
not they remained there or were employed (Fairweather 1969,
p. 276).

Fairweather (1964, p. 278) found that adjustment to
the community is relatively independent of attitudes about
hospital personnel and treatment programs. He cites the
example where the behavior of an inmate assigned to a laun-
dry job in a hospital is not predictive of the person's
behavior on a laundry job in the community. He found that
the relative independence of hospital behavior, perception,
and community adjustment is attributable, somewhat, to very
different situations,namely, to a shift from a member of
the majority in the hospital to a minority in the community,
and to the fact that attitudes, fantasies, and behaviors,
even with the same situation, are, at least, only marginally
related (Fairweather, 1969, p. 282).

Perhaps the wrong questions are being asked. A focus
on appointment keeping is frequently done because it is an
obvious aspect of a health care practice; it involves fin-
ancial loss as well as possible damage to the patient's
health. In conducting a preventive program such as health
screening, which in Michigan costs eleven to twelve million
dollars a year, the focus should go beyond whether or not a
person keeps an appointment to look at whether, over time,
the screening did in fact prevent damage to the person's
health and decrease the escalating cost of health care

services. The national health screening program, namely,

147

EPSDT, has been in existence for over ten years. To date,
no comprehensive study has been made to measure the impact
of the program on the health and well being of those who
in fact did keep their appointments for comprehensive

screening.

APPENDICES

APPENDIX A

Letter of Invitation to Participate in
Health Screening and the Survey

Dear

 

Your children are eligible for a complete health
screening that includes vision and hearing tests, nutri-
tional assessment. tests for diabetes, and a comprehensive
physical exam — to name some of the screening tests. These
tests are conducted at no cost to you.

We are anxious to make these screening services as
convenient as possible for you and other families. I
would also like to invite you to give your opinion in an
Opinion Survey about health services which you have re-
ceived. Your responses will help us improve our services.

The opinions you express in our survey are completely
confidential. They will not be shared with the Department
of Social Services and will have no effect on the assist-
ance you are now getting.

It is especially important that I hear from you,
because we are asking the opinions of only a few families
and feel that the opinions of many families will be reflected.

Your help is important to us in this survey and is
greatly appreciated.

Sincerely,

Health Screening
Coordinator

148

10.

11.

12.

APPENDIX B

Client Questionnaire

What is your age group?

a) 13-17 years b) 18-24 years c) 25-34 years
d) 35-44 years e) 45 years or older
Your gender? a) Male b) Female

In addition to being an American, what is your ethnic
group or ancestry?

a) White b) Black c) Spanish American
d) Asian e) Other

What is your religion?

a) Protestant b) Catholic c) Jewish

d) Other e) None

Do you speak or understand a language other than English?
a) Yes b) No

Were you born in Michigan? 3) Yes b) No

The number of persons in your family living in this
household is?
a) 1-2 b) 3-4 c) 5-6 d) 7-8 e) 9 or more

The oldest child in this family living in this house-
hold is?

a) 0-3 years b) 3-6 years c) 7-9 years

d) 10-12 years e) 13 years or older

Your appointment with the doctor generally is?
a) Before 10 A.M. b) 10 A.M.-12 P.M. c) 12:01-2 P.M.

d) 2:01-4 P.M. e) 4:01 P.M. or later

Your appointment with the doctor is generally on?
a) Monday b) Tuesday c) Wednesday
d) Thursday e) Friday or weekends

What is the last grade of school you completed?
a) 8th grade or less b) Some high school

c) Graduated high school d) Some college

e) Graduated college

Which of the following income groups represents your
highest income for any year?

a) Under $5,000 b) $5,000-9,999 c) $10,000-l4,000
d) $15,000-19,999 e) $20,000 and over

149

l3.

14.

15.

l6.

l7.

l8.

19.

20.

21.

22.

23.

150

Which of the following income represents the income of
your family for 1984?

a) Under $5,000 b) $5,000-9,999 c) $10,000-14,999
d) $15,000-l9,999 e) $20,000 and over

The occupation of your father was that of a?
a) Unskilled worker b) Semi-skilled worker
c) Skilled Worker

You have had the same doctor for?

a) Less than 2 years b) 2 years but less than 4 years
b) 4 years but less than 6 years d) 6 years but less
than 8 years e) 8 years or more

You go to a medical clinic for yourself or with your
child(ren)?

a) Once a year b) Twice a year c) 3 times a year
d) 4 or 5 times a year e) 6 times or more a year

Which of the following items describes the place where
you live?

a) Within the city/village limit b) Within 1 mile from
the city/village limit c) Between 1 and 5 miles from
the city/village limit d) Between 6 and 10 miles from
the city/village limit e) More than 11 miles from the
city/village limit

The number of miles from your house to the health
screening clinic is?

a) Less than 1 mile b) 1 mile but less than 2

c) 2 miles but less than 4 d) 4 miles but less than
8 e) 8 miles or more

You have been on public assistance for?

a) Less than 1 year b) 1 year but less than 2

c) 2 years but less than 3 years d) 4 years but less
than 5 e) 5 years or more

Transportation is unavailable to you to go to the
medical clinic?
a) Always b) Often c) Sometimes d) Rarely e) Never

You have no telephone in your home?
a) Always b) Often c) Sometimes d) Rarely e) Never

Your doctor does not prescribe medication when you
visit him/her?
a) Always b) Often c) Sometimes d) Rarely e) Never

Your doctor does not put you on a diet when you visit
him/her?
a) Always b) Often c) Sometimes d) Rarely e) Never

24.

25.

151

You miss appointments with the doctor?
a) Always b) Often c) Sometimes d) Rarely

You avoid the health department for services?
a) Always b) Often c) Sometimes d) Rarely

e) Never

e) Never

10.

11.

12.

13.

APPENDIX C

Client's Opinion Questionnaire

You tend to stay well and not get sick.
3) Always b) Often c) Sometimes d) Rarely e) Never

Most sicknesses do not seriously affect a person's
health and well being.
a) Always b) Often c) Sometimes d) Rarely e) Never

You do not tend to get a particular sickness, for
example, heart problems, cancer, diabetes, or the
like.

a) Always b) Often c) Sometimes d) Rarely e) Never

Treatment for most sicknesses are effective.
a) Always b) Often c) Sometimes d) Rarely e) Never

Treatment used to relieve sickness is not painful.
a) Always b) Often c) Sometimes d) Rarely e) Never

God rewards those who do good with good health.
a) Always b) Often c) Sometimes d) Rarely e) Never

Doctors have ability to heal.
a) Always b) Often c) Sometimes d) Rarely e) Never

You have an above knowledge of sickness people get
and how to care for them.
a) Always b) Often c) Sometimes d) Rarely e) Never

You have an above average intelligence.
a) Always b) Often c) Sometimes d) Rarely e) Never

Health care professionals are friendly and capable.
a) Always b) Often c) Sometimes d) Rarely e) Never

In the treatment of your health problems, drugs are
needed.
a) Always b) Often c) Sometimes d) Rarely e) Never

You are head of the household.
3) Always b) Often c) Sometimes d) Rarely e) Never

It is important to have such items as a thermometer

to keep check on the state of your health and that

of your children.

a) Always b) Often c) Sometimes d) Rarely e) Never

152

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

153

The doctor says sickness you have had were due to
physical causes and not to mental or emotional stress.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your past sicknesses have not been severe.
a) Always b) Often c) Sometimes d) Rarely e) Never

You have received treatment for a health problem in
the past.
a) Always b) Often c) Sometimes d) Rarely e) Never

The doctor finds health problems that you did not know
you had.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your health problem improves after you do what your
doctor says.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your past health problems have not required immediate
attention from the doctor.
a) Always b) Often c) Sometimes d) Rarely e) Never

Missing an appointment with the doctor would be serious.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your health problems, in the past, made it hard for
you to get around.
a) Always b) Often c) Sometimes d) Rarely e) Never

The doctor prescribes weak medication for your health
problem.
a) Always b) Often c) Sometimes d) Rarely e) Never

The treatment prescribed by your doctor lasts a short
time.
a) Always b) Often c) Sometimes d) Rarely e) Never

The cost of visiting your doctor is low.
a) Always b) Often c) Sometimes d) Rarely e) Never

Treatment ordered by your doctor has no side effects.
a) Always b) Often c) Sometimes d) Rarely e) Never

The dosage of medication prescribed by your doctor is
low.
a) Always b) Often c) Sometimes d) Rarely e) Never

In following the treatment ordered by your doctor,
there is little change of your routine.
a) Always b) Often c) Sometimes d) Rarely e) Never

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

154

You have been satisfied with the medical clinics you
have visited.
3) Always b) Often c) Sometimes d) Rarely e) Never

Your expectations were met when you made a visit to a
medical clinic.
a) Always b) Often c) Sometimes d) Rarely e) Never

You can get all the medical services you need when
you visit the medical clinic you attend.
a) Always b) Often c) Sometimes d) Rarely e) Never

The doctor takes time to explain to you what he does,
and why he wants you to do certain things.
a) Always b) Often c) Sometimes d) Rarely e) Never

The doctor keeps a close check on you in regard to
following his directions.
a) Always b) Often c) Sometimes d) Rarely e) Never

Even though Medicaid pays for health services, there
is no cost to you.
a) Always b) Often c) Sometimes d) Rarely e) Never

When you go to your medical clinic, you wait a short
time for services.
3) Always b) Often c) Sometimes d) Rarely e) Never

Recommendation from family or friends is important to
selecting a clinic for yourself.
a) Always b) Often c) Sometimes d) Rarely e) Never

You much prefer being sent to a specific doctor rather
than merely sent to a clinic for services.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your doctor has enough staff to adequately serve all
patients that come.
3) Always b) Often c) Sometimes d) Rarely e) Never

Your doctor sets a specific time to see you exclusively.
a) Always b) Often c) Sometimes d) Rarely e) Never

Parking close to the medical clinic has been available.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your doctor calls or sends notices to remind you of
your appointment.
a) Always b) Often c) Sometimes d) Rarely e) Never

41.

42.

43.

44.

45.

46.

47.

155

Your doctor calls or sends notices when you miss an
appointment.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your doctor does not make a mistake in scheduling
your appointment.
a) Always b) Often c) Sometimes d) Rarely e) Never

If the weather is bad, you would still keep a medical
appointment.
a) Always b) Often c) Sometimes d) Rarely e) Never

The influence of family and friends is important to
you in keeping your appointment with the doctor.
a) Always b) Often c) Sometimes d) Rarely e) Never

Your sicknesses have lasted a short time.
a) Always b) Often c) Sometimes d) Rarely e) Never

You go to the doctor when there is a sickness in your
family.
a) Always b) Often c) Sometimes d) Rarely e) Never

If your children are not doing well in school, you take
them to your doctor.
a) Always b) Often c) Sometimes d) Rarely e) Never

10.

11.

APPENDIX D

Medical Outreach Worker Questionnaire

Medical Outreach Worker's age?
a) 18-24 years b) 25-34 years c) 35-44 years
d) 45-54 years e) 55 years or more

Number of years since high school graduation?

a) 0 to less than 1 year b) 1 to less than 2 years
c) 2 to less than 3 years d) 4 to less than 5 years
e) 5 years or more

Medical Outreach Worker's gender? a) Male b) Female

Ethnic origin of medical social worker is?
a) White b) Black c) Spanish American d) Asian
e) Other

Medical Outreach Worker's years of outreach work?

a) 0 to less than 1 year b) l to less than 2 years
c) 2 to less than 3 years c) 4 to less than 5 years
e) 5 years or more

The number of families I am given each month to contact
on an average is?

a) 0-99 b) 100-149 c) 150-199 d) 200-299

e) 300 or more

I feel that EPSDT should be required for families to
receive their welfare check?
a) Always b) Often c) Sometimes d) Rarely e) Never

The number of days from the date of scheduling this
appointment to and including the day of the appointment
is?

a) 5 days or less b) 6 to 10 days c) 11 to 14 days
d) 15 days or more e) No appointment needed

Outreach Outcome — The Client:

a) Refused an appointment for screening.

b) Accepted an appointment but did not keep it without
previous canceling or changing it.

c) Accepted an appointment but called to cancel or
change it.

d) Accepted an appointment and came in on the day of
the appointment.

Date of Interview:

 

Date of Appointment:

 

156

12.

13.

Time of Appointment:

My region is:
a) Delta-Menominee
c) Barry-Eaton

157

 

b) Grand Traverse
d) Monroe e) St.

Clair

BIBLIOGRAPHY

BIBLIOGRAPHY

Adler, L. M., Goin M. Yammamoto, J., Failed Psychiatric Clinic
Appointments. California Medicine, 1963, 99:388.

 

Alpert, J. J., Broken Appointments. Pediatrics, 1964, 34:127

 

Armstrong, M. L., Bakke, J. L., Dodge, H., Conrad, L. L.,
Freis, E. D., Fremont, R. E., Kirkendall, W. M., Pilz, C.

G., Ramierz, E. A., Richardson, D. W., and Williams, J. H.,
Double Blind Control Study of Hyptertensive Agents. Archives
of Internal Medicine, 1962, 110, 222-239.

 

Badgley, R. F., Furnal MA. Appointment Breaking in a Ped-
iatric Clinic. Yale Journal of Biological Medicine, 1961,
34:117.

 

Baekeland, F., Lundwall, 1., Dropping Out of Treatment: A
Critical Review. Psychology Bulletin, 1975, 82:738.

 

Baekeland, F., Lundwall, L., Dropping Out of Treatment: A
Critical Review. Psychology Bulletin, 1975, 82:738.

 

Barry, Daniels, Effecting Change in Outpatient Appointments.
Journal of Family Practice, May, 1984, 18(5), p. 739-42.

 

Baum, O. E., Felzer, S. B., D'Zmura, T. L., and Shumaker, E.
Psychotherapy, Dropouts and Lower Socioeconomic Patients.
American Journal of Orthopsychiatry, 1966, 36, 629-635.

 

Becker, M. H., Maiman, L. A., Kirscht, J. P., et al. A Test
of the Health Belief Model in Obesity. Journal of Health
Social Behavior, 1977, 18:348.

 

 

Becker, M. H., Nathanson, C. A., Drachman, R. H., and Kirscht,
J. P., Mother's Health Beliefs and Children's Clinic Visits.
Journal of Community Health, 1977, 3:125.

 

Berne, E., Group Astendance: Clinic and Theoretical Consider—
ations. International Journal of Group Psychotherapy, 1955,
5, 392—402.

 

Bessman, A. N., Comparison of Medical Care in Nurse Clinician
and Physician Clinics in Medical School Affiliated Hospitals.
Journal of Chronic Disease, 1974, 27:115.

 

158

159

Bigger, J. F., A Comparison of Patient Compliance in Treat-
ed vs Untreated Occular Hypertension. Ophthalmology, 1976,
81:277.

 

Birch, Herbert G., and Gussow, Joan Dye, Disadvantaged
Children and School Failure, New York, Harcourt, Brace, and
World, 1970, p. 243.

 

 

Blane, H. T., & Meyers, W. R., Social Class and the Estab-
lishment of Treatment Relations by Alcoholics. Journal of
Clinical Psychology, 1964, 20, 287-290.

 

Bowen, W. T., & Androes, A Follow-up Study of 79 Alcoholic
Patients: 1963-1965. Bulletin of the Menninger Clinic,
1968, 32, 26-34.

 

Burnip, R. Erickson R., Barr, G. D., et al, Well-Child Care
by Pediatric Nurse Practictioners in a Large Group Practice:
A Controlled Study in 1,152 Preschool Children. American
Journal of Disadvantaged Child, 1976, 130:51.

 

Caldwell, J. R., Cobb, 8., Dowling, M. D., et al., The
Dropout Problem in Antihypertensive Treatment. Journal
of Chronic Diseases, 1970, 22:579.

 

Camp, Bonnie W., Pediatric Screening Tests. Charles C.
Thomas Publisher, Springfield, IL, 1957, p. 9.

Crag, T. J., Huffine, C. L., Correlates of Patient Attendance
in an Inner City Mental Health Clinic. American Journal of
Psychiatry, 1976, 133:61.

Davis, Karen and Schoen, Cathy, Health and the War on Pover-
ty: A Ten Year Appraisal; Washington, D. C. The Urban
Institute, 1972.

 

Delk, J. L., Johnson, W. E., Treatment Continuers and Dis-
continuers in an Adult Oupatient Psychiatry Clinic. Journal
of Arkansas Medical Society, 1975, 72(2):92.

Demographic Yearbook of the United States, New York, 1973.

Dervin, J. V., Stone, D. L., Beck, C. H., The No-Show Patient
in the Model Family Practice Unit. Journal of Family Prac-
tice, 1978, 7:1177.

Dinnen, A., Change of Therapists as a Cause of Absence From
Group Psychotherapy. British Journal of Psychiatry, 1971,
119, 625-628.

160

Dodd, J., A Retrospective Analysis of Variables Related
to Duration of Treatment in a University Psychiatric Clinic.
Journal of Nervous and Mental Disorders, 1971, 151:75.

 

Endicott, N. A., & Endicott, J., "Improvement" in Untreated

Pscyhiatric Patients. Archives of General Psychiatry, 1963,
9, 575-585.

Fairweather, G. W., Social Psychology in Treating Mental
Illness. University Microfilms Limited, High Wycomb, Eng-
land, 1964, p. 211.

Fairweather, G. W., Community Life for the Mentally Ill.
Aldine Publishing, Chicago, IL, 1969, p. 262.

 

Fairweather, G. W., Experiemental Methods for Social Policy
Research. Pergannon Press, New York, 1977, p. 353.

Fiester, A. R., Mahrer, A. R., Giambra, L. M., et al,
Shaping a Clinic Population: The Dropout Problem ReconSi-
dered. Journal of Communigy Mental Health, 1974, 10:173.

 

Fiester, A. R., Rudestam, K. E., A Multivariate Analysis
of the Early Dropout Process. Journal of Consulting
Clinical Psychology, 1975, 43:528.

 

Foltz, Anne-Marie. An Ounce of Prevention, MIT Press,
Cambridge, MA, 1982, p. l.

 

Frank, J. D., The Influence of Patients' and Therapists'
Expectations on the Outcome of Psychotherapy. British
Journal of Medical Psychology, 1968, 41:349-356.

Frankenburg.William R., and Camp, Bonnie. Pediatric
Screening Tests. Charles C. Thomas Publisher, Springfield,
IL, 1978, p. 10.

 

Gallagher, E. B., & Kanter, S. S., The Duration of Out-
Patient Psychotherapy. Psychiatric anrterly Supplement,
1961, 35, 312-331.

Gates, S. J., Dolborn, D. K., Lowering Appointment Failures
in a Neighborhood Health Center. Medical Care, 1976,
14:263.

 

Genzberg, Eli, and Solow, Robert M., The Great Sociegy:
Lessons for the Future, New York, Basic Books, 1974.

Click, B. S., Dropout in an Outpatient Doubleblind Study.
Psychosomatics, 1965, 6:44.

 

161

Glowgow, E., Effects of Health Education Methods on Appoint-
ment Breaking. Public Health Repprt, 1970, 85:441.

 

Glowgow, E., Noncompliance - A Dilema. Sight Saver Report,
1973, 43-29.

 

Gottlieb, S., Kramer, H., Compliance With Recommendations
Following Executive Health Examinations. Journal of Occupa-
tional Medicine, 1962, 4:709.

 

 

Gray, R. M., Kesler, J. P., Moody, P. M., Effects of Social
Class and Friend's Expectations on Oral Polio Vaccination
Participation. American Journal of Public Health, 1966,
56:2028.

 

Green, L. W., The Potential of Health Education Includes
Cost-Effectiveness. Hospitals, 1976, 50(9):57.

 

Harr, Charles M., Between the Idea and the Reality: A Study
in the Origigy Fate and Legacy of the Model Cities Program.
Boston, Little, Brown and Company, 1975.

 

Haefner, D. P., Kirscht, J. P., Motivational and Behavioral
Effects of Modifying Health Beliefs. Public Health Reporpy
1970, 85:478.

 

Hagerman, G. A., Testing of the Mailed Appointment Reminder
in Family Practice. Journal of Family Practice, 1978, 7:199.

 

Hansen, A. C., Broken Appointments in a Child Health Confer-
ence. Nurses Outlook, 1953, 1:417.

 

Haynes, R. B., A Critical Review of the "Determinants" of
Patient Compliance With Therapeutic Regimens. In: Sackett,
D. L., Haynes, R. B., eds. Compliance With Therapeutic
Regimens. Baltimore, Johns HOpkins University Press, 1976.

 

 

 

Healy, W., and Breener, A., The Child Guidance Clinic: Birth
and Growth of an Idea. Orthopsychiatryi 1948.

 

Hertz, P., Stamps, P. L., Appointment Keeping Behavior Re-
evaluated. American Journal of Public Health, 1977, 67:1033.

 

Hildebrandt, D. E., Savis, J. M., Home Visits: A Method of
Reducing the Preintake Dropout Rate. Journal of Psychiatric

 

Nursing, 1975, 13(5);43.

Hoen-Sarie, R., Frank, J. D., Imber, S. D., et a1, Systema-
tic Preparation of Patients for Psychotherapy, and Effects
on Therapy Behavior and Outcome. Journal of Psychiatric
Res., 1964, 2:267.

 

 

162

Hoen-Sarie, R., Frank J. D., Imber, S. D., et al, Systema-
tic Preparation of Patients for Psychotherapy, and Effects
on Therapy Behavior and Outcome. Journal of Psychiatric
Rg§., 1964, 2:267.

 

Hoenig, J., Raff, N., The Non-attending Psychiatric Out-
patient - An Administrative Problem. Medical Care, 1966, 4:96.

 

Hofmann, P. B., Rockart, J. F., Implications of the No-show
Rate for Scheduling OPD Appointments. Hospital Program,
1969, 50:35.

 

Howard, K., Rickels, K., Mock, J. E., Lipman, R. S., Covi,
L., and Bauman, N. C. Therapeutic Style and Attrition Rate
From Psychiatric Drug Treatment. Journal of Nervous and
Mental Disease, 1970, 150, 102—110.

 

 

Howard, K. I., Orlinsky, D. E., and Hille, J. A., The Pa—
tient's Experience of Psychotherapy: Some Dimensions and
Determinants. Multivariate Behavioral Research, 1968,
Special Issue, 55-72.

 

Hurley, Roger, ngerty and Mental Retardation: A Casual
Relationship. New York, Random House, 1969, p. 134.

 

 

Hurtado, A. V., Greenlick, M. R., Colombo, T. J., Deter-
minants of Medical Care Utilization: Failure to Keep Ap-
pointments. Mgdical Care, 1973, 11:180.

 

Johnson, W. L., Rosenfeld, L. 8., Factors Affecting Waiting
Time in Ambulatory Care Services. Health Services Research,
1968, 3:286.

 

Jonas, 8., Appointment Breaking in a General Medical Clinic.
Medical Care, 1971, 9:82.

 

Karier, C., Testing for Order and Control in the Corporate
Liberal State. Educational Theory, 1972, 22 (Spring):155-l80.

 

Kegles, S. 8., Why People Seek Dental Care: A Test of Con-
ceptual Formulation. Journal of Health Human Behavior, 1973,
4:166.

 

Leedom, Nancy J., Energy Conservation Educapion: A Task-
Oriented Approach, Master's Thesis, Michigan State Univer—
sity, 1980.

 

 

Linley, Jayne., Mother's Attitude Regarding Health Care for
Their Children. AMCN, 1984, Jan-Feb., Vol. 9:1, p. 37-39.

Litt, Iris, Cuskey, Walter, Satisfaction With Health Care.
Journal of Adolescent Health Care, 1984, 5:196-200.

 

163

Lounsbury, Karen M., Survey of Attitudes, Job Satisfaction
and Training Suggestions of Attendants in Mental Retardation
Training_Schools, Dissertation, Michigan State University,
1972, p. 42.

 

 

 

Moyniham, Daniel P., Maximum Feasible Misunderstanding. New
York: The Free Press, 1969.

 

Michigan EPSDT Annual Report, Michigan Department of Public
Health, Lansing, p. 2.

 

Morris, Peter and Martin, Reis., Dilemmas of Social Reform,
New York: Atherton Press, 1969.

 

National Center for Health Statistics, Vital Statistics of
the United States. HEW, Washington, D.C.: U. S. Printing
Office, Volume W., Part A, 1974.

 

 

National Center for Health Statistics, Monthly Vital Statis-
tics Report Fund, Mortality Statistics. HEW, Washington, D.C.
U. 8., Printing Office, Volume 23, No. 11, 1975.

 

 

National Committee for Mental Hygiene, 1912, Origins, Ob-
jects and Plans. Publication No. 1, New York, p. 20.

 

Nazarian, L. F., Mechaber, J., Charney, E., et al: Effect
of Mailed Appointment Reminder on Appointment Keeping.
Pediatrics, 1974, 53:349.

 

Newman, J., and Sparer, P. J., Verbal Behavior and Other
Correlates of Adjustment to Hospitalization Among Tuber-
culosis Patients. In: P. J. Sparer, Ed, Personality, Stress
and Tuberculosis. New York: International Unversities
Press, 1956

 

 

Oppenheim, G. L., Bergman, J. J., English, E. C., Failed
Appointments: A Review. Journal of Family Practice, 1979,
8:789.

 

Overall, B., and Aronson, H., Expectations of Psychotherapy
in Patients of Lower Socioeconomic Class. American Journal
of Orthopsychiatry, 1963, 33, 421-430.

 

 

Pam, A., Bryskin, L., Rachlin, S., and Rosenblatt, A.,
Community Adjustment of Self-discharged Patients. Psychia—
tric Quarterly, 1973, 47, 175-183.

 

Park, L. C., Covi, L., Non-blind Placebo Trial: An Explora-
tion of Neurotic Patients' Responses to Placebo When its

Inert Content is Disclosed. Archives of General Psychiatgy,
1965, 12:336-345.

 

164

Piven, Frances Fox, and Cloward, Richard A., Regulating“
the Poor: The Functions of Public Welfare. New York: Pan-
theon, 1971.

 

 

Pratt, T. C., Linn, M. W., Carmichael, J. S., and Webb, N.
L., The Alcoholic's Perception of the Ward as a Predictor
of Aftercare Attendance. Journal of Clinical Psychology,

1977, 33, 915-918.

 

Pressman, Jeffrey L., and Wildarsky, Aaron B., Implementa-
tion: How Great Expectations in Washington Are Dashed in Oak—
land or Why It's Amazing Thangederal Programs Work at All.
Berkeley, University of California Press, 1973.

 

 

Quatrone, D., Profile of Active and Terminated Patients in a
Methadone Maintenance Program. Proceedings of the Fifth
National Conference on Methadone Treatment, 1973, 1, 760-766.

 

 

Raynes, A. E., Patch, V. D., and Fisch, A., Distinquishing
Features of Heroin Addicts Benefitting From Detoxification.
Proceedings of the Fourth National Conference on Methadone
Treatment, 1972, 465-467.

 

 

Renton, C. A., Affleck, J. W., Carstairs, F. M-, and Forrest,
A. B., A Follow-up of Schizophrenic Patients in Edinburgh.
Acta Psychiatrica Neurologica Scandinavia, 1963, 39, 548-600.

 

Rickels, K., Hesbacher, P., Vandervort, W., et al., Tybamite:
A Perplexing Drug., American Journal of Psyphiatpy, 1968,
125:320.

 

Rockart, J. F., Hofmann, P. B., Physician and Patient Behavior
Under Different Scheduling Systems in a Hospital Outpatient
Department. Medical Care, 1961, 7:463.

 

Rosenblut, A., Jonas, 8., Wassertheil, S., et a1, OPD Wait-
ing Time Reduced by_Use of an Individual Appointment System.

 

Rubenstein, H. 8., Behavior in a Medical Clinic of Patients
With Well Controlled Asthma. Lancet, 1976, 1(7967):101l.

Russell, W., Recent Progress and Further Needs in the Recog-
nition, Examination and Commitment of the Insane. Proceed-
ings of the Mental Hygiene Conference, National Committee
for Mental Hygiene, New York, 1912, p. 189.

 

Ryan, W., Blaming the Victim, Pantheon Books, 1971, New York.

 

Sackett, D. L., The Magnitude of Compliance and Non—Compliance.
In: Sackett, D. L., Haynes, R. B., eds., Compliance With
Therapeutic Regimens. Baltimore: John Hopkins University
Press, 1976

 

 

165

Schroeder, S. A., Lowering Broken Appointment Rates at a
Medical Clinic. Medical Care, 1973, 11:75.

 

Sethna, E. R., and Harrington, J. A., A Study of Patients
Who Lapsed From Group Psychotherapy. British Journal of
of Psychiatry, 1971, 119, 59-69.

 

 

Shah, C. P., MacBride, J. R., Lamb, G. A., Appointment
Systems: Why Does A Patient Not Return? Canadian Journal
of Public Health, 1977, 68:148.

 

 

Shepard, D. S., Moseley, T. A. E., Mailed vs. Telephoned
Appointment Reminders to Reduce Broken Appointments in a
Hospital Outpatient Department. Medical Care, 1976, 14:268.

 

Shmarak, K. L., Reduce Your Broken Appointment Rate: How
One Children and Youth Project Reduced its Broken Appoint-
ment Rate. Public Health, 1971, 61:2400.

 

Shmarak, K. L., Reduce Your Broken Appointment Rate: How
One Children and Youth Project Reduced its Broken Appoint-
ment Rate. Public Healppy 1971, 61:2400.

 

Shonick, W., Klein, B. W., An Approach to Reducing the Ad-
verse Effects of Broken Appointments in Primary Care Systems:
Development of a Decision Rule Based on Estimated Conditional
Probability. Medical Care, 1977, 15:419.

 

Shonick, W., Klein, B. W., An Approach to Reducing the Ad-
verse Effect of Broken Appointments in Primary Care Systems.
Medical Care, 1977, 15:419.

 

Spitzer, W. 0., Sackett, D. L., Sibley, J. C., et al. The
Burlington Randomized Trial of the Nurse Practitioner. New
England Journal of Medicine, 1974, 290:251.

 

Steiner, Gilbert, The State of Welfare, Washington, D. C.,
The Bookings Institute, 1971.

 

Stevenson, C., Child Guidance and the National Committee
for Mental Hygiene. Orthopsychiatry, 1948.

 

Stine, O. C., Chuagui, C., Jimenez, C., et al, Broken Appoint-
ments at a Comprehensive Clinic for Chidren. Medical Care,
1968, 6:322. -

 

Suchman, E. A., Health Attitudes and Behavior. Chicago Ar-
chives on Environmental Health, Jan: 1970, 20:195-10.

 

 

Tagliacozzo, D. M., Luskin, D. B., Lashof, J. C., et al,
Nurse Intervention and Patient Behavior. American Journal
of Public Health, 1974, 64:596.

 

 

166

Tagliacozzo, D. M., Ima, K., Knowledge of Illness as a
Predictor of Patient Behavior. Journal of Chronic Disease,
1970, 22:765.

 

Tash, R. H., O'Shea, R. M., Cohen, L. K., Testing a Pre-
ventive-Symptomatic Theory of Dental Health Behavior.
American Journal of Public Health, 1969, 59:514.

 

Taylor, D. W., A Test of the Health Belief Model in Hyper-
tension. In: Haynes, R. B., Taylor, W. W., Sackett, S. L.,
eds. Compliances in Health Care. Baltimore: John HOpkins
University Press, 1979, 103.

 

Tomsovic, M. A Follow-up Study of Discharged Alcoholics.
Hospital and Community Psychiatry, 1970, 21, 94-97.

 

Tryon, Robert C., and Bailey, Daniel, Cluster Analysis.
McGraw-Hill, New York, 1970, p. 47.

 

Tucker, Charles H., The Role of Readingi Speaking Dialect
and Associative Bridging in Behavioral Achievement and
Attitude Change, Dissertation, Michigan State University,
1974, p. 57.

 

 

 

U. S. Commission on Chronic Illness: Chronic Illness in.
the United Stateg, Volume 1, Prevention of Chronic Illness,
Cambridge, Howard University Press, 1957, p. 45.

 

 

 

U. S. Comptroller General, Improvements Needed to Speed
Implementation of Medicaid's Early and Periodic Screening,
Diagnosis and Treatment Program. Washington, D. C., General
Accounting Office, 1975.

 

 

U. S. Congress, House Subcommittee on Overweight and Investi-
gations of the Committee on Interstate and Foreign Commerce
Report: Department of Health, Education and Welfare's Admin-
istration of Health Programs; Shortchanging Children, 94th
Congress, 2nd Session, September, 1976, p. iii.

 

U. S. Department of Health, Education, and Welfare, Data on
the Medicaid Program: Eligibilipy/ServiceslEgpenditures, 1979
Edition (revised), Medicaid/Medicare Management Institute,
Health Care Financing Administration, 1979, p. 53.

 

U. S. Health, Education and Welfare, Vital Statistics Report,
Washington, D. C., Table 6, 1967.

 

Urban Health Study, East Los Angeles Health Systems, Inc.,
December, 1973.

 

Walsh, J. J., Bouton, J. L., Arnold, I. G., Why Patients
Break Appointments. Hogpital Top, 1967, 45(2):67.

 

167

Winslow, C., The Evolution and Significance of the Modern
Public Health Campaign. Yale University Press, 1923, New
Haven.

 

 

White, M. K., Alpert, J. J., Kosa, J., Hard-to—Reach Families
in a Comprehensive Care Program. JAMA, 1967, 201:123.

Wilmer, H. A., The Relationship of the Physician to the Self—

discharge Behavior of Tuberculosis Patients. In: P. J. Sparer
(Ed.), Personalityy Stress and Tuberculosis. New York, Inter—

national Universities Press, 1956.

 

Wolkon, G. H., and Moriwaki, S., Race and Social Class as
Factors in the Orientation Toward Psychotherapy. Journal
of Counseling Psychology, 1973, 20, 312-316.